From dede10b34c14a4f704c55d385f81c1b73ff61182 Mon Sep 17 00:00:00 2001 From: Antony Lewis Date: Fri, 9 Aug 2024 17:09:53 +0200 Subject: [PATCH] Add --allow-changes for resume (#374) --- CHANGELOG.md | 6 + cobaya/__init__.py | 2 +- cobaya/output.py | 22 +- cobaya/run.py | 7 + docs/cobaya-example.ipynb | 3923 ++++++++++++++++++++----------------- docs/example.rst | 2 +- docs/installation.rst | 2 +- tests/test_cosmo_run.py | 8 +- tests/test_input.py | 1 + 9 files changed, 2137 insertions(+), 1836 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index edb6e1d51..defd741c8 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,3 +1,9 @@ +## 3.5.3 + +- added --allow-changes option to cobaya-run to allow changes in the input file +- Updates for deprecation warnings +- Minor optimization refactor and doc update + ## 3.5.2 - Updates for numpy 2 and other compatibility fixes diff --git a/cobaya/__init__.py b/cobaya/__init__.py index aaef97168..b8e00240d 100644 --- a/cobaya/__init__.py +++ b/cobaya/__init__.py @@ -15,7 +15,7 @@ __author__ = "Jesus Torrado and Antony Lewis" -__version__ = "3.5.2" +__version__ = "3.5.3" __obsolete__ = False __year__ = "2024" __url__ = "https://cobaya.readthedocs.io" diff --git a/cobaya/output.py b/cobaya/output.py index ef9a9ff28..30e1cd426 100644 --- a/cobaya/output.py +++ b/cobaya/output.py @@ -463,7 +463,7 @@ def check_and_dump_info(self, input_info, updated_info, check_compatible=True, - idem, populated with the components' defaults. If resuming a sample, checks first that old and new infos and versions are - consistent. + consistent unless allow_changes is True. """ # trim known params of each likelihood: for internal use only self.check_lock() @@ -471,7 +471,7 @@ def check_and_dump_info(self, input_info, updated_info, check_compatible=True, updated_info_trimmed["version"] = get_version() for like_info in updated_info_trimmed.get("likelihood", {}).values(): (like_info or {}).pop("params", None) - if check_compatible: + if check_compatible or cache_old: # We will test the old info against the dumped+loaded new info. # This is because we can't actually check if python objects do change try: @@ -481,21 +481,16 @@ def check_and_dump_info(self, input_info, updated_info, check_compatible=True, # for example, when there's a dynamically generated class that cannot # be found by the yaml loader (could use yaml loader that ignores them) old_info = None - if old_info: - # use consistent yaml read-in types - # TODO: could probably just compare full infos here, with externals? - # for the moment cautiously keeping old behaviour - old_info = yaml_load(yaml_dump(old_info)) # type: ignore - if old_info.get("test"): - old_info = None - if old_info: + if check_compatible and old_info and not old_info.get("test"): + old_info = yaml_load(yaml_dump(old_info)) new_info = yaml_load(yaml_dump(updated_info_trimmed)) if not is_equal_info(old_info, new_info, strict=False, ignore_blocks=list(ignore_blocks) + [ "output"]): raise LoggedError( self.log, "Old and new run information not compatible! " - "Resuming not possible!") + "Resuming not possible!\n" + "Use --allow-changes to proceed anyway.") # Deal with version comparison separately: # - If not specified now, take the one used in resume info # - If specified both now and before, check new older than old one @@ -540,11 +535,8 @@ def check_and_dump_info(self, input_info, updated_info, check_compatible=True, for f, info in [(self.file_input, input_info), (self.file_updated, updated_info_trimmed)]: if info: - for k in ignore_blocks: + for k in tuple(ignore_blocks) + ("debug", "force", "resume"): info.pop(k, None) - info.pop("debug", None) - info.pop("force", None) - info.pop("resume", None) # make sure the dumped output prefix does only contain the file prefix, # not the folder, since it's already been placed inside it info["output"] = self.updated_prefix() diff --git a/cobaya/run.py b/cobaya/run.py index 4d9f0a9c1..8b62e34be 100644 --- a/cobaya/run.py +++ b/cobaya/run.py @@ -34,6 +34,7 @@ def run(info_or_yaml_or_file: Union[InputDict, str, os.PathLike], minimize: Optional[bool] = None, no_mpi: bool = False, test: Optional[bool] = None, override: Optional[InputDict] = None, + allow_changes: bool = False ) -> Tuple[InputDict, Union[Sampler, PostResult]]: """ Run from an input dictionary, file name or yaml string, with optional arguments @@ -51,6 +52,7 @@ def run(info_or_yaml_or_file: Union[InputDict, str, os.PathLike], :param test: only test initialization rather than actually running :param override: option dictionary to merge into the input one, overriding settings (but with lower precedence than the explicit keyword arguments) + :param allow_changes: if true, allow input option changes when resuming or minimizing :return: (updated_info, sampler) tuple of options dictionary and Sampler instance, or (updated_info, post_results) if using "post" post-processing """ @@ -103,6 +105,7 @@ def run(info_or_yaml_or_file: Union[InputDict, str, os.PathLike], # 3. If output requested, check compatibility if existing one, and dump. # 3.1 First: model only out.check_and_dump_info(info, updated_info, cache_old=True, + check_compatible=not allow_changes, ignore_blocks=["sampler"]) # 3.2 Then sampler -- 1st get the first sampler mentioned in the updated.yaml if not (info_sampler := updated_info.get("sampler")): @@ -177,6 +180,10 @@ def run_script(args=None): help=("Replaces the sampler in the input and runs a minimization " "process (incompatible with post-processing)."), **trueNone_kwargs) + parser.add_argument("--allow-changes", + help="Allow changing input parameters when resuming " + "or minimizing, skipping consistency checks", + **trueNone_kwargs) parser.add_argument("--version", action="version", version=get_version()) parser.add_argument("--no-mpi", help=("disable MPI when mpi4py installed but MPI does " diff --git a/docs/cobaya-example.ipynb b/docs/cobaya-example.ipynb index fb12eca70..a7eef950c 100644 --- a/docs/cobaya-example.ipynb +++ b/docs/cobaya-example.ipynb @@ -104,756 +104,524 @@ "[mcmc] Getting initial point... (this may take a few seconds)\n", "[mcmc] Initial point: x:0.5, y:0.5\n", "[model] Measuring speeds... (this may take a few seconds)\n", - "[model] Setting measured speeds (per sec): {ring: 5990.0}\n", + "[model] Setting measured speeds (per sec): {ring: 6820.0}\n", "[mcmc] Covariance matrix not present. We will start learning the covariance of the proposal earlier: R-1 = 30 (would be 2 if all params loaded).\n", "[mcmc] Sampling!\n", - "[mcmc] Progress @ 2024-08-09 09:38:16 : 1 steps taken, and 0 accepted.\n", + "[mcmc] Progress @ 2024-08-09 15:13:15 : 1 steps taken, and 0 accepted.\n", "[mcmc] Learn + convergence test @ 80 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.688\n", - "[mcmc] - Convergence of means: R-1 = 19.398481 after 64 accepted steps\n", + "[mcmc] - Acceptance rate: 0.753\n", + "[mcmc] - Convergence of means: R-1 = 6.266116 after 64 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.525\n", - "[mcmc] - Convergence of means: R-1 = 3.315150 after 128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.527\n", + "[mcmc] - Convergence of means: R-1 = 0.635726 after 128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.413\n", - "[mcmc] - Convergence of means: R-1 = 1.784527 after 192 accepted steps\n", + "[mcmc] - Acceptance rate: 0.343\n", + "[mcmc] - Convergence of means: R-1 = 0.749305 after 192 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.304\n", - "[mcmc] - Convergence of means: R-1 = 0.858393 after 256 accepted steps\n", + "[mcmc] - Acceptance rate: 0.226\n", + "[mcmc] - Convergence of means: R-1 = 0.386923 after 256 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.246\n", - "[mcmc] - Convergence of means: R-1 = 0.476199 after 320 accepted steps\n", + "[mcmc] - Acceptance rate: 0.176\n", + "[mcmc] - Convergence of means: R-1 = 0.325198 after 320 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.211\n", - "[mcmc] - Convergence of means: R-1 = 0.391530 after 384 accepted steps\n", + "[mcmc] - Acceptance rate: 0.152\n", + "[mcmc] - Convergence of means: R-1 = 0.195328 after 384 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.214224 after 448 accepted steps\n", + "[mcmc] - Acceptance rate: 0.135\n", + "[mcmc] - Convergence of means: R-1 = 0.240948 after 448 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.171\n", - "[mcmc] - Convergence of means: R-1 = 0.163997 after 512 accepted steps\n", + "[mcmc] - Acceptance rate: 0.128\n", + "[mcmc] - Convergence of means: R-1 = 0.079528 after 512 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.156\n", - "[mcmc] - Convergence of means: R-1 = 0.089496 after 576 accepted steps\n", + "[mcmc] - Acceptance rate: 0.121\n", + "[mcmc] - Convergence of means: R-1 = 0.104897 after 576 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.142\n", - "[mcmc] - Convergence of means: R-1 = 0.101825 after 640 accepted steps\n", + "[mcmc] - Acceptance rate: 0.115\n", + "[mcmc] - Convergence of means: R-1 = 0.070654 after 640 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.134\n", - "[mcmc] - Convergence of means: R-1 = 0.103353 after 704 accepted steps\n", + "[mcmc] - Acceptance rate: 0.110\n", + "[mcmc] - Convergence of means: R-1 = 0.055986 after 704 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.129\n", - "[mcmc] - Convergence of means: R-1 = 0.063573 after 768 accepted steps\n", + "[mcmc] - Acceptance rate: 0.110\n", + "[mcmc] - Convergence of means: R-1 = 0.132901 after 768 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.123\n", - "[mcmc] - Convergence of means: R-1 = 0.085604 after 832 accepted steps\n", + "[mcmc] - Acceptance rate: 0.108\n", + "[mcmc] - Convergence of means: R-1 = 0.056798 after 832 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.120\n", - "[mcmc] - Convergence of means: R-1 = 0.064863 after 896 accepted steps\n", + "[mcmc] - Acceptance rate: 0.107\n", + "[mcmc] - Convergence of means: R-1 = 0.046266 after 896 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.117\n", - "[mcmc] - Convergence of means: R-1 = 0.068390 after 960 accepted steps\n", + "[mcmc] - Acceptance rate: 0.105\n", + "[mcmc] - Convergence of means: R-1 = 0.068523 after 960 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.114\n", - "[mcmc] - Convergence of means: R-1 = 0.076408 after 1024 accepted steps\n", + "[mcmc] - Acceptance rate: 0.101\n", + "[mcmc] - Convergence of means: R-1 = 0.061209 after 1024 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.113\n", - "[mcmc] - Convergence of means: R-1 = 0.049658 after 1088 accepted steps\n", + "[mcmc] - Acceptance rate: 0.098\n", + "[mcmc] - Convergence of means: R-1 = 0.016922 after 1088 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.109\n", - "[mcmc] - Convergence of means: R-1 = 0.031811 after 1152 accepted steps\n", + "[mcmc] - Acceptance rate: 0.097\n", + "[mcmc] - Convergence of means: R-1 = 0.007255 after 1152 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.108\n", - "[mcmc] - Convergence of means: R-1 = 0.036533 after 1216 accepted steps\n", + "[mcmc] - Acceptance rate: 0.097\n", + "[mcmc] - Convergence of means: R-1 = 0.006309 after 1216 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.106\n", - "[mcmc] - Convergence of means: R-1 = 0.019146 after 1280 accepted steps\n", + "[mcmc] - Acceptance rate: 0.096\n", + "[mcmc] - Convergence of means: R-1 = 0.004943 after 1280 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.105\n", - "[mcmc] - Convergence of means: R-1 = 0.018104 after 1344 accepted steps\n", + "[mcmc] - Acceptance rate: 0.096\n", + "[mcmc] - Convergence of means: R-1 = 0.011111 after 1344 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.104\n", - "[mcmc] - Convergence of means: R-1 = 0.011420 after 1408 accepted steps\n", + "[mcmc] - Acceptance rate: 0.096\n", + "[mcmc] - Convergence of means: R-1 = 0.012277 after 1408 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.104\n", - "[mcmc] - Convergence of means: R-1 = 0.013207 after 1472 accepted steps\n", + "[mcmc] - Acceptance rate: 0.096\n", + "[mcmc] - Convergence of means: R-1 = 0.017863 after 1472 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.103\n", - "[mcmc] - Convergence of means: R-1 = 0.008931 after 1536 accepted steps\n", + "[mcmc] - Acceptance rate: 0.096\n", + "[mcmc] - Convergence of means: R-1 = 0.029332 after 1536 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.103\n", - "[mcmc] - Convergence of means: R-1 = 0.007207 after 1600 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.014933 after 1600 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.102\n", - "[mcmc] - Convergence of means: R-1 = 0.008045 after 1664 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.011300 after 1664 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.101\n", - "[mcmc] - Convergence of means: R-1 = 0.015833 after 1728 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.010939 after 1728 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.101\n", - "[mcmc] - Convergence of means: R-1 = 0.011108 after 1792 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.022174 after 1792 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.101\n", - "[mcmc] - Convergence of means: R-1 = 0.006649 after 1856 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.012540 after 1856 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.101\n", - "[mcmc] - Convergence of means: R-1 = 0.009323 after 1920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.011564 after 1920 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.101\n", - "[mcmc] - Convergence of means: R-1 = 0.007083 after 1984 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.012802 after 1984 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.100\n", - "[mcmc] - Convergence of means: R-1 = 0.014704 after 2048 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.005467 after 2048 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.098\n", - "[mcmc] - Convergence of means: R-1 = 0.007670 after 2112 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.008068 after 2112 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.098\n", - "[mcmc] - Convergence of means: R-1 = 0.002129 after 2176 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.013097 after 2176 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.098\n", - "[mcmc] - Convergence of means: R-1 = 0.003661 after 2240 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.008621 after 2240 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.098\n", - "[mcmc] - Convergence of means: R-1 = 0.007467 after 2304 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.008002 after 2304 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.098\n", - "[mcmc] - Convergence of means: R-1 = 0.007551 after 2368 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.011965 after 2368 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.097\n", - "[mcmc] - Convergence of means: R-1 = 0.010929 after 2432 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.009179 after 2432 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.097\n", - "[mcmc] - Convergence of means: R-1 = 0.008957 after 2496 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.011461 after 2496 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.096\n", - "[mcmc] - Convergence of means: R-1 = 0.007616 after 2560 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.009252 after 2560 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.009236 after 2624 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.006048 after 2624 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.010228 after 2688 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.004257 after 2688 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.096\n", - "[mcmc] - Convergence of means: R-1 = 0.012496 after 2752 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.005328 after 2752 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.005920 after 2816 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.007179 after 2816 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.009187 after 2880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.007871 after 2880 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.010817 after 2944 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.008274 after 2944 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.095\n", - "[mcmc] - Convergence of means: R-1 = 0.011547 after 3008 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.007519 after 3008 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.094\n", - "[mcmc] - Convergence of means: R-1 = 0.011021 after 3072 accepted steps\n", + "[mcmc] - Acceptance rate: 0.095\n", + "[mcmc] - Convergence of means: R-1 = 0.008983 after 3072 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.013130 after 3136 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.008339 after 3136 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.013201 after 3200 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.006079 after 3200 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.011063 after 3264 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.011399 after 3264 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4160 samples accepted.\n", "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.006389 after 3328 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.007168 after 3328 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4240 samples accepted.\n", "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.004359 after 3392 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.008040 after 3392 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4320 samples accepted.\n", "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.003583 after 3456 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.011545 after 3456 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003166 after 3520 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.016155 after 3520 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002959 after 3584 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.014921 after 3584 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002789 after 3648 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.011640 after 3648 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004750 after 3712 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009169 after 3712 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004879 after 3776 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.011311 after 3776 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.006282 after 3840 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.014650 after 3840 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.009770 after 3904 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.012881 after 3904 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.009808 after 3968 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007246 after 3968 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.012804 after 4032 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005343 after 4032 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.010722 after 4096 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005707 after 4096 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.009025 after 4160 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.008508 after 4160 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.005523 after 4224 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007347 after 4224 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.005670 after 4288 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.006988 after 4288 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003722 after 4352 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.008235 after 4352 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.005524 after 4416 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009142 after 4416 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.005142 after 4480 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.010026 after 4480 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006035 after 4544 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009948 after 4544 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006572 after 4608 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009252 after 4608 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.007118 after 4672 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009557 after 4672 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006108 after 4736 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.011702 after 4736 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.007209 after 4800 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.011643 after 4800 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006582 after 4864 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.010201 after 4864 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.007438 after 4928 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.009908 after 4928 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.007098 after 4992 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007405 after 4992 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.005754 after 5056 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007382 after 5056 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.004541 after 5120 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005927 after 5120 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006471 after 5184 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.006265 after 5184 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.005718 after 5248 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005988 after 5248 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.004837 after 5312 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.004893 after 5312 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.007932 after 5376 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.003638 after 5376 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.008692 after 5440 accepted steps\n", + "[mcmc] - Acceptance rate: 0.094\n", + "[mcmc] - Convergence of means: R-1 = 0.004163 after 5440 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.006741 after 5504 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.003660 after 5504 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.005652 after 5568 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.002628 after 5568 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.006149 after 5632 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.002172 after 5632 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.006081 after 5696 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.002378 after 5696 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.006059 after 5760 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.002432 after 5760 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.005150 after 5824 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.002580 after 5824 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004687 after 5888 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.004389 after 5888 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004536 after 5952 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005195 after 5952 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.005047 after 6016 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.003990 after 6016 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7600 samples accepted.\n", "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.003580 after 6080 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004920 after 6080 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7680 samples accepted.\n", "[mcmc] - Acceptance rate: 0.093\n", - "[mcmc] - Convergence of means: R-1 = 0.002452 after 6144 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.006196 after 6144 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002359 after 6208 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007735 after 6208 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.001920 after 6272 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007969 after 6272 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002082 after 6336 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.007043 after 6336 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002381 after 6400 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005783 after 6400 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.001735 after 6464 accepted steps\n", + "[mcmc] - Acceptance rate: 0.093\n", + "[mcmc] - Convergence of means: R-1 = 0.005003 after 6464 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8160 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002507 after 6528 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004460 after 6528 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8240 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002514 after 6592 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.005134 after 6592 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8320 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002987 after 6656 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004082 after 6656 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8400 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002788 after 6720 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003654 after 6720 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8480 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.002593 after 6784 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.002103 after 6784 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8560 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004979 after 6848 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.002716 after 6848 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8640 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004596 after 6912 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003148 after 6912 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8720 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003821 after 6976 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003059 after 6976 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8800 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003076 after 7040 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003920 after 7040 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8880 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.004044 after 7104 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004102 after 7104 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8960 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003302 after 7168 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003767 after 7168 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9040 samples accepted.\n", "[mcmc] - Acceptance rate: 0.092\n", - "[mcmc] - Convergence of means: R-1 = 0.003229 after 7232 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004214 after 7232 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003304 after 7296 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.003839 after 7296 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003141 after 7360 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.002829 after 7360 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003793 after 7424 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.003377 after 7424 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003425 after 7488 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.002425 after 7488 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003755 after 7552 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.001849 after 7552 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003263 after 7616 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.001623 after 7616 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.002766 after 7680 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.002538 after 7680 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003265 after 7744 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.002354 after 7744 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.003385 after 7808 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.002453 after 7808 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.091\n", - "[mcmc] - Convergence of means: R-1 = 0.002471 after 7872 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.001297 after 7872 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003365 after 7936 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.000715 after 7936 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004816 after 8000 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.001113 after 8000 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004275 after 8064 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.001161 after 8064 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004094 after 8128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.092\n", + "[mcmc] - Convergence of means: R-1 = 0.000398 after 8128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004773 after 8192 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003986 after 8256 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003718 after 8320 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003032 after 8384 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003104 after 8448 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002946 after 8512 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003269 after 8576 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002784 after 8640 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002533 after 8704 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 10960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003806 after 8768 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003817 after 8832 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002566 after 8896 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002113 after 8960 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002430 after 9024 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002677 after 9088 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003654 after 9152 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002486 after 9216 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002130 after 9280 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003207 after 9344 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003260 after 9408 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003368 after 9472 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 11920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004029 after 9536 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004274 after 9600 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004112 after 9664 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003833 after 9728 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004482 after 9792 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002700 after 9856 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002888 after 9920 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003047 after 9984 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003249 after 10048 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003269 after 10112 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003202 after 10176 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003553 after 10240 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004191 after 10304 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 12960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.004887 after 10368 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003628 after 10432 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002933 after 10496 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002995 after 10560 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.003280 after 10624 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002982 after 10688 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002180 after 10752 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002091 after 10816 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001455 after 10880 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001703 after 10944 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.089\n", - "[mcmc] - Convergence of means: R-1 = 0.001821 after 11008 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001538 after 11072 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 13920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.002043 after 11136 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001952 after 11200 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001146 after 11264 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001149 after 11328 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001053 after 11392 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001401 after 11456 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001536 after 11520 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001936 after 11584 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001918 after 11648 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001235 after 11712 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.001469 after 11776 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.000876 after 11840 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 14880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.090\n", - "[mcmc] - Convergence of means: R-1 = 0.000935 after 11904 accepted steps\n", - "[mcmc] - Convergence of bounds: R-1 = 0.019702 after 14880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.091\n", + "[mcmc] - Convergence of means: R-1 = 0.000174 after 8192 accepted steps\n", + "[mcmc] - Convergence of bounds: R-1 = 0.022435 after 10240 accepted steps\n", "[mcmc] The run has converged!\n", - "[mcmc] Sampling complete after 14880 accepted steps.\n" + "[mcmc] Sampling complete after 10240 accepted steps.\n" ] } ], @@ -875,12 +643,12 @@ "name": "stdout", "output_type": "stream", "text": [ - "[mcmc] *WARNING* When combining chains, it is recommended to remove some initial fraction, e.g. 'skip_samples=0.3'\n" + "[root] *WARNING* auto bandwidth for theta very small or failed (h=0.0006228019329792012,N_eff=2423.001184451471). Using fallback (h=0.0801279372160761)\n" ] }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -895,7 +663,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -912,7 +680,7 @@ "source": [ "import getdist.plots as gdplt\n", "\n", - "gdsamples = sampler.products(to_getdist=True)[\"sample\"]\n", + "gdsamples = sampler.products(to_getdist=True, skip_samples=0.3)[\"sample\"]\n", "gdplot = gdplt.get_subplot_plotter(width_inch=5)\n", "gdplot.triangle_plot(gdsamples, [\"x\", \"y\"], filled=True)\n", "gdplot.export(\"example_adv_ring.png\")\n", @@ -944,569 +712,1569 @@ "[mcmc] Getting initial point... (this may take a few seconds)\n", "[mcmc] Initial point: x:0.5, y:0.5\n", "[model] Measuring speeds... (this may take a few seconds)\n", - "[model] Setting measured speeds (per sec): {ring: 6020.0}\n", + "[model] Setting measured speeds (per sec): {ring: 12700.0}\n", "[mcmc] Covariance matrix not present. We will start learning the covariance of the proposal earlier: R-1 = 30 (would be 2 if all params loaded).\n", "[mcmc] Sampling!\n", - "[mcmc] Progress @ 2024-08-09 09:38:33 : 1 steps taken, and 0 accepted.\n", + "[mcmc] Progress @ 2024-08-09 15:13:29 : 1 steps taken, and 0 accepted.\n", "[mcmc] Learn + convergence test @ 80 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.727\n", - "[mcmc] - Convergence of means: R-1 = 1.089851 after 64 accepted steps\n", + "[mcmc] - Acceptance rate: 0.681\n", + "[mcmc] - Convergence of means: R-1 = 13.938110 after 64 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.650\n", - "[mcmc] - Convergence of means: R-1 = 0.588121 after 128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.542\n", + "[mcmc] - Convergence of means: R-1 = 0.676284 after 128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.443\n", - "[mcmc] - Convergence of means: R-1 = 0.104657 after 192 accepted steps\n", + "[mcmc] - Acceptance rate: 0.428\n", + "[mcmc] - Convergence of means: R-1 = 0.327621 after 192 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.394\n", - "[mcmc] - Convergence of means: R-1 = 0.105382 after 256 accepted steps\n", + "[mcmc] - Acceptance rate: 0.357\n", + "[mcmc] - Convergence of means: R-1 = 0.108285 after 256 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.347\n", - "[mcmc] - Convergence of means: R-1 = 0.077792 after 320 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.092346 after 320 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.315\n", - "[mcmc] - Convergence of means: R-1 = 0.036589 after 384 accepted steps\n", + "[mcmc] - Acceptance rate: 0.263\n", + "[mcmc] - Convergence of means: R-1 = 0.114896 after 384 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.062128 after 448 accepted steps\n", + "[mcmc] - Acceptance rate: 0.245\n", + "[mcmc] - Convergence of means: R-1 = 0.029267 after 448 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.289\n", - "[mcmc] - Convergence of means: R-1 = 0.054072 after 512 accepted steps\n", + "[mcmc] - Acceptance rate: 0.233\n", + "[mcmc] - Convergence of means: R-1 = 0.053584 after 512 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.275\n", - "[mcmc] - Convergence of means: R-1 = 0.065704 after 576 accepted steps\n", + "[mcmc] - Acceptance rate: 0.224\n", + "[mcmc] - Convergence of means: R-1 = 0.014337 after 576 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.276\n", - "[mcmc] - Convergence of means: R-1 = 0.030672 after 640 accepted steps\n", + "[mcmc] - Acceptance rate: 0.210\n", + "[mcmc] - Convergence of means: R-1 = 0.010908 after 640 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.269\n", - "[mcmc] - Convergence of means: R-1 = 0.058332 after 704 accepted steps\n", + "[mcmc] - Acceptance rate: 0.208\n", + "[mcmc] - Convergence of means: R-1 = 0.039703 after 704 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.261\n", - "[mcmc] - Convergence of means: R-1 = 0.047144 after 768 accepted steps\n", + "[mcmc] - Acceptance rate: 0.203\n", + "[mcmc] - Convergence of means: R-1 = 0.034227 after 768 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.258\n", - "[mcmc] - Convergence of means: R-1 = 0.054727 after 832 accepted steps\n", + "[mcmc] - Acceptance rate: 0.204\n", + "[mcmc] - Convergence of means: R-1 = 0.026114 after 832 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.258\n", - "[mcmc] - Convergence of means: R-1 = 0.024700 after 896 accepted steps\n", + "[mcmc] - Acceptance rate: 0.197\n", + "[mcmc] - Convergence of means: R-1 = 0.021122 after 896 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.247\n", - "[mcmc] - Convergence of means: R-1 = 0.027643 after 960 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.010483 after 960 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.242\n", - "[mcmc] - Convergence of means: R-1 = 0.039965 after 1024 accepted steps\n", + "[mcmc] - Acceptance rate: 0.193\n", + "[mcmc] - Convergence of means: R-1 = 0.006988 after 1024 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.239\n", - "[mcmc] - Convergence of means: R-1 = 0.032398 after 1088 accepted steps\n", + "[mcmc] - Acceptance rate: 0.193\n", + "[mcmc] - Convergence of means: R-1 = 0.012615 after 1088 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.240\n", - "[mcmc] - Convergence of means: R-1 = 0.033837 after 1152 accepted steps\n", + "[mcmc] - Acceptance rate: 0.192\n", + "[mcmc] - Convergence of means: R-1 = 0.017592 after 1152 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.232\n", - "[mcmc] - Convergence of means: R-1 = 0.037282 after 1216 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.018392 after 1216 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.230\n", - "[mcmc] - Convergence of means: R-1 = 0.011148 after 1280 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.011165 after 1280 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.229\n", - "[mcmc] - Convergence of means: R-1 = 0.011706 after 1344 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.015537 after 1344 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.227\n", - "[mcmc] - Convergence of means: R-1 = 0.023239 after 1408 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.009587 after 1408 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.224\n", - "[mcmc] - Convergence of means: R-1 = 0.014407 after 1472 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.008229 after 1472 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.223\n", - "[mcmc] - Convergence of means: R-1 = 0.031844 after 1536 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.010150 after 1536 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.219\n", - "[mcmc] - Convergence of means: R-1 = 0.018587 after 1600 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.005544 after 1600 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.217\n", - "[mcmc] - Convergence of means: R-1 = 0.036447 after 1664 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.013067 after 1664 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.215\n", - "[mcmc] - Convergence of means: R-1 = 0.032098 after 1728 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.015911 after 1728 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.213\n", - "[mcmc] - Convergence of means: R-1 = 0.017706 after 1792 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.011833 after 1792 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.210\n", - "[mcmc] - Convergence of means: R-1 = 0.020787 after 1856 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004696 after 1856 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.208\n", - "[mcmc] - Convergence of means: R-1 = 0.020573 after 1920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.010256 after 1920 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.206\n", - "[mcmc] - Convergence of means: R-1 = 0.029298 after 1984 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.014342 after 1984 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.204\n", - "[mcmc] - Convergence of means: R-1 = 0.026224 after 2048 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.002624 after 2048 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.202\n", - "[mcmc] - Convergence of means: R-1 = 0.032009 after 2112 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.009186 after 2112 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.199\n", - "[mcmc] - Convergence of means: R-1 = 0.028089 after 2176 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.008363 after 2176 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.197\n", - "[mcmc] - Convergence of means: R-1 = 0.019481 after 2240 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005572 after 2240 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.195\n", - "[mcmc] - Convergence of means: R-1 = 0.021746 after 2304 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004926 after 2304 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.017489 after 2368 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.009103 after 2368 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.013915 after 2432 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.017323 after 2432 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.023999 after 2496 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.024379 after 2496 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.016602 after 2560 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.025962 after 2560 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.014377 after 2624 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.013716 after 2624 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.009942 after 2688 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.011409 after 2688 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.004169 after 2752 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.010922 after 2752 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.005804 after 2816 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.014524 after 2816 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.186\n", - "[mcmc] - Convergence of means: R-1 = 0.004960 after 2880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.012646 after 2880 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.186\n", - "[mcmc] - Convergence of means: R-1 = 0.003894 after 2944 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.011555 after 2944 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.185\n", - "[mcmc] - Convergence of means: R-1 = 0.007415 after 3008 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.008615 after 3008 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.183\n", - "[mcmc] - Convergence of means: R-1 = 0.008796 after 3072 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.007832 after 3072 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3920 samples accepted.\n", "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.009581 after 3136 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.013598 after 3136 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.182\n", - "[mcmc] - Convergence of means: R-1 = 0.011340 after 3200 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.016308 after 3200 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.181\n", - "[mcmc] - Convergence of means: R-1 = 0.016534 after 3264 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.018936 after 3264 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.016574 after 3328 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.018008 after 3328 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.022134 after 3392 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.017373 after 3392 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.012939 after 3456 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.012660 after 3456 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.012632 after 3520 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.011953 after 3520 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.008629 after 3584 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.009585 after 3584 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.008906 after 3648 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.009447 after 3648 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.008736 after 3712 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.007406 after 3712 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.009689 after 3776 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.009306 after 3776 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.005913 after 3840 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.007175 after 3840 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.004685 after 3904 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006910 after 3904 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004551 after 3968 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004486 after 3968 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004988 after 4032 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.004522 after 4032 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004418 after 4096 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.008197 after 4096 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.003522 after 4160 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.008564 after 4160 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.002096 after 4224 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.011973 after 4224 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.002288 after 4288 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.012761 after 4288 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003493 after 4352 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.011487 after 4352 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003516 after 4416 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.008833 after 4416 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.005606 after 4480 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.009340 after 4480 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007315 after 4544 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.008772 after 4544 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.005295 after 4608 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.011085 after 4608 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007401 after 4672 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.010488 after 4672 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.005815 after 4736 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.012798 after 4736 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007394 after 4800 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.013283 after 4800 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.006893 after 4864 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.014349 after 4864 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003819 after 4928 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.016225 after 4928 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003832 after 4992 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.016431 after 4992 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.003324 after 5056 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.014183 after 5056 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004147 after 5120 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.012560 after 5120 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.003184 after 5184 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.011596 after 5184 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002396 after 5248 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.011483 after 5248 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.001613 after 5312 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.011572 after 5312 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.003335 after 5376 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.012469 after 5376 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.001717 after 5440 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.012740 after 5440 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.000818 after 5504 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.012415 after 5504 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.001633 after 5568 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.011804 after 5568 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.001343 after 5632 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.011565 after 5632 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.000485 after 5696 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.011681 after 5696 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.002195 after 5760 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.006748 after 5760 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.002767 after 5824 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005124 after 5824 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.002102 after 5888 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005804 after 5888 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.002656 after 5952 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005564 after 5952 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.002618 after 6016 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005871 after 6016 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.003497 after 6080 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.006721 after 6080 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.003861 after 6144 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.008626 after 6144 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.004709 after 6208 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.007694 after 6208 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.003794 after 6272 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.008406 after 6272 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003095 after 6336 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005480 after 6336 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.004428 after 6400 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.006353 after 6400 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003367 after 6464 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005572 after 6464 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003051 after 6528 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.006030 after 6528 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001930 after 6592 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.006780 after 6592 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003589 after 6656 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.005639 after 6656 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.004062 after 6720 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.006802 after 6720 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.005021 after 6784 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.004907 after 6784 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.005686 after 6848 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.005269 after 6848 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.004544 after 6912 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.004002 after 6912 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003314 after 6976 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.004577 after 6976 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.002997 after 7040 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.004130 after 7040 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.002281 after 7104 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.003359 after 7104 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.173\n", - "[mcmc] - Convergence of means: R-1 = 0.003783 after 7168 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.003473 after 7168 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.003248 after 7232 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.002242 after 7232 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001818 after 7296 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.002827 after 7296 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001399 after 7360 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.002494 after 7360 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001683 after 7424 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.001662 after 7424 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001180 after 7488 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.002151 after 7488 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.002225 after 7552 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002793 after 7552 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001517 after 7616 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003946 after 7616 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001370 after 7680 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004203 after 7680 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001514 after 7744 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004841 after 7744 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001332 after 7808 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004519 after 7808 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.001517 after 7872 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004892 after 7872 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.000500 after 7936 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005868 after 7936 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.172\n", - "[mcmc] - Convergence of means: R-1 = 0.000646 after 8000 accepted steps\n", - "[mcmc] - Convergence of bounds: R-1 = 0.036217 after 10000 accepted steps\n", - "[mcmc] The run has converged!\n", - "[mcmc] Sampling complete after 10000 accepted steps.\n" - ] - } - ], - "source": [ - "info[\"prior\"] = {\"x_eq_y_band\": lambda x, y: stats.norm.logpdf(x - y, loc=0, scale=0.3)}\n", - "\n", - "from cobaya import run\n", - "updated_info_x_eq_y, sampler_x_eq_y = run(info)" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": { - "image/png": { - "height": 488, - "width": 489 - } - }, - "output_type": "display_data" - } - ], - "source": [ - "# load samples, removing first 30% as burn in\n", - "gdsamples_x_eq_y = sampler_x_eq_y.products(to_getdist=True, skip_samples=0.3)[\"sample\"]\n", - "gdplot = gdplt.get_subplot_plotter(width_inch=5)\n", - "gdplot.triangle_plot(gdsamples_x_eq_y, [\"x\", \"y\"], filled=True)\n", - "gdplot.export(\"example_adv_band.png\")" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "collapsed": true, - "jupyter": { - "outputs_hidden": true - } - }, - "source": [ - "## Alternative: $r$ and $\\theta$ as derived parameters of the likelihood" - ] - }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": { - "scrolled": true - }, + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005125 after 8000 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005526 after 8064 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.008050 after 8128 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.007630 after 8192 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.006623 after 8256 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.006166 after 8320 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005382 after 8384 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003095 after 8448 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003347 after 8512 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002809 after 8576 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.001833 after 8640 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002182 after 8704 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 10960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002876 after 8768 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003567 after 8832 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002696 after 8896 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003508 after 8960 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002674 after 9024 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002205 after 9088 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002818 after 9152 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003031 after 9216 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002965 after 9280 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002787 after 9344 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003446 after 9408 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003005 after 9472 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 11920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.002624 after 9536 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003148 after 9600 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004497 after 9664 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.005824 after 9728 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.004860 after 9792 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.005840 after 9856 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.005946 after 9920 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.006701 after 9984 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.007136 after 10048 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.005283 after 10112 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.004954 after 10176 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.004157 after 10240 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003356 after 10304 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 12960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002675 after 10368 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002262 after 10432 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002874 after 10496 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003498 after 10560 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002518 after 10624 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002297 after 10688 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002092 after 10752 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001991 after 10816 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002277 after 10880 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002654 after 10944 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003231 after 11008 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002406 after 11072 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 13920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002392 after 11136 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002694 after 11200 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003022 after 11264 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002986 after 11328 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002981 after 11392 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002897 after 11456 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002467 after 11520 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.002202 after 11584 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002106 after 11648 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002293 after 11712 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002557 after 11776 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002884 after 11840 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003232 after 11904 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 14960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004273 after 11968 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004662 after 12032 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003850 after 12096 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003775 after 12160 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004048 after 12224 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003693 after 12288 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003631 after 12352 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004076 after 12416 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004586 after 12480 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004768 after 12544 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005436 after 12608 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005342 after 12672 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 15920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005450 after 12736 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005052 after 12800 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005295 after 12864 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004712 after 12928 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005247 after 12992 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004832 after 13056 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004003 after 13120 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004226 after 13184 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004036 after 13248 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004285 after 13312 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004222 after 13376 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004126 after 13440 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004019 after 13504 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 16960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.004494 after 13568 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.004145 after 13632 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.004067 after 13696 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.003827 after 13760 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003195 after 13824 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003378 after 13888 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003390 after 13952 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003240 after 14016 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.003400 after 14080 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002630 after 14144 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.001707 after 14208 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002161 after 14272 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 17920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002543 after 14336 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002403 after 14400 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002125 after 14464 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002255 after 14528 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002540 after 14592 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003362 after 14656 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003418 after 14720 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002717 after 14784 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002783 after 14848 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003099 after 14912 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003579 after 14976 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003968 after 15040 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004399 after 15104 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 18960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004771 after 15168 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004464 after 15232 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003857 after 15296 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003558 after 15360 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003827 after 15424 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004742 after 15488 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005091 after 15552 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005932 after 15616 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005830 after 15680 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006057 after 15744 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006724 after 15808 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.007455 after 15872 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 19920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006803 after 15936 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006274 after 16000 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006503 after 16064 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005717 after 16128 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005191 after 16192 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004699 after 16256 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004654 after 16320 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003839 after 16384 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003835 after 16448 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003360 after 16512 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003326 after 16576 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003641 after 16640 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003417 after 16704 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 20960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003196 after 16768 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003564 after 16832 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004106 after 16896 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004273 after 16960 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004779 after 17024 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004587 after 17088 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003986 after 17152 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003311 after 17216 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003013 after 17280 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003609 after 17344 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003456 after 17408 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002981 after 17472 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 21920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002869 after 17536 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002902 after 17600 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.002611 after 17664 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002423 after 17728 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002407 after 17792 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002026 after 17856 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001765 after 17920 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001625 after 17984 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001581 after 18048 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001359 after 18112 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001158 after 18176 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001277 after 18240 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001506 after 18304 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 22960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001582 after 18368 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001748 after 18432 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001595 after 18496 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001790 after 18560 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001778 after 18624 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002194 after 18688 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002763 after 18752 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002664 after 18816 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002636 after 18880 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002593 after 18944 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002685 after 19008 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002715 after 19072 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 23920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003399 after 19136 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003612 after 19200 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003638 after 19264 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003987 after 19328 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003738 after 19392 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004147 after 19456 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004464 after 19520 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004692 after 19584 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004337 after 19648 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004099 after 19712 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004741 after 19776 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004687 after 19840 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004701 after 19904 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 24960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005083 after 19968 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005530 after 20032 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005860 after 20096 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005615 after 20160 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005144 after 20224 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004883 after 20288 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004422 after 20352 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003966 after 20416 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.003656 after 20480 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004040 after 20544 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004668 after 20608 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004638 after 20672 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 25920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004248 after 20736 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004259 after 20800 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004522 after 20864 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004647 after 20928 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004375 after 20992 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004702 after 21056 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004853 after 21120 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004297 after 21184 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003175 after 21248 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002883 after 21312 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002784 after 21376 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002568 after 21440 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002780 after 21504 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 26960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002646 after 21568 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003025 after 21632 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003185 after 21696 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002826 after 21760 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002651 after 21824 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002682 after 21888 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002976 after 21952 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002932 after 22016 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002595 after 22080 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002459 after 22144 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002477 after 22208 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002387 after 22272 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 27920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002484 after 22336 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002571 after 22400 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28080 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002454 after 22464 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28160 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003118 after 22528 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28240 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003096 after 22592 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28320 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002908 after 22656 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28400 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003203 after 22720 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28480 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003187 after 22784 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28560 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002732 after 22848 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28640 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002507 after 22912 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002274 after 22976 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002203 after 23040 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.002156 after 23104 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 28960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001551 after 23168 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29040 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.001895 after 23232 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29120 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001783 after 23296 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29200 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001670 after 23360 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29280 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001478 after 23424 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29360 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001615 after 23488 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29440 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001691 after 23552 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29520 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001552 after 23616 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29600 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001580 after 23680 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29680 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001328 after 23744 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29760 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001091 after 23808 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29840 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.001113 after 23872 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 29920 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.000795 after 23936 accepted steps\n", + "[mcmc] - Updated covariance matrix of proposal pdf.\n", + "[mcmc] Learn + convergence test @ 30000 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.000738 after 24000 accepted steps\n", + "[mcmc] - Convergence of bounds: R-1 = 0.027844 after 30000 accepted steps\n", + "[mcmc] The run has converged!\n", + "[mcmc] Sampling complete after 30000 accepted steps.\n" + ] + } + ], + "source": [ + "info[\"prior\"] = {\"x_eq_y_band\": lambda x, y: stats.norm.logpdf(x - y, loc=0, scale=0.3)}\n", + "\n", + "from cobaya import run\n", + "updated_info_x_eq_y, sampler_x_eq_y = run(info)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "image/png": { + "height": 488, + "width": 489 + } + }, + "output_type": "display_data" + } + ], + "source": [ + "# load samples, removing first 30% as burn in\n", + "gdsamples_x_eq_y = sampler_x_eq_y.products(to_getdist=True, skip_samples=0.3)[\"sample\"]\n", + "gdplot = gdplt.get_subplot_plotter(width_inch=5)\n", + "gdplot.triangle_plot(gdsamples_x_eq_y, [\"x\", \"y\"], filled=True)\n", + "gdplot.export(\"example_adv_band.png\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": true, + "jupyter": { + "outputs_hidden": true + } + }, + "source": [ + "## Alternative: $r$ and $\\theta$ as derived parameters of the likelihood" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "scrolled": true + }, "outputs": [ { "name": "stdout", @@ -1517,1148 +2285,860 @@ "[mcmc] Getting initial point... (this may take a few seconds)\n", "[mcmc] Initial point: x:0.5, y:0.5\n", "[model] Measuring speeds... (this may take a few seconds)\n", - "[model] Setting measured speeds (per sec): {ring: 14900.0}\n", + "[model] Setting measured speeds (per sec): {ring: 12400.0}\n", "[mcmc] Covariance matrix not present. We will start learning the covariance of the proposal earlier: R-1 = 30 (would be 2 if all params loaded).\n", "[mcmc] Sampling!\n", - "[mcmc] Progress @ 2024-08-09 09:38:43 : 1 steps taken, and 0 accepted.\n", + "[mcmc] Progress @ 2024-08-09 15:14:03 : 1 steps taken, and 0 accepted.\n", "[mcmc] Learn + convergence test @ 80 samples accepted.\n", "[mcmc] - Acceptance rate: 0.660\n", - "[mcmc] - Convergence of means: R-1 = 4.057590 after 64 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 4.216064 after 64 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.479\n", - "[mcmc] - Convergence of means: R-1 = 0.625906 after 128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.607\n", + "[mcmc] - Convergence of means: R-1 = 1.282216 after 128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.399\n", - "[mcmc] - Convergence of means: R-1 = 0.125060 after 192 accepted steps\n", + "[mcmc] - Acceptance rate: 0.471\n", + "[mcmc] - Convergence of means: R-1 = 0.397623 after 192 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.347\n", - "[mcmc] - Convergence of means: R-1 = 0.197280 after 256 accepted steps\n", + "[mcmc] - Acceptance rate: 0.412\n", + "[mcmc] - Convergence of means: R-1 = 0.273554 after 256 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.310\n", - "[mcmc] - Convergence of means: R-1 = 0.328387 after 320 accepted steps\n", + "[mcmc] - Acceptance rate: 0.350\n", + "[mcmc] - Convergence of means: R-1 = 0.090995 after 320 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.262\n", - "[mcmc] - Convergence of means: R-1 = 0.343965 after 384 accepted steps\n", + "[mcmc] - Acceptance rate: 0.323\n", + "[mcmc] - Convergence of means: R-1 = 0.118634 after 384 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.243\n", - "[mcmc] - Convergence of means: R-1 = 0.160004 after 448 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.078772 after 448 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.236\n", - "[mcmc] - Convergence of means: R-1 = 0.140111 after 512 accepted steps\n", + "[mcmc] - Acceptance rate: 0.281\n", + "[mcmc] - Convergence of means: R-1 = 0.067353 after 512 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.218\n", - "[mcmc] - Convergence of means: R-1 = 0.127194 after 576 accepted steps\n", + "[mcmc] - Acceptance rate: 0.268\n", + "[mcmc] - Convergence of means: R-1 = 0.058858 after 576 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.217\n", - "[mcmc] - Convergence of means: R-1 = 0.024876 after 640 accepted steps\n", + "[mcmc] - Acceptance rate: 0.260\n", + "[mcmc] - Convergence of means: R-1 = 0.061793 after 640 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.211\n", - "[mcmc] - Convergence of means: R-1 = 0.064113 after 704 accepted steps\n", + "[mcmc] - Acceptance rate: 0.254\n", + "[mcmc] - Convergence of means: R-1 = 0.060368 after 704 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.203\n", - "[mcmc] - Convergence of means: R-1 = 0.096964 after 768 accepted steps\n", + "[mcmc] - Acceptance rate: 0.248\n", + "[mcmc] - Convergence of means: R-1 = 0.028681 after 768 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.197\n", - "[mcmc] - Convergence of means: R-1 = 0.076215 after 832 accepted steps\n", + "[mcmc] - Acceptance rate: 0.243\n", + "[mcmc] - Convergence of means: R-1 = 0.026099 after 832 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.194\n", - "[mcmc] - Convergence of means: R-1 = 0.030546 after 896 accepted steps\n", + "[mcmc] - Acceptance rate: 0.240\n", + "[mcmc] - Convergence of means: R-1 = 0.046500 after 896 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.022476 after 960 accepted steps\n", + "[mcmc] - Acceptance rate: 0.230\n", + "[mcmc] - Convergence of means: R-1 = 0.043646 after 960 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.015050 after 1024 accepted steps\n", + "[mcmc] - Acceptance rate: 0.224\n", + "[mcmc] - Convergence of means: R-1 = 0.010687 after 1024 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.008218 after 1088 accepted steps\n", + "[mcmc] - Acceptance rate: 0.222\n", + "[mcmc] - Convergence of means: R-1 = 0.007791 after 1088 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.005482 after 1152 accepted steps\n", + "[mcmc] - Acceptance rate: 0.222\n", + "[mcmc] - Convergence of means: R-1 = 0.009398 after 1152 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.024816 after 1216 accepted steps\n", + "[mcmc] - Acceptance rate: 0.221\n", + "[mcmc] - Convergence of means: R-1 = 0.007502 after 1216 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.065226 after 1280 accepted steps\n", + "[mcmc] - Acceptance rate: 0.219\n", + "[mcmc] - Convergence of means: R-1 = 0.006141 after 1280 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.073834 after 1344 accepted steps\n", + "[mcmc] - Acceptance rate: 0.218\n", + "[mcmc] - Convergence of means: R-1 = 0.006892 after 1344 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.022047 after 1408 accepted steps\n", + "[mcmc] - Acceptance rate: 0.219\n", + "[mcmc] - Convergence of means: R-1 = 0.014339 after 1408 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.019303 after 1472 accepted steps\n", + "[mcmc] - Acceptance rate: 0.218\n", + "[mcmc] - Convergence of means: R-1 = 0.014733 after 1472 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.024009 after 1536 accepted steps\n", + "[mcmc] - Acceptance rate: 0.215\n", + "[mcmc] - Convergence of means: R-1 = 0.008307 after 1536 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.039192 after 1600 accepted steps\n", + "[mcmc] - Acceptance rate: 0.215\n", + "[mcmc] - Convergence of means: R-1 = 0.012009 after 1600 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.057054 after 1664 accepted steps\n", + "[mcmc] - Acceptance rate: 0.214\n", + "[mcmc] - Convergence of means: R-1 = 0.024034 after 1664 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.054239 after 1728 accepted steps\n", + "[mcmc] - Acceptance rate: 0.212\n", + "[mcmc] - Convergence of means: R-1 = 0.011217 after 1728 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.034443 after 1792 accepted steps\n", + "[mcmc] - Acceptance rate: 0.210\n", + "[mcmc] - Convergence of means: R-1 = 0.010158 after 1792 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.020870 after 1856 accepted steps\n", + "[mcmc] - Acceptance rate: 0.210\n", + "[mcmc] - Convergence of means: R-1 = 0.008392 after 1856 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.008405 after 1920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.210\n", + "[mcmc] - Convergence of means: R-1 = 0.012585 after 1920 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.194\n", - "[mcmc] - Convergence of means: R-1 = 0.005660 after 1984 accepted steps\n", + "[mcmc] - Acceptance rate: 0.209\n", + "[mcmc] - Convergence of means: R-1 = 0.013590 after 1984 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.005744 after 2048 accepted steps\n", + "[mcmc] - Acceptance rate: 0.208\n", + "[mcmc] - Convergence of means: R-1 = 0.008465 after 2048 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.013320 after 2112 accepted steps\n", + "[mcmc] - Acceptance rate: 0.207\n", + "[mcmc] - Convergence of means: R-1 = 0.010054 after 2112 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.013439 after 2176 accepted steps\n", + "[mcmc] - Acceptance rate: 0.206\n", + "[mcmc] - Convergence of means: R-1 = 0.013183 after 2176 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.010923 after 2240 accepted steps\n", + "[mcmc] - Acceptance rate: 0.204\n", + "[mcmc] - Convergence of means: R-1 = 0.016063 after 2240 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.011884 after 2304 accepted steps\n", + "[mcmc] - Acceptance rate: 0.202\n", + "[mcmc] - Convergence of means: R-1 = 0.006625 after 2304 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.194\n", - "[mcmc] - Convergence of means: R-1 = 0.008684 after 2368 accepted steps\n", + "[mcmc] - Acceptance rate: 0.202\n", + "[mcmc] - Convergence of means: R-1 = 0.007850 after 2368 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.011378 after 2432 accepted steps\n", + "[mcmc] - Acceptance rate: 0.202\n", + "[mcmc] - Convergence of means: R-1 = 0.006210 after 2432 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.019896 after 2496 accepted steps\n", + "[mcmc] - Acceptance rate: 0.200\n", + "[mcmc] - Convergence of means: R-1 = 0.008302 after 2496 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.025008 after 2560 accepted steps\n", + "[mcmc] - Acceptance rate: 0.199\n", + "[mcmc] - Convergence of means: R-1 = 0.012942 after 2560 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.024482 after 2624 accepted steps\n", + "[mcmc] - Acceptance rate: 0.198\n", + "[mcmc] - Convergence of means: R-1 = 0.011173 after 2624 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.193\n", - "[mcmc] - Convergence of means: R-1 = 0.018174 after 2688 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.014484 after 2688 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.192\n", - "[mcmc] - Convergence of means: R-1 = 0.010338 after 2752 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.008461 after 2752 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.005501 after 2816 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.009024 after 2816 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.006378 after 2880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.010454 after 2880 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.004518 after 2944 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.011916 after 2944 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.191\n", - "[mcmc] - Convergence of means: R-1 = 0.004186 after 3008 accepted steps\n", + "[mcmc] - Acceptance rate: 0.196\n", + "[mcmc] - Convergence of means: R-1 = 0.011760 after 3008 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.008537 after 3072 accepted steps\n", + "[mcmc] - Acceptance rate: 0.195\n", + "[mcmc] - Convergence of means: R-1 = 0.008250 after 3072 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.008949 after 3136 accepted steps\n", + "[mcmc] - Acceptance rate: 0.195\n", + "[mcmc] - Convergence of means: R-1 = 0.011390 after 3136 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.011153 after 3200 accepted steps\n", + "[mcmc] - Acceptance rate: 0.193\n", + "[mcmc] - Convergence of means: R-1 = 0.010055 after 3200 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.016791 after 3264 accepted steps\n", + "[mcmc] - Acceptance rate: 0.193\n", + "[mcmc] - Convergence of means: R-1 = 0.006458 after 3264 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.012787 after 3328 accepted steps\n", + "[mcmc] - Acceptance rate: 0.193\n", + "[mcmc] - Convergence of means: R-1 = 0.007350 after 3328 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.011183 after 3392 accepted steps\n", + "[mcmc] - Acceptance rate: 0.192\n", + "[mcmc] - Convergence of means: R-1 = 0.007275 after 3392 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.007699 after 3456 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.007019 after 3456 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.005074 after 3520 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.004576 after 3520 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.006968 after 3584 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.006705 after 3584 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.189\n", - "[mcmc] - Convergence of means: R-1 = 0.007269 after 3648 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.005638 after 3648 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.190\n", - "[mcmc] - Convergence of means: R-1 = 0.006993 after 3712 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.005668 after 3712 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.006492 after 3776 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.008082 after 3776 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.010356 after 3840 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.008782 after 3840 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.007349 after 3904 accepted steps\n", + "[mcmc] - Acceptance rate: 0.191\n", + "[mcmc] - Convergence of means: R-1 = 0.009024 after 3904 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.006682 after 3968 accepted steps\n", + "[mcmc] - Acceptance rate: 0.190\n", + "[mcmc] - Convergence of means: R-1 = 0.009559 after 3968 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.002893 after 4032 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.008551 after 4032 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.003804 after 4096 accepted steps\n", + "[mcmc] - Acceptance rate: 0.189\n", + "[mcmc] - Convergence of means: R-1 = 0.005942 after 4096 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5200 samples accepted.\n", "[mcmc] - Acceptance rate: 0.188\n", - "[mcmc] - Convergence of means: R-1 = 0.003617 after 4160 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.009502 after 4160 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.004083 after 4224 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.008728 after 4224 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.003918 after 4288 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.006023 after 4288 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5440 samples accepted.\n", "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.007343 after 4352 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.004052 after 4352 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.186\n", - "[mcmc] - Convergence of means: R-1 = 0.006027 after 4416 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.004659 after 4416 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5600 samples accepted.\n", "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.006909 after 4480 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.003468 after 4480 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5680 samples accepted.\n", "[mcmc] - Acceptance rate: 0.187\n", - "[mcmc] - Convergence of means: R-1 = 0.010158 after 4544 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.002939 after 4544 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.186\n", - "[mcmc] - Convergence of means: R-1 = 0.007671 after 4608 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.003040 after 4608 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.185\n", - "[mcmc] - Convergence of means: R-1 = 0.008531 after 4672 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.001309 after 4672 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 5920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.185\n", - "[mcmc] - Convergence of means: R-1 = 0.014211 after 4736 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001405 after 4736 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.185\n", - "[mcmc] - Convergence of means: R-1 = 0.012870 after 4800 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001102 after 4800 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.185\n", - "[mcmc] - Convergence of means: R-1 = 0.012899 after 4864 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001967 after 4864 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.014443 after 4928 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001266 after 4928 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.013124 after 4992 accepted steps\n", + "[mcmc] - Acceptance rate: 0.188\n", + "[mcmc] - Convergence of means: R-1 = 0.000437 after 4992 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.006513 after 5056 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001056 after 5056 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.004874 after 5120 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.000620 after 5120 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.184\n", - "[mcmc] - Convergence of means: R-1 = 0.005462 after 5184 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001837 after 5184 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.183\n", - "[mcmc] - Convergence of means: R-1 = 0.004326 after 5248 accepted steps\n", + "[mcmc] - Acceptance rate: 0.187\n", + "[mcmc] - Convergence of means: R-1 = 0.001371 after 5248 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.183\n", - "[mcmc] - Convergence of means: R-1 = 0.004316 after 5312 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.001562 after 5312 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.183\n", - "[mcmc] - Convergence of means: R-1 = 0.002554 after 5376 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002976 after 5376 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.182\n", - "[mcmc] - Convergence of means: R-1 = 0.001582 after 5440 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.002018 after 5440 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.182\n", - "[mcmc] - Convergence of means: R-1 = 0.001625 after 5504 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.005083 after 5504 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 6960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.182\n", - "[mcmc] - Convergence of means: R-1 = 0.002093 after 5568 accepted steps\n", + "[mcmc] - Acceptance rate: 0.186\n", + "[mcmc] - Convergence of means: R-1 = 0.007129 after 5568 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.181\n", - "[mcmc] - Convergence of means: R-1 = 0.002427 after 5632 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.005760 after 5632 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.003879 after 5696 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.006171 after 5696 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.004208 after 5760 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004153 after 5760 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.005789 after 5824 accepted steps\n", + "[mcmc] - Acceptance rate: 0.185\n", + "[mcmc] - Convergence of means: R-1 = 0.004264 after 5824 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.004985 after 5888 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004634 after 5888 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.005706 after 5952 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004903 after 5952 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.004835 after 6016 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.004077 after 6016 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.004241 after 6080 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.004108 after 6080 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.004133 after 6144 accepted steps\n", + "[mcmc] - Acceptance rate: 0.184\n", + "[mcmc] - Convergence of means: R-1 = 0.003550 after 6144 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.004470 after 6208 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.003707 after 6208 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.180\n", - "[mcmc] - Convergence of means: R-1 = 0.006229 after 6272 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.004501 after 6272 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 7920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.007944 after 6336 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.005912 after 6336 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.008986 after 6400 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.005232 after 6400 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.011205 after 6464 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.005462 after 6464 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.013475 after 6528 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.004058 after 6528 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.011333 after 6592 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.003187 after 6592 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.011098 after 6656 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.003279 after 6656 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.008289 after 6720 accepted steps\n", + "[mcmc] - Acceptance rate: 0.183\n", + "[mcmc] - Convergence of means: R-1 = 0.003738 after 6720 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.007392 after 6784 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.002116 after 6784 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.006794 after 6848 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002414 after 6848 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.005793 after 6912 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.005097 after 6912 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.179\n", - "[mcmc] - Convergence of means: R-1 = 0.005783 after 6976 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.007096 after 6976 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004502 after 7040 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.005941 after 7040 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004956 after 7104 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.004673 after 7104 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 8960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003668 after 7168 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.004968 after 7168 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003715 after 7232 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003671 after 7232 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004007 after 7296 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002549 after 7296 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003215 after 7360 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002536 after 7360 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004171 after 7424 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002590 after 7424 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003998 after 7488 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003314 after 7488 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003692 after 7552 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.005040 after 7552 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003690 after 7616 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.004704 after 7616 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003333 after 7680 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.003020 after 7680 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004493 after 7744 accepted steps\n", + "[mcmc] - Acceptance rate: 0.182\n", + "[mcmc] - Convergence of means: R-1 = 0.001730 after 7744 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.004132 after 7808 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.001791 after 7808 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004372 after 7872 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003823 after 7872 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 9920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004077 after 7936 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.005447 after 7936 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004529 after 8000 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003073 after 8000 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003956 after 8064 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003114 after 8064 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.004653 after 8128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003807 after 8128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.003100 after 8192 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003234 after 8192 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.002883 after 8256 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002622 after 8256 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.002519 after 8320 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.001807 after 8320 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.002750 after 8384 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002271 after 8384 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.178\n", - "[mcmc] - Convergence of means: R-1 = 0.002609 after 8448 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003492 after 8448 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.003316 after 8512 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002351 after 8512 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.005088 after 8576 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003511 after 8576 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.006260 after 8640 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.004958 after 8640 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.007032 after 8704 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.006246 after 8704 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 10960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.006047 after 8768 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.005981 after 8768 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.005174 after 8832 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003060 after 8832 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.177\n", - "[mcmc] - Convergence of means: R-1 = 0.005536 after 8896 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002648 after 8896 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.006265 after 8960 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.003231 after 8960 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.006080 after 9024 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002072 after 9024 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.005361 after 9088 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.001726 after 9088 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.005797 after 9152 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002003 after 9152 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004737 after 9216 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002929 after 9216 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.005000 after 9280 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003398 after 9280 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.005677 after 9344 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003039 after 9344 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007101 after 9408 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003347 after 9408 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007275 after 9472 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002898 after 9472 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 11920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007344 after 9536 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002469 after 9536 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007161 after 9600 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002133 after 9600 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.006501 after 9664 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002151 after 9664 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.006637 after 9728 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002251 after 9728 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.006533 after 9792 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001711 after 9792 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.007208 after 9856 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001471 after 9856 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.006181 after 9920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001301 after 9920 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.006377 after 9984 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001631 after 9984 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.006805 after 10048 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001215 after 10048 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004547 after 10112 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002251 after 10112 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004784 after 10176 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002441 after 10176 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.003796 after 10240 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.002530 after 10240 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004155 after 10304 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001815 after 10304 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 12960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.004238 after 10368 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001450 after 10368 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.003301 after 10432 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001296 after 10432 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.003617 after 10496 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.001995 after 10496 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003932 after 10560 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003202 after 10560 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004930 after 10624 accepted steps\n", + "[mcmc] - Acceptance rate: 0.181\n", + "[mcmc] - Convergence of means: R-1 = 0.002867 after 10624 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004949 after 10688 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003065 after 10688 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004134 after 10752 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003519 after 10752 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003713 after 10816 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.004379 after 10816 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003397 after 10880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.180\n", + "[mcmc] - Convergence of means: R-1 = 0.003953 after 10880 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003499 after 10944 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.003287 after 10944 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003073 after 11008 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001811 after 11008 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003484 after 11072 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001794 after 11072 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 13920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003330 after 11136 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001707 after 11136 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002856 after 11200 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001229 after 11200 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003377 after 11264 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001627 after 11264 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003262 after 11328 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001319 after 11328 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003353 after 11392 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001296 after 11392 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003302 after 11456 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001198 after 11456 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.002812 after 11520 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001795 after 11520 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003043 after 11584 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002905 after 11584 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003304 after 11648 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002895 after 11648 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003178 after 11712 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002643 after 11712 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003303 after 11776 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002812 after 11776 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003686 after 11840 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002192 after 11840 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003632 after 11904 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002196 after 11904 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 14960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004590 after 11968 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002343 after 11968 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004171 after 12032 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002115 after 12032 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.005136 after 12096 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.002124 after 12096 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003544 after 12160 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001946 after 12160 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003027 after 12224 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001888 after 12224 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003115 after 12288 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001835 after 12288 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002561 after 12352 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001943 after 12352 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002140 after 12416 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001358 after 12416 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002239 after 12480 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001411 after 12480 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002326 after 12544 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001127 after 12544 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002132 after 12608 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001018 after 12608 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002370 after 12672 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.000975 after 12672 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 15920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002558 after 12736 accepted steps\n", + "[mcmc] - Acceptance rate: 0.179\n", + "[mcmc] - Convergence of means: R-1 = 0.001219 after 12736 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003282 after 12800 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001232 after 12800 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003757 after 12864 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001282 after 12864 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004677 after 12928 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001060 after 12928 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004618 after 12992 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001091 after 12992 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.005116 after 13056 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001216 after 13056 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004112 after 13120 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001094 after 13120 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003784 after 13184 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.000996 after 13184 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003920 after 13248 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001037 after 13248 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 16640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003926 after 13312 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 16720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004258 after 13376 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 16800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004410 after 13440 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 16880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003958 after 13504 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 16960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003860 after 13568 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003653 after 13632 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004174 after 13696 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003881 after 13760 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003772 after 13824 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004118 after 13888 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003746 after 13952 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003213 after 14016 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002446 after 14080 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002340 after 14144 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002291 after 14208 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002233 after 14272 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 17920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002140 after 14336 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002224 after 14400 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002647 after 14464 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003267 after 14528 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003496 after 14592 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003346 after 14656 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003379 after 14720 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004168 after 14784 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004631 after 14848 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004689 after 14912 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004691 after 14976 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004206 after 15040 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.004979 after 15104 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 18960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.002786 after 15168 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002704 after 15232 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002610 after 15296 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002527 after 15360 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003201 after 15424 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003542 after 15488 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.003016 after 15552 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.002771 after 15616 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.002419 after 15680 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.174\n", - "[mcmc] - Convergence of means: R-1 = 0.002888 after 15744 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002629 after 15808 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002677 after 15872 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 19920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002845 after 15936 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002983 after 16000 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002491 after 16064 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002595 after 16128 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002732 after 16192 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002783 after 16256 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002811 after 16320 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002078 after 16384 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001738 after 16448 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001732 after 16512 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002273 after 16576 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002022 after 16640 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002143 after 16704 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 20960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002202 after 16768 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002488 after 16832 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002833 after 16896 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002395 after 16960 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002618 after 17024 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002337 after 17088 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002402 after 17152 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.002360 after 17216 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001730 after 17280 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001647 after 17344 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001650 after 17408 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001669 after 17472 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 21920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001693 after 17536 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001408 after 17600 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.176\n", - "[mcmc] - Convergence of means: R-1 = 0.001392 after 17664 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001334 after 17728 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001262 after 17792 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001192 after 17856 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001067 after 17920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.000885 after 13312 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001046 after 17984 accepted steps\n", + "[mcmc] Learn + convergence test @ 16720 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001009 after 13376 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.001034 after 18048 accepted steps\n", + "[mcmc] Learn + convergence test @ 16800 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.001049 after 13440 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.000899 after 18112 accepted steps\n", + "[mcmc] Learn + convergence test @ 16880 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.000392 after 13504 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 22720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.175\n", - "[mcmc] - Convergence of means: R-1 = 0.000979 after 18176 accepted steps\n", - "[mcmc] - Convergence of bounds: R-1 = 0.017527 after 22720 accepted steps\n", + "[mcmc] Learn + convergence test @ 16960 samples accepted.\n", + "[mcmc] - Acceptance rate: 0.178\n", + "[mcmc] - Convergence of means: R-1 = 0.000327 after 13568 accepted steps\n", + "[mcmc] - Convergence of bounds: R-1 = 0.022699 after 16960 accepted steps\n", "[mcmc] The run has converged!\n", - "[mcmc] Sampling complete after 22720 accepted steps.\n" + "[mcmc] Sampling complete after 16960 accepted steps.\n" ] } ], @@ -2696,7 +3176,7 @@ "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -2720,7 +3200,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -2766,424 +3246,228 @@ "[mcmc] Getting initial point... (this may take a few seconds)\n", "[mcmc] Initial point: r:1, theta:0\n", "[model] Measuring speeds... (this may take a few seconds)\n", - "[model] Setting measured speeds (per sec): {ring: 13300.0}\n", + "[model] Setting measured speeds (per sec): {ring: 13500.0}\n", "[mcmc] Covariance matrix not present. We will start learning the covariance of the proposal earlier: R-1 = 30 (would be 2 if all params loaded).\n", "[mcmc] Sampling!\n", - "[mcmc] Progress @ 2024-08-09 09:39:06 : 1 steps taken, and 0 accepted.\n", + "[mcmc] Progress @ 2024-08-09 15:14:22 : 1 steps taken, and 0 accepted.\n", "[mcmc] Learn + convergence test @ 80 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.255\n", - "[mcmc] - Convergence of means: R-1 = 0.417672 after 64 accepted steps\n", + "[mcmc] - Acceptance rate: 0.233\n", + "[mcmc] - Convergence of means: R-1 = 0.672297 after 64 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.322\n", - "[mcmc] - Convergence of means: R-1 = 0.237821 after 128 accepted steps\n", + "[mcmc] - Acceptance rate: 0.281\n", + "[mcmc] - Convergence of means: R-1 = 0.199335 after 128 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.149561 after 192 accepted steps\n", + "[mcmc] - Acceptance rate: 0.296\n", + "[mcmc] - Convergence of means: R-1 = 0.067744 after 192 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.291\n", - "[mcmc] - Convergence of means: R-1 = 0.169192 after 256 accepted steps\n", + "[mcmc] - Acceptance rate: 0.285\n", + "[mcmc] - Convergence of means: R-1 = 0.057994 after 256 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.015867 after 320 accepted steps\n", + "[mcmc] - Acceptance rate: 0.288\n", + "[mcmc] - Convergence of means: R-1 = 0.059267 after 320 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.039520 after 384 accepted steps\n", + "[mcmc] - Acceptance rate: 0.292\n", + "[mcmc] - Convergence of means: R-1 = 0.006517 after 384 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.042018 after 448 accepted steps\n", + "[mcmc] - Acceptance rate: 0.301\n", + "[mcmc] - Convergence of means: R-1 = 0.076374 after 448 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.294\n", - "[mcmc] - Convergence of means: R-1 = 0.066488 after 512 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.019831 after 512 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.051451 after 576 accepted steps\n", + "[mcmc] - Acceptance rate: 0.297\n", + "[mcmc] - Convergence of means: R-1 = 0.017072 after 576 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.282\n", - "[mcmc] - Convergence of means: R-1 = 0.041952 after 640 accepted steps\n", + "[mcmc] - Acceptance rate: 0.292\n", + "[mcmc] - Convergence of means: R-1 = 0.031449 after 640 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.282\n", - "[mcmc] - Convergence of means: R-1 = 0.038841 after 704 accepted steps\n", + "[mcmc] - Acceptance rate: 0.292\n", + "[mcmc] - Convergence of means: R-1 = 0.022334 after 704 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.285\n", - "[mcmc] - Convergence of means: R-1 = 0.023623 after 768 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.013646 after 768 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.289\n", - "[mcmc] - Convergence of means: R-1 = 0.016116 after 832 accepted steps\n", + "[mcmc] - Acceptance rate: 0.290\n", + "[mcmc] - Convergence of means: R-1 = 0.007357 after 832 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.030891 after 896 accepted steps\n", + "[mcmc] - Acceptance rate: 0.291\n", + "[mcmc] - Convergence of means: R-1 = 0.016123 after 896 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.292\n", - "[mcmc] - Convergence of means: R-1 = 0.027060 after 960 accepted steps\n", + "[mcmc] - Acceptance rate: 0.289\n", + "[mcmc] - Convergence of means: R-1 = 0.017400 after 960 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.291\n", - "[mcmc] - Convergence of means: R-1 = 0.010647 after 1024 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.014329 after 1024 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.291\n", - "[mcmc] - Convergence of means: R-1 = 0.012560 after 1088 accepted steps\n", + "[mcmc] - Acceptance rate: 0.302\n", + "[mcmc] - Convergence of means: R-1 = 0.009698 after 1088 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.289\n", - "[mcmc] - Convergence of means: R-1 = 0.013742 after 1152 accepted steps\n", + "[mcmc] - Acceptance rate: 0.305\n", + "[mcmc] - Convergence of means: R-1 = 0.003543 after 1152 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.008243 after 1216 accepted steps\n", + "[mcmc] - Acceptance rate: 0.304\n", + "[mcmc] - Convergence of means: R-1 = 0.016627 after 1216 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.007098 after 1280 accepted steps\n", + "[mcmc] - Acceptance rate: 0.302\n", + "[mcmc] - Convergence of means: R-1 = 0.014153 after 1280 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.016344 after 1344 accepted steps\n", + "[mcmc] - Acceptance rate: 0.301\n", + "[mcmc] - Convergence of means: R-1 = 0.005330 after 1344 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.294\n", - "[mcmc] - Convergence of means: R-1 = 0.006989 after 1408 accepted steps\n", + "[mcmc] - Acceptance rate: 0.303\n", + "[mcmc] - Convergence of means: R-1 = 0.002717 after 1408 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.293\n", - "[mcmc] - Convergence of means: R-1 = 0.006382 after 1472 accepted steps\n", + "[mcmc] - Acceptance rate: 0.301\n", + "[mcmc] - Convergence of means: R-1 = 0.002195 after 1472 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 1920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.294\n", - "[mcmc] - Convergence of means: R-1 = 0.017384 after 1536 accepted steps\n", + "[mcmc] - Acceptance rate: 0.298\n", + "[mcmc] - Convergence of means: R-1 = 0.008654 after 1536 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.004281 after 1600 accepted steps\n", + "[mcmc] - Acceptance rate: 0.299\n", + "[mcmc] - Convergence of means: R-1 = 0.006339 after 1600 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.294\n", - "[mcmc] - Convergence of means: R-1 = 0.002011 after 1664 accepted steps\n", + "[mcmc] - Acceptance rate: 0.300\n", + "[mcmc] - Convergence of means: R-1 = 0.008106 after 1664 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.005941 after 1728 accepted steps\n", + "[mcmc] - Acceptance rate: 0.299\n", + "[mcmc] - Convergence of means: R-1 = 0.003025 after 1728 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.292\n", - "[mcmc] - Convergence of means: R-1 = 0.010301 after 1792 accepted steps\n", + "[mcmc] - Acceptance rate: 0.300\n", + "[mcmc] - Convergence of means: R-1 = 0.009877 after 1792 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.292\n", - "[mcmc] - Convergence of means: R-1 = 0.008729 after 1856 accepted steps\n", + "[mcmc] - Acceptance rate: 0.300\n", + "[mcmc] - Convergence of means: R-1 = 0.012106 after 1856 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.292\n", - "[mcmc] - Convergence of means: R-1 = 0.009529 after 1920 accepted steps\n", + "[mcmc] - Acceptance rate: 0.298\n", + "[mcmc] - Convergence of means: R-1 = 0.014341 after 1920 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.292\n", - "[mcmc] - Convergence of means: R-1 = 0.005150 after 1984 accepted steps\n", + "[mcmc] - Acceptance rate: 0.300\n", + "[mcmc] - Convergence of means: R-1 = 0.007829 after 1984 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.294\n", - "[mcmc] - Convergence of means: R-1 = 0.005779 after 2048 accepted steps\n", + "[mcmc] - Acceptance rate: 0.300\n", + "[mcmc] - Convergence of means: R-1 = 0.006487 after 2048 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.003222 after 2112 accepted steps\n", + "[mcmc] - Acceptance rate: 0.298\n", + "[mcmc] - Convergence of means: R-1 = 0.002336 after 2112 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.005019 after 2176 accepted steps\n", + "[mcmc] - Acceptance rate: 0.296\n", + "[mcmc] - Convergence of means: R-1 = 0.006143 after 2176 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.004043 after 2240 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.003613 after 2240 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.006073 after 2304 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.009772 after 2304 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 2960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.006375 after 2368 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.005537 after 2368 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.007279 after 2432 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.004719 after 2432 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.005678 after 2496 accepted steps\n", + "[mcmc] - Acceptance rate: 0.294\n", + "[mcmc] - Convergence of means: R-1 = 0.004082 after 2496 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.012607 after 2560 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.000609 after 2560 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.007040 after 2624 accepted steps\n", + "[mcmc] - Acceptance rate: 0.294\n", + "[mcmc] - Convergence of means: R-1 = 0.001567 after 2624 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.013938 after 2688 accepted steps\n", + "[mcmc] - Acceptance rate: 0.291\n", + "[mcmc] - Convergence of means: R-1 = 0.002956 after 2688 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.008919 after 2752 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.002004 after 2752 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.012795 after 2816 accepted steps\n", + "[mcmc] - Acceptance rate: 0.292\n", + "[mcmc] - Convergence of means: R-1 = 0.001446 after 2816 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.008280 after 2880 accepted steps\n", + "[mcmc] - Acceptance rate: 0.291\n", + "[mcmc] - Convergence of means: R-1 = 0.000743 after 2880 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.007181 after 2944 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.001193 after 2944 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.008900 after 3008 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.002548 after 3008 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.006949 after 3072 accepted steps\n", + "[mcmc] - Acceptance rate: 0.293\n", + "[mcmc] - Convergence of means: R-1 = 0.006011 after 3072 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 3920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.005171 after 3136 accepted steps\n", + "[mcmc] - Acceptance rate: 0.294\n", + "[mcmc] - Convergence of means: R-1 = 0.003374 after 3136 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.002986 after 3200 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.001349 after 3200 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.003759 after 3264 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.000918 after 3264 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.003197 after 3328 accepted steps\n", + "[mcmc] - Acceptance rate: 0.295\n", + "[mcmc] - Convergence of means: R-1 = 0.002217 after 3328 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.007217 after 3392 accepted steps\n", + "[mcmc] - Acceptance rate: 0.294\n", + "[mcmc] - Convergence of means: R-1 = 0.000512 after 3392 accepted steps\n", "[mcmc] - Updated covariance matrix of proposal pdf.\n", "[mcmc] Learn + convergence test @ 4320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.007969 after 3456 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.006087 after 3520 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.007335 after 3584 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.005143 after 3648 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.008002 after 3712 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.300\n", - "[mcmc] - Convergence of means: R-1 = 0.007937 after 3776 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.008392 after 3840 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.009095 after 3904 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 4960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.010298 after 3968 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.003948 after 4032 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.004811 after 4096 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.005192 after 4160 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.004992 after 4224 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.003436 after 4288 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.003404 after 4352 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.003236 after 4416 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.004209 after 4480 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.004339 after 4544 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.004442 after 4608 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.005166 after 4672 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 5920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.005896 after 4736 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.004085 after 4800 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.005044 after 4864 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.007241 after 4928 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6240 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.007582 after 4992 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6320 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.009228 after 5056 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6400 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.007509 after 5120 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6480 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.009997 after 5184 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6560 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.009134 after 5248 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6640 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.004783 after 5312 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6720 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.004570 after 5376 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6800 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.005240 after 5440 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6880 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.003147 after 5504 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 6960 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.003148 after 5568 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7040 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.002668 after 5632 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7120 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.299\n", - "[mcmc] - Convergence of means: R-1 = 0.002301 after 5696 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7200 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.001899 after 5760 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7280 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.002450 after 5824 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7360 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.002573 after 5888 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7440 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.002392 after 5952 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7520 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.298\n", - "[mcmc] - Convergence of means: R-1 = 0.002256 after 6016 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7600 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.002191 after 6080 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7680 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.001097 after 6144 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7760 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.297\n", - "[mcmc] - Convergence of means: R-1 = 0.001130 after 6208 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7840 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.001547 after 6272 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 7920 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.296\n", - "[mcmc] - Convergence of means: R-1 = 0.001251 after 6336 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 8000 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.001451 after 6400 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 8080 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.001110 after 6464 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 8160 samples accepted.\n", - "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.000907 after 6528 accepted steps\n", - "[mcmc] - Updated covariance matrix of proposal pdf.\n", - "[mcmc] Learn + convergence test @ 8240 samples accepted.\n", "[mcmc] - Acceptance rate: 0.295\n", - "[mcmc] - Convergence of means: R-1 = 0.000753 after 6592 accepted steps\n", - "[mcmc] - Convergence of bounds: R-1 = 0.045890 after 8240 accepted steps\n", + "[mcmc] - Convergence of means: R-1 = 0.000761 after 3456 accepted steps\n", + "[mcmc] - Convergence of bounds: R-1 = 0.031435 after 4320 accepted steps\n", "[mcmc] The run has converged!\n", - "[mcmc] Sampling complete after 8240 accepted steps.\n" + "[mcmc] Sampling complete after 4320 accepted steps.\n" ] } ], @@ -3224,9 +3508,16 @@ "execution_count": 12, "metadata": {}, "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[root] *WARNING* outlier fraction 0.0016534391534391533 \n" + ] + }, { "data": { - "image/png": "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", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9MAAAPRCAYAAAD+xftOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAB7CAAAewgFu0HU+AADL5ElEQVR4nOzdd3hc5Zn+8fs958yMpJEsd0u23G2Me8MGYzqmYyChGUwLgSSk10022c2S3WSTbMgm+SXZQEiC6cVU04tDNeCCe8fdcu9WnXLO+f0xkizARWOVad/PdfmywoxGrxRrZu7zPO/zGt/3fQEAAAAAgCazUr0AAAAAAAAyDWEaAAAAAIAkEaYBAAAAAEgSYRoAAAAAgCQRpgEAAAAASBJhGgAAAACAJBGmAQAAAABIEmEaAAAAAIAkEaYBAAAAAEgSYRoAAAAAgCQRpgEAAAAASBJhGgAAAACAJBGmAQAAAABIEmEaAAAAAIAkEaYBAAAAAEgSYRoAAAAAgCQRpgEAAAAASJKT6gWgeWpra7VkyRJJUpcuXeQ4/F8KAEiteDyuXbt2SZKGDx+uvLy8FK8IAICWR/LKcEuWLNH48eNTvQwAAA5rzpw5GjduXKqXAQBAi6PNGwAAAACAJFGZznBdunRp+HjOnDkqLS1N4WoAAJC2bdvW0DXV+HUKAIBsQpjOcI33SJeWlqqsrCyFqwEA4JOY5QEAyFa0eQMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJIkwDAAAAAJAkwjQAAAAAAEkiTAMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJIkwDAAAAAJAkwjQAAAAAAEkiTAMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJIkwDAAAAAJAkwjQAAAAAAEkiTAMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJIkwDAAAAAJAkwjQAAAAAAEkiTAMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJIkwDAAAAAJAkwjQAAAAAAEkiTAMAAAAAkCTCNAAAAAAASSJMAwAAAACQJMI0AAAAAABJclK9AABAetq0z9Uzi2NaucNVbdzX3mpffTpaGtXD0eeGB9S+gOuxAAAgdxGmAQCfsGqHq9sfr9K76+JHuEdE33vO6D8uyNeXTw0pL2DadH0AAADpgDANAGjwztqYrplWqR0VvvIC0sWDAzp7YEDhoFGHfKO1e1y9uDymNz+O69vPVOu+2RG99fUiqtQAACDnEKYBAJKkmatjuuieCsVcadIJjh69qVCdCz8bkr97Vp7+9mFEv3itVou2urp6WqVe+nKRAjYVagAAkDsoJQAA9PaamC7/WyJIf+uMkF75StFhg7QkGWN0+4Q8vfetdiptZ/TG6rguvLtCnue38aoBAABShzANADluyda4Lv9bpaqi0tSxQf32igLZ1rGrzGXtLb34pSK1zzf658dxPb4g2garBQAASA+EaQDIYeX7PV10T4UO1Pq6dnRQD0wNNylI1xtd5uh/ryiQJH31yWpt2OO21lIBAADSCmEaAHLU/upEkN5ywNdZAxzdPzUsK4kgXe+W8UF9bkRA+2t8TXmgUtE47d4AACD7EaYBIAf5vq/rH6zS0m2uhpXaeubWQoWc4xsgZozR36eE1aejpdkbXf3XazUtvFoAAID0Q5gGgBz09OKYXl4RU6ew0ctfbv7RVh0KLD1yY1jGSL96o1aLtx7pjGoAAIDsQJgGgBxTFfH1nWeqJUm/ujRfZe1b5qVgQt+AvnVGSHFP+uKjVYq7tHsDAIDsRZgGgBzz32/UaPN+T+N62br15FCLPvbPLy5Qn46W5m129fu3a1v0sQEAANIJYRoAcsjHu1zd9c9aGSP9+arjGzh2NOGQ0b3XhiVJP325Rmt3M90bAABkJ8I0AOQI3/f1zaeqFXWl204JaVwvp1W+zqRBAd0yPqiamPQfLzOMDAAAZCfCNADkiNdXxfXKypg6Fhj99yX5rfq1fn5xgYK29Oj8qD7eRXUaAABkH8I0AOSIX81MVIl/fF6eOhe27tN/j/aWbj05JM+Xfvk61WkAAJB9CNMAkAPmbYrrzY/jKs4z+tKEvDb5mj88N0+OJT04L6oNe6hOAwCA7EKYBoAc8Jt/JiZr3zExpKK8lh06diR9Otm68aSg4p70q5lM9gYAANmFMA0AWW7dbldPLooqaEvfPKNtqtL1fnxeviwjTZsT0d4qr02/NgAAQGsiTANAlvvft2rl+dKN40IqLW7bp/0BXWxdODigSFx6aF60Tb82AABAayJMA0AW213p6R+zI5Kk753VtlXperefEpIk3fthRL7vp2QNAAAALY0wDQBZ7M/vRVQTky4bFtDgEjsla7hkaEAl7YyWbnM1eyODyAAAQHYgTANAlqqO+vrTu4nBX/9yTmqq0pIUsI2+ML6uOv0Bg8gAAEB2IEwDQJa6e1atdlf5mtDH0cR+gZSu5Yt1rd6PLYjqYC2t3gAAIPMRpgEgC1XU+vrlG4kq8J0X5qd4NVL/zrbOPcFRdVR69KNIqpcDAADQbIRpAMhCf/swot1Vvk7v5+i8QU6qlyPpk4PIAAAAMh1hGgCyTNz19Ye3E1XpH03KkzEmxStKuGJEUJ3CRh9tdrWgPJ7q5QAAADQLYRoAssxTi6LauM/TkBJbF56Y2r3SjYUcoxtPCkqSHp3PmdMAACCzEaYBIIv4vq/fvpWoSn/3rDxZVnpUpetdPSoRpp9aFOXMaQAAkNEI0wCQRd5bF9fcTa66FhpNHRtM9XI+45TejkrbGa3b42nRFs6cBgAAmYswDQBZ5K43E1Xpr5+ep7xAelWlJcmyjD43IhHyn15MqzcAAMhchGkAyBJrdrl6fllM+QHpjomhVC/niK6sC9NPLY6leCUAAADHjzANAFni7x9G5PvS9WND6lyYvk/vZ/R31ClstHy7q5U7aPUGAACZKX3fbQEAmizu+rp/buL85ttOSd+qtCQ5ttEVwxPV6ScW0OoNAAAyE2EaALLAyyti2nbQ15ASWyf3tlO9nGOaMrr+iKwIU70BAEBGIkwDQBb4+4eJqvQXTw7KmPQbPPZpZw901K3IaOVOpnoDAIDMRJgGgAy3eV9i8FjQlm4cl94t3vVsy+iaUUz1BgAAmYswDQAZ7q8fROT50jWjg+qSxoPHPu2yYYkw/fwypnoDAIDMkznvugAAnxFzff2trsU7nY/DOpwz+jtql2e0cIurzfto9QYAAJmFMA0AGeyVFTFtP+hreKmtCX2cVC8nKUHH6IITA5KkF6hOAwCADEOYBoAM9shHif3GN43LjMFjn3bZsESYnrGUMA0AADILYRoAMlRlxNeMZVEZI00Zk1kt3vUuGhyQZaR/fhxTZYQjsgAAQOYgTANAhnp0fkTVUenM/o7K2mfm03mnsKWJfR1FXemNVVSnAQBA5sjMd18AAN09KzF47MunZmZVut7FQxKt3q8RpgEAQAYhTANABlpYHtf8cledwkafGxFM9XKa5bxBhGkAAJB5CNMAkIH+MTtRlb7ppKBCTuYNHmtsdA9bncJGa3d7WrebI7IAAEBmIEwDQIZxPV+PL0hM8b7l5Mxu8ZYkyzKadEKiOv061WkAAJAhCNMAkGHeXhPXzkpfg7tZGl5qp3o5LeL8ulbv11cTpgEAQGYgTANAhnliYaIqfc2ozDxb+nDOG+RIkmaujivuckQWAABIf4RpAMggcdfXU4vqwvTozG/xrtezg60Tu1raX+Nr3mb2TQMAgPRHmAaADPLG6ph2V/kaWmJrSEl2tHjXO//ERKv3KyuiKV4JAADAsRGmASCDPDQvETRvOCmzj8M6nIsHJ8L0SyvYNw0AANIfYRoAMkRlxNczSxJh+vqx2RemzxwQUH5AmrfZ1c4KL9XLAQAAOCrCNABkiGeXRFUdlc7s76hXh+xq8ZakvIDR2QMD8n3pNY7IAgAAaY4wDQAZ4uG6Fu+pWdjiXa/+iKyZHJEFAADSHGEaADLArkpPr6+OKWBLV43M3jA96YTEEVlvrI7L9zkiCwAApC/CNABkgOkLo3I96aLBAXUoyN6n7iEltkraGZXv97R6J/umAQBA+sred2QAkEXqp3hfNyZ7q9KSZIzRpBMSrd5v0OoNAADSGGEaANLcml2uPtgQV1FIunxYdodpSTqPMA0AADIAYRoA0tzDHyWq0leNDCo/aFK8mtZ3bl2YfvPjuOIu+6YBAEB6IkwDQJqbvrB+incoxStpGz3aWxrczdKBWl8fbXZTvRwAAIDDIkwDQBpbvdPVsu2uOhYYndnfSfVy2gz7pgEAQLojTANAGntmcaIqffmwgBw7+1u8600aRJgGAADpjTANAGns6bow/fksPlv6cM7s78gY6YMNcUXi7JsGAADphzANAGlq0z5Xcza5KgwdanvOFcX5lkb1sBWJS3M3xVO9HAAAgM8gTANAmnq0bor3ZcOCygvkTot3vTP6JfaIv7OWMA0AANIPYRoA0lT9kVhTx+ZWi3e9M/onqvFvr2HfNAAASD+EaQBIQyt3uFqyzVWnsNF5g3Krxbve6XXTy2et57xpAACQfgjTAJCGnl2SqEpfMTyoQA5N8W6sS2HivOmqqDS/nPOmAQBAeiFMA0AaOhSmc7MqXe/0fonv//317JsGAADphTANAGlm6wFPsze6KghK5w7M7TA9sW99qzf7pgEAQHohTANAmpmxNFGVvvDEgPKDudniXe/UujD9/oa4fJ990wAAIH0QpgEgzTy3JFGFvWJ4bk7xbqx/Z0tdC422HvC1ca+X6uUAAAA0IEwDQBo5WOtr5scx2ZZ0yZDcbvGWJGPMJ6rTAAAA6YIwDQBp5OXlUcVc6cz+jjqGeYqWGrV6M4QMAACkEd6pAUAaebauxfvyYbR41zs0hIwwDQAA0gdhGgDSRDTu66UVdWE6x4/EamxMmaOgLS3e6qqiliFkAAAgPRCmASBNvLUmroO1vkaX2erd0U71ctJGXsDopF6OPF+avZHqNAAASA+EaQBIE88uSRyJdQUt3p9xah/2TQMAgPRCmAaANOB5vp6rO1/6ihG0eH8aE70BAEC6IUwDQBqYt9nV1gO++nS0NLyUFu9Pqw/TH2yIy/XYNw0AAFKPMA0AaaChKj08IGNMileTfroVWerf2dLBWl/LtrupXg4AAABhGgDSQf2RWFcMZ7/0kUyo2zc9dxNhGgAApB5hGgBSbPVOV8u3u+oUNg1nKuOzTuqZ+NnM28S+aQAAkHqEaQBIsfoW78lDA3JsWryP5KSeib3k8zYTpgEAQOoRpgEgxWjxbppRPRxZRlq01VUkzhAyAACQWoRpAEihHRWePtgQV35AOm8QR2IdTThkNKTEVsyVlm5j3zQAAEgtwjQApNDzS6Pyfen8QQEVBGnxPpb6Vu+57JsGAAApRpgGgBSixTs543rVDSHbTGUaAACkFmEaAFKkotbXG6tjsox06VBavJuCid4AACBdEKYBIEVeXRlTJC6d3s9R50KejptiRHdbjiUt3e6qJsoQMgAAkDq8ewOAFJlRdyTWZcOoSjdVXsBoeHdbrpeY6g0AAJAqhGkASAHP8/XyisR+6UuHsl86GQ2t3pw3DQAAUogwDQApMHeTq91VvgZ0tnRCVzvVy8ko9RO92TcNAABSiTANACnw4vJEi/clQ2jxTtahyjRt3gAAIHUI0wCQAi8sT7R4X0KLd9KGldoK2tKKHa4qIwwhAwAAqUGYBoA2tmW/pwXlrgpD0pn9nVQvJ+MEHaORPWx5vrSgnFZvAACQGoRpAGhj9S3eF5wYUNAxKV5NZqLVGwAApBphGgDa2PPL6qZ4D6HF+3gx0RsAAKQaYRoA2lB11Ncbq2MyRrqI4WPHbVwvJnoDAIDUIkwDQBt68+OYamPS+F62uhXxFHy8BnezlR+QVu/ydKDGS/VyAABADuKdHAC0oRfqW7yZ4t0sjm00uizR6j2/nH3TAACg7RGmAaCN+L7fcCTW5KG0eDfXST0Trd5zafUGAAApQJgGgDayeKur8v2eytpbGtHdTvVyMh4TvQEAQCoRpgGgjTS0eA8JyBiOxGouJnoDAIBUIkwDQBt5flnifOlLafFuESd0tVQYktbv8bSniiFkAACgbRGmAaAN7KjwNGeTq/yAdM5AwnRLsC2jMXVDyD6i1RsAALQxwjQAtIGXl8fk+9K5JwSUH6TFu6XQ6g0AAFKFMA0AbeCF5YkWb6Z4tywmegMAgFQhTANAK4vGfb26MjF87JIhnC/dkurbvBduoc0bAAC0LcI0ALSyd9bGVRmRRpfZ6tGep92WNKCLpYKgtGGvp/3VDCEDAABth3d1ANDKXqif4j2EFu+WZltGI7snqtOLtlKdBgAAbYcwDQCtyPd9PV9/vvRQWrxbw6geiX3TtHoDAIC2RJgGgFa0coendXs8dSsyDcOy0LIOhWmGkAEAgLZDmAaAVlQ/xfuSIQFZFkditYZRPRhCBgAA2h5hGgBa0Qu0eLe6YaW2LCMt2+4qGvdTvRwAAJAjCNMA0Er2VnmatT6uoC1NOoHhY62lIGg0qKulmCut2EF1GgAAtA3CNAC0kldXxuR60lkDHBXl0eLdmmj1BgAAbY0wDQCt5IXltHi3FYaQAQCAtkaYBoBWEHd9vbwiEaYv4XzpVldfmV5QTmUaAAC0DcI0ALSC9zfEta/a15ASW/06cyRWaxvZ6Kxp32cIGQAAaH2EaQBoBfVTvCcPpSrdFroVWSptZ3Sg1tfGvV6qlwMAAHIAYRoAWsELyxLnS19KmG4zo8sYQgYAANoOYRoAWtja3a5W7PDUscDolN5OqpeTMxhCBgAA2hJhGgBa2It1Ld4XDQ7IsTkSq61wPBYAAGhLhGkAaGEvLKfFOxVGNRpCBgAA0NoI0wDQgipqfb21Ji7bki44kTDdlvp3shQOShv3edpXzRAyAADQugjTANCCXlsVU8yVTu/nqEMBT7FtybKMRta1ei+iOg0AAFoZ7/QAoAUxxTu1aPUGAABthTANAC3E9Xy9uDwxfOySIcEUryY3MdEbAAC0FcI0ALSQDzfEtavS18AulgZ15ek1Feonei+gMg0AAFoZ7/YAoIU8tzRRlb58WFDGcCRWKgwrsWVb0vLtriJxP9XLAQAAWYwwDQAtwPd9PbsksV/68uHsl06V/KDRiV1txb1EoAYAAGgthGkAaAELyl19vMtTSTujCX2cVC8npzGEDAAAtAXCNAC0gCcXJarS144KyrZo8U6l+jC9oJwhZAAAoPUQpgGgBcyo2y/9uRFM8U61+rOmF2+lMg0AAFoPYRoAmmndblfLtrvqUGA0sS8t3qk2onuiMr14myvfZwgZAABoHYRpAGim55clqtIXDw7IsWnxTrVuRZa6Fhrtq/a15QBhGgAAtA7CNAA004ylif3Sk4cyxTtdNFSnt7JvGgAAtA7CNAA0w75qT++sjcuxpAsHE6bTxYjuiXb7JeybBgAArYQwDQDN8NLymOKedM5AR8X5PKWmi8b7pgEAAFoD7/wAoBmeXZJo8b58OFO808mhNm/CNAAAaB2EaQA4TrUxXy+vSAwfu2wYYTqdDO5my7aklTtcReIMIQMAAC2PMA0Ax2nm6piqotK4XrbK2vN0mk7yAkaDutqKe4lADQAA0NJ49wcAx+nZJYmq9BW0eKelEaW0egMAgNZDmAaA4+B6fsORWITp9MS+aQAA0JoI0wBwHGZvjGtnpa+BXSwN7sZTaTrirGkAANCaeAcIAMehvsX78mFBGWNSvBocDsdjAQCA1kSYBoAk+b6vZxbXt3gHUrwaHElZe0vt8422H/S1s8JL9XIAAECWIUwDQJJW7PC0ZrenroVGp/RxUr0cHIExpqE6vYTqNAAAaGGEaQBI0nNLElXpy4YFZVu0eKez4Uz0BgAArYQwDQBJenYJLd6ZoqEyzRAyAADQwgjTAJCELfs9zdnkKhyUzj2BMJ3uRnRPtOEzhAwAALQ0wjQAJKH+bOkLBweUF6DFO90Nq2vzXrbdVdz1U7waAACQTQjTAJCEQy3ewRSvBE1RGDLq39lSbUxas5uJ3gAAoOUQpgGgiQ7UeHpzTVy2JV0yhBbvTFE/hGwprd4AAKAFEaYBoIleXhFTzJXOGuCoQwFPn5mivtV7yTaGkAEAgJbDu0EAaKJnl8Qk0eKdaYaVUJkGAAAtjzANAE0Qift6aXliv/Tlw2jxziTD6yZ6LyFMAwCAFkSYBoAmePPjmCoi0pgyWz072KleDpIwsIulgJ0YQFYTZaI3AABoGYRpAGgCWrwzV8A2GtzNlu9LK3ZQnQYAAC2DMA0Ax+B5fsP50lcMp8U7Ex0aQkaYBgAALYMwDQDHMHeTq20HffXrZDWEMmQWhpABAICWRpgGgGN4dkl9VTooY0yKV4PjMbw7lWkAANCyCNMAcAyHwjQt3pnqUGWas6YBAEDLIEwDwFGs3OFq5U5PncNGp/Z1Ur0cHKfeHS0VhqQtB3ztq/ZSvRwAAJAFCNMAcBTP1VWlJw8NyLZo8c5Uxhj2TQMAgBZFmAaAo2ho8R7BkViZblhporOAMA0AAFoCYRoAjqB8v6cPN7oKB6XzTmC/dKYbzvFYAACgBRGmAeAInlmcqEpfPCSo/CAt3pmu/lgzKtMAAKAlEKYB4AieXJQI01eOoCqdDeqPx1q63ZXv+yleDQAAyHSEaQA4jE37XL2zNq6CoHTJUPZLZ4MuhZa6Fhrtq/a19QBhGgAANA9hGgAO4+F5iar054YHVRiixTtbNFSnOW8aAAA0E2EaAD7F9309OC8iSbrxJKrS2aT+eCyGkAEAgOYiTAPAp8wvd7Vih6eSdkbnMsU7qzCEDAAAtBTCNAB8ykN1Venrx4Tk2LR4Z5Ph9WdNbydMAwCA5iFMA0AjcdfXo/MT+6Vp8c4+Q+ravJdtd+V6DCEDAADHjzANAI3MXB3TjgpfQ0tsjexhp3o5aGFFeUZ9O1mqjUnr9nipXg4AAMhghGkAaOTJRTFJ0nVjgjKGFu9s1DCEbCut3gAA4PgRpgGgTtz19eySRIv3lSNp8c5WDCEDAAAtgTANAHXeXRfX7ipfQ0psndiNFu9sNbw+TDOEDAAANANhGgDqPLUoUZX+/AiOw8pm9ZXpJVvjKV4JAADIZIRpAJDkeb6eXlzX4j2CFu9sNqirLceSPt7tqTbGRG8AAHB8CNMAIOnDjXFtO+irXyeLKd5ZLugYDepqy/WkVTtp9QYAAMeHMA0Akp5ceGjwGFO8s19DqzdDyAAAwHEiTAPIeb7v6+nFiSOxaPHODcOZ6A0AAJqJMA0g583b7GrjPk+9Olga35sW71xAZRoAADQXYRpAzpte1+J9FS3eOYPKNAAAaC7CNICc5vu+nlxUH6Y5EitX9OloqSAobdrn6WAtE70BAEDyCNMActr8clfr93gqa2/p5N5OqpeDNmJZRkNL6qvTnDcNAACSR5gGkNPqW7yvHBGQZdHinUto9QYAAM1BmAaQsxq3eF89iineuWZYaaITgSFkAADgeBCmAeSshVtcrd3tqXux0YQ+tHjnGirTAACgOQjTAHLWoRbvIC3eOajx8Vi+zxAyAACQHMI0gJzk+76eWEiLdy7rVmTUKWy0p8rXjgrCNAAASA5hGkBOmrsp0eJd1t7SxL60eOciYwyt3gAA4LgRpgHkpEfnRyRJ146mxTuXNW71BgAASAZhGkDO8Ty/Yb/0lNG0eOeyQ5VpzpoGAADJIUwDyDmzN8a15YCvvp0sje1pp3o5SKFhtHkDAIDjRJgGkHOeXBSTJF01MihjaPHOZUNLEmF62XZXnscQMgAA0HSEaQA5xfd9Pbko0eJ91UhavHNdcb6lXh0sVUWlDXu9VC8HAABkEMI0gJwyb7OrTfs89epgaVwvWrzBEDIAAHB8CNMAcsqTCw9VpWnxhiSOxwIAAMeFMA0gZ3yyxTuQ4tUgXTCEDAAAHA/CNICcsaDc1bo9nnoUG53c20n1cpAmhpXQ5g0AAJJHmAaQM+qr0leODMqyaPFGwondbNmWtGqnq2icid4AAKBpCNMAcoLv+3piIVO88Vl5AaOBnS3FPWn1LqrTAACgaQjTAHLCR5tdrd2daPGe2JcWb3wS+6YBAECyCNMAcsJjCxJV6WtHh2jxxmcMK01cYCFMAwCApiJMA8h6nufr8bowPWUMLd74LCrTAAAgWYRpAFnv/Q1xle/31K+TpZN62qleDtJQ/UTvpdsJ0wAAoGkI0wCy3uMNLd5BGUOLNz6rf2dLIUdat8dTVYSJ3gAA4NgI0wCymuv5mr7wUJgGDsexjQZ3s+X70oodVKcBAMCxEaYBZLW318S1o8LXoK6WRnSnxRtHxr5pAACQDMI0gKxGizeain3TAAAgGYRpAFmrNnZoivd1Y0IpXg3SHZVpAACQDMI0gKz13JKoDtT6Orm3rRO70eKNozsUpuMpXgkAAMgEhGkAWev+uYmq9M3jqErj2Hp1sFQYkrYc8LWv2kv1cgAAQJojTAPIStsOeHp1ZUxBmyneaBpjTMO+6WXsmwYAAMdAmAaQlR7+KCLPly4bFlDHME91aJphpY4k9k0DAIBj4x0mgKzj+/6hFu/xtHij6RhCBgAAmoowDSDrLCh3tXSbq66FRhecGEj1cpBBCNMAAKCpCNMAss79cyOSpKljgwrYnC2NpmsI09td+b6f4tUAAIB0RpgGkFWicV+PfESLN45P10KjLoVGe6p8bT1AmAYAAEdGmAaQVV5eEdPuKl8ju9sa2cNJ9XKQYYwxGtk9UZ1etJXzpgEAwJERpgFklfoWb6rSOF6j6i7CLNzCvmkAAHBkhGkAWWN3pacXlsVkW9L1YzlbGsdnVI9EZZowDQAAjoYwDSBrPDQvqpgrXTw4oG5FPL3h+BwK07R5AwCAI+PdJoCs4Pu+/vZhosX7tgm0eOP4DepqK+RIa3Z7qowwhAwAABweYRpAVpi90dWy7a5K2xldPJizpXH8HNtoWKkt35eWMIQMAAAcAWEaQFb4e11V+pbxITmcLY1mYggZAAA4FsI0gIxXUevr0fmJMH3rybR4o/kYQgYAAI6FMA0g4z2xMKqqqHTWAEcDutipXg6yAEPIAADAsRCmAWS8+hbv206hKo2WMaJ7os178TZXcZchZAAA4LMI0wAy2rJtcX2wIa7iPKPPj+BsabSMdnlG/Ttbqo1JH+/yUr0cAACQhgjTADJafVX6hpOCyg8yeAwtZ2T3RKv3IiZ6AwCAwyBMA8hYkbivB+dFJdHijZbHRG8AAHA0hGkAGWvG0ph2V/kaU2ZrVJmT6uUgyzDRGwAAHA1hGkDG+tsHtZKoSqN1MNEbAAAcDWEaQEbauNfV66vjyg9I141h8BhaXll7Sx0LjHZU+Np+kCFkAADgkwjTADLSfbMj8n3pqpFBtS/gqQwtzxhDdRoAABwR70ABZBzX8/WP2XWDxybQ4o3W0zDRm33TAADgUwjTADLOG6ti2rzf08Aulk7vx+AxtB4megMAgCMhTAPIOH+rO1v6iyeHZAxnS6P1jCqjzRsAABweYRpARtlV6em5pTHZlnTzeFq80bpO7GoraEurdnmqivipXg4AAEgjhGkAGeXBuRHFXGny0IBK2vEUhtYVdIyGltryfWnpNlq9AQDAIbwTBZAxfN8/1OLN2dJoI6O60+oNAAA+izANIGN8sCGuFTs8dS82uvDEQKqXgxwxsm4I2aKtVKYBAMAhhGkAGePvdVXpL4wPybEZPIa2ceisacI0AAA4hDANICMcrPX12ILE2dK3nkyLN9pO/VnTi7fG5XoMIQMAAAmEaQAZ4fEFEVVHpXMGOurX2U71cpBD2hdY6tPRUlVUWrvbS/VyAABAmiBMA8gI9S3etzF4DClwqNWbIWQAACCBMA0g7S3ZGtfsja46FBh9bkQw1ctBDhrZnX3TAADgkwjTANLeXz9IVKVvGBtUXoDBY2h7o5joDQAAPoUwDSCt7a/2dN/sRJj+ysS8FK8GuYo2bwAA8GmEaQBpbdqcqKqi0vmDHA0pYfAYUqN3R0vFeUZbD/jaWcEQMgAAQJgGkMZ839ff6gaPfe10qtJIHWNMQ3WaVm8AACARpgGksXmbXS3b7qqkndHFgwOpXg5yHK3eAACgMcI0gLR1b93gsRtPCsmxGTyG1BpZN4SMid4AAEAiTANIU/uqPT00LyJjpC9N4GxppF5DmzdhGgAAiDANIE39Y3ZENTHposEBDejC4DGk3pASW44lrdzpqibqp3o5AAAgxQjTANKO6/n6v/cSLd5fP42qNNJDyDEaUmLL9aRl26lOAwCQ6wjTANLOyytiWrfH04DOli44kcFjSB8MIQMAAPUI0wDSzp/erTsO67Q8WRaDx5A+RjGEDAAA1CFMA0grq3e6enVlTOGgdMv4YKqXA3zCyO71lWnCNAAAuY4wDSCt/Pm9WkmJ47DaF/AUhfQysn6i99a4PI8hZAAA5DLeqQJIGxW1vqbNiUqSvn46g8eQfjqFLfXpaKkyIq3e5aV6OQAAIIUI0wDSxoPzIjpY6+vsgY6GljqpXg5wWCf1TFSn521mCBkAALmMMA0gLfi+rz+9m2jx/vppeSleDXBk43olLvTM3USYBgAglxGmAaSFf34c14odnnq2t3TZMI7DQvo6qWciTM/bzBAyAAByGWEaQFqor0rfMTEkx+Y4LKSvMWWJNu8F5XHFXYaQAQCQqwjTAFJu415XM5bGFHKk2yYweAzprX2BpYFdLNXEpOU7qE4DAJCrCNMAUu4vsyLyfGnK6KC6FPK0hPRXv2963ibCNAAAuYp3rQBSqibq628fRiRJXz+dwWPIDIf2TTOEDACAXEWYBpBSjy2Iak+Vr1N62zqpF8dhITNwPBYAACBMA0gZ3/f1x/rjsKhKI4OMLnNkGWnRFlfROEPIAADIRYRpACnzwYa4FpS76lZkdPWoYKqXAzRZYchocDdbUVdaso190wAA5CLCNICU8H1fP36hRpL0lVNDCjoch4XM0tDqvYlWbwAAchFhGkBKzFga09tr4+paaPTds/NTvRwgafV7/OdtpjINAEAuIkwDSIlfvZGoSt95Yb7a5VGVRuYZVzfRey6VaQAAchJhGkCbe399TB9udNWl0OiW8aFULwc4LiO623Isael2VzVRhpABAJBrCNMA2txdbyYmeH91Ykj5QarSyEz5QaNhpbZcT1q0lVZvAAByDWEaQJtas8vVs0tiCjnSV0/jOCxktpNo9QYAIGcRpgG0qd/8s1a+L900LqSuRTwFIbON61U30XszYRoAgFzDO1kAbWbzPlf3zYnIMtK/nENVGpmvvjLNRG8AAHIPYRpAm/nDOxHFXOn6sUEN6GKnejlAsw0rtRW0pRU7XFVGGEIGAEAuIUwDaBMVtb7u/SAiSfr+2VSlkR2CjtHIHrZ8X5pfTqs3AAC5hDANoE38/cOIDtb6Onugo5E9nFQvB2gx43ol/j3P2UiYBgAglxCmAbS6SNzXb96skST9gKo0ssyEPokwPWs9YRoAgFxCmAbQ6u6fE9HWA77GlNm6cHAg1csBWtTEvofCtO+zbxoAgFxBmAbQquKur1/PrJUk/fi8fBljUrwioGX16WiptJ3Rrkpfa3Z7qV4OAABoI4RpAK3qsQVRrdvjaXA3S58bTlUa2ccYc6g6vY5WbwAAcgVhGkCr8Txfv3wjUZX+10n5siyq0shOE/smLhSxbxoAgNxBmAbQap5dEtPy7a76drJ03ZhgqpcDtJqJ/er3TcdSvBIAANBWCNMAWoXv+/rF64kJ3j88J0+OTVUa2WtUD1sFQWnFDk97q9g3DQBALiBMA2gVr66MaX65q+7FRrecHEr1coBWFbCNTu6dqE6/v4FWbwAAcgFhGkCr+MXrib3S3z87XyGHqjSyH0PIAADILYRpAC3unbUxvbcurs5hoy9NoCqN3ND4vGkAAJD9CNMAWtwvXkvslf72mXkKh6hKIzdM6OPIGGnu5riicT/VywEAAK2MMA2gRc3dFNdrq+Jql2f0tdOoSiN3FOdbGlZiqzYmzS93U70cAADQygjTAFrUna8kqtJfPz2k9gU8xSC31Ld6v7eOI7IAAMh2vNMF0GJmb4jrpeUxFYWk752Vl+rlAG3u0HnT7JsGACDbEaYBtJj/eOXQXumOYZ5ekHsaDyHzffZNAwCQzXi3C6BFzFoX06srYyrOM/oOVWnkqD4dLZW2M9pV6WvNbi/VywEAAK2IMA2g2Xzf149fTFSlv3tWnjqwVxo5yhjDedMAAOQI3vECaLbXV8X1ztrEudJUpZHrJvYNSGLfNAAA2Y4wDaBZElXpaknSv07KU1Ee50ojtx0aQsZEbwAAshlhGkCzPL04po82uyprb+mrp1GVBkb1sFUQlFbs8LS3in3TAABkK8I0gOPmer7+ra4q/dML8pQXoCoNBGyj8b0S1en3N9DqDQBAtiJMAzhu98+JauVOTwM6W7plfCjVywHSBkPIAADIfoRpAMelKuLr315KVKV/cUm+AjZVaaBe4/OmAQBAdiJMAzgud71Zq20HfU3o4+jqUcFULwdIKxP6ODJGmrs5rmjcT/VyAABAKyBMA0ja1gOe/uefiXOl77o8X8ZQlQYaa19gaWiJrdqYNL/cTfVyAABAKyBMA0jav79UreqodM2ooE6tO1MXwCed1pcjsgAAyGaEaQBJWVge131zogra0q8m56d6OUDaOnTeNPumAQDIRoRpAE3m+76+91y1fF/61pl56tvJTvWSgLRVX5l+Z21cnse+aQAAsg1hGkCTvbAspn9+HFfnsNGPJ+WlejlAWuvTyVbfTpb2VPlauIV90wAAZBvCNIAmibm+fjAjcRTWnRfmq30BTx/AsZx3QmKmwOur2DcNAEC24d0wgCa5b3ZEq3Z6OrGrpS+dGkr1coCMcN6gujC9mjANAEC2IUwDOKZI3NfPX6uVJP3ikgIFbI7CAprinIGJ86bfWxdXTZR90wAAZBPCNIBj+t1btdq839OoHrauGM5RWEBTdQxbOqmnrUhcencdU70BAMgmhGkAR7V5n6v/eq1GkvS7KwpkWVSlgWQ0tHqzbxoAgKxCmAZwVN97rkbVUWnK6KDOGkhVGkgWQ8gAAMhOhGkAR/TGqpimL4yqMCTddXlBqpcDZKQJfR0VBKVFW13tqPBSvRwAANBCCNMADqs25uvrT1VJkv7jgnz1aM/TBXA8Qo7Rmf0T1emZTPUGACBr8O4YwGH94vUardrpaViprW+ekZfq5QAZjX3TAABkH8I0gM94d21M//16rYyR/npNgYIOQ8eA5jh/kCNJenVlTJ7HEVkAAGQDwjSAT9hX7Wnqg1XyfOnfzsvThL4MHQOaa0iJrT4dLW076Gt+uZvq5QAAgBZAmAbwCd9+plqb93ua2NfRTy/IT/VygKxgjNFlwxIXpmYsjaZ4NQAAoCUQpgE0mLE0qgfmRlUQlO6fGpZj094NtJTJQ4OSpBlL2TcNAEA2IEwDkCTtqPB022OJ6d2/nlyg/p3tFK8IyC5n9HfULs9o0VZXG/fS6g0AQKYjTAOQ7/u69dEq7ar0dd4gR1+dGEr1koCsE3SMLhqcaPV+fhnVaQAAMh1hGoD+Miuil5bH1ClsNO36QlkW7d1Aa6jfN/08+6YBAMh4hGkgx63Y7up7z1VLku69NqzuxTwtAK3losEB2Zb05pq4DtZyRBYAAJmMd81ADovGfU19qFK1Mem2U0L63IhgqpcEZLUOBZbO6O8o5ibOnAYAAJmLMA3ksJ+8WKMF5a4GdLb0u88VpHo5QE64rGGqN63eAABkMsI0kKNeWh7VXW/WyrGkh28sVGGIfdJAW5hct2/6xeUxxV1avQEAyFSEaSAHle/3dNPDiWOwfjU5X+N7OyleEZA7+ne2NaTE1r5qX+9viKd6OQAA4DgRpoEcE3d9Xf9ApfZU+bp0aEDfPSsv1UsCcs5lQxPV6RlL2TcNAECmIkwDOebfXqrRu+viKmtvadr1YRlDezfQ1uqPyHpuSVS+T6s3AACZiDAN5JAH50b065mJfdKP3hRWpzBPAUAqjO/tqGuh0Zrdnlbu8FK9HAAAcBx4Jw3kiPfXx3TbY4l90n+5ukCn9QukeEVA7rIto8nDElO9py9kqjcAAJmIMA3kgE37XF3x90pFXek7Z4Z02wT2SQOpdt2YRJh+ZH6EVm8AADIQYRrIcjVRX5//R6V2Vfq6aHBAv7mc86SBdHDWAEcl7YxW7fS0cIub6uUAAIAkEaaBLOb7vr4yvUofbXY1oLOlR24My7YYOAakA9syumZUojr96HxavQEAyDSEaSBL+b6v7z5brQfmRlUYkp79YqHaF/ArD6ST68eEJEmPfBSR69HqDQBAJuGdNZCFfN/XD5+v0e/fjijkSE/eUqihpU6qlwXgU8b3tjWoq6UtB3y9sYozpwEAyCSEaSDL+L6vn7xYo9/8s1ZBW3r61kJdMDiY6mUBOAxjjG4Zn6hOT5tDqzcAAJmEMA1kmTtfqdEv36hVwJae/EKhLh5CkAbS2Y0nhWQZ6ZklUe2v5sxpAAAyBWEayCL/9WqN/vPVWtmW9PjNhQ3n2AJIXz3aWzpvUECRuPTYAqrTAABkCsI0kCV+9UaNfvpyjWxLevSmQn1uBEEayBRfGJ/4ff37h5EUrwQAADQVYRrIAnf9s0b/+kKNLCM9dENYV48iSAOZ5PLhQXUKG83b7GrepniqlwMAAJqAMA1kuN+/VasfzKiRMdL9U8OaUnfUDoDMkRcwuvXkxO/uX2bVpng1AACgKQjTQAb707u1+s6z1TJG+seUsG44iSANZKovn5r4/X10flT7GEQGAEDaI0wDGcj3ff3bi9X6xlPVkqS/XlOgW04mSAOZrH9nWxecGFBNTHpgLoPIAABId4RpIMNE4r5ueKhKv3i9Vo4l3XddWLdNyEv1sgC0gDsmJi6K3T2rVr7vp3g1AADgaAjTQAbZU+XpvP+r0CMfRdUuz+jlLxdRkQayyCVDAiprb2nlTk+vr2IQGQAA6YwwDWSItbtdnfr7g3p3XVy9Olia9a0iTRoUSPWyALQgxzb65hmJC2S/fKMmxasBAABHQ5gGMsCHG+I65XcHtXqXpzFltj78djsNK3VSvSwAreArp+apQ4HRW2vien99LNXLAQAAR0CYBtLcU4uiOvvPB7W7ytclQwJ6+xvtVFrMry6QrYryjL55eqI6/d+vc0wWAADpinfkQJryPF//+UqNrp5WqdqY9LXTQnr2i4UqDJlULw1AK/vGGXkKB6UXl8e0sJy90wAApCPCNJCGdld6uvivlfqPV2pkJP328nz98coCOTZBGsgFncKW7piYmNL/i9fZOw0AQDoiTANpxPd9TV8Y1Yj/OaBXV8bUtdDo9TuK9N2z82UMQRrIJd87O0/5AenJRTF9tJnqNAAA6YYwDaSJDXtcXXpvpa6ZVqltB32dNcDR/O8X65wTmNgN5KKSdpa+dUaiOv3jF6pTvBoAAPBphGkgxWKur9/8s0ZDf31ALy2PqUOB0d+nhDXzq0Xq0Z5fUSCX/cu5eWqfb/Taqrje+pjJ3gAApBPeqQMpNHtDXCf99qD+ZUaNqqPSjScFtfJfi3XrKSFZFm3dQK7rUGDph+cmqtPfe65arueneEUAAKAeYRpIgQM1nr72ZJUm/OGgFm91NaCzpdfvKNIDNxSqaxG/lgAO+dYZeerT0dL8cld/mRVJ9XIAAEAd3rUDbSga93X3rFoN/uUB/d97ETmW9G/n52nxvxRr0iD2RgP4rPyg0R+vLJAk/eTFGm0/6KV4RQAAQJKcVC8AyAXVUV8PzYvoF6/XatO+xBvh0/o5uueasIaU2CleHYB0d+nQoK4YHtCzS2L6/nPVeujGwlQvCQCAnEeYBlrRgvK4/vZhRA/Pi+pAbWKv48jutu68MF+XDw9w3BWAJvv95wr02qoDevijqG49OcakfwAAUowwDbSwg7W+Hvkoor99GNFHm92G/z6xr6Nvn5mnz48IMFwMQNJ6d7T10/Pz9aMXanTro1Va8IN26lDAbi0AAFKFMA20AN/39f76RBX6iYVRVUcT/71z2Ojm8SF98eSQBtPODaCZvnt2np5bGtMHG+L64mNVeuoLhXS4AACQIoRpoBl2VXp6cG6iCr1ix6GhQOcNcnT7hDxdNiygkMMbXQAtI2AbPXpTWKN+c1DPLI7p/70T0bfOzEv1sgAAyEmEaSBJcdfXzNUx/WN2VM8siSpW18ndvdjo1pNDuvXkkPp2ogoNoHX07mjrvuvC+tw/KvW956o1uJut809k/zQAAG2NMA00QW3M19trYnpyUUzPLIlqT1VimJhtSZcPC+j2CSFdcGJAjk0VGkDru2JEUHdemK87X6nR1dMq9cG323EyAAAAbYwwDRzBpn2uXloe00vLY5r5caxhH7QkDSu1df2YoG4ZH1JpMQOAALS9n16Qp1U7XT06P6rz/nJQb3+9nQZ0IVADANBWCNNAnZibGCL20vKYXloR09Jt7iduP6mnrcuGBXX1qKBO7MYbVgCpZYzRP64La1elpzdWx3XOnyv01teL1K8zz08AALQFwjRy2vaDnl5ZkQjPr62MNZwFLUnFeUYXnBjQxUMCunBwQN2KqEADSC95AaPnbivSJX+t0Ftr4pr4/w7qpS8VaXQZL+8AALQ2Xm2RU6qjvuaXx/X6qpheXB77xDnQknRiZ0+XjyzQxUMCmtDHUYA90A3Ky8vVs2dPSdLmzZtVVlaW4hWlH35GTcPP6diS+RkVBI2ev71IV/6jQq+tiuvMPx7U07cWadIghpIBANCaCNPIWq7na8UOV3M2upqzKa7ZG+Nass2Ve+gEKxUEpYk943r9r/8ibXhdr6+YrbKyzqlbNAAch8JQIlDf9liVHpwX1YX3VOi/L8nX98/Ok2VxURAAgNZAmEZW8H1fG/d6mrvZ1dxNcc3dFNe8zXFVRj55P8eSxpTZOr2fo4uHBHVGf0e7d2xRz2/en5qFA0ALCTpG908Nq09HS//1Wq1++HyN3vw4rnunhFXWnm0qAAC0NMI0Mo7r+Vq909OCLXEt3OJqQXlcC7a4DcdVNda/s6XxvRyN7+Xo5N62RvVwlB+kSgMgOxlj9J8XF+jUvo5ufKhKr6yMacgv9+unF+Tr66fnKS/A8x8AAC2FMI205Xm+Nu/39PEuTx/vcrVsu6v55a4WbY1/4piqet2KjMb1cjSup6OTetka38tR50KqMQByz4WDg1ryQ0fffLpa0xdG9YMZNfrTuxH966Q83TQuxEVFAABaAGEaKeX7vrYe8PXxLlcf73b18S5Pq3e6+ni3p7W7XUXih/+8fp0sjS5LVJpH90j83b3YyBjeIAKAJJW0s/TELYV6Y1VMP5hRrYVbXH1lerX+7aUa3XZKSFNPCmpYKW8DAAA4XryKolV5nq9dVb7K93vauNfThr2uNuz1tHGfp/V7PK3d4x62ylyvR7HRwC62BnaxNbibXRecbbUvoOIMAE0xaVBAH32vnZ5ZEtNv36zVBxvi+tXMWv1qZq2Glti6dnRQ5wx0NLanQxs4AABJIEwjKZG4rwM1vg7UfvLvPdWeth3wte2gp60HvcTfBzxtr/A/MT37cEraGQ3sbGtgF6shOA/sYmlAZ1sFtCICQLNZltGVI4O6cmRQH26I68F5EU1fGNWy7a5++nKNfvqyFLATAxpP7ePolD6OBnax1aejpQ5cvAQA4LAI0xmusrKy4eN3Fm1Wh61xxV0p7vmKe6r7kwi0MU9yPSnu+oq7Um3cV01cqon6qq37uyYu1cZ8Vcd8VdT6qoj4Olj3d0XEVySW5AKN1KXQqFuhpbL2lnq2t1VWbFTWwVJZsaVeHWwV5h0mMHvS3p3S3ub9eJpk27Zth/0Yn8TP6dj4GTUNP6dja82fUZkj/esp0g/G+Xp/Q1yvroxp3qa4VuxwNXu/NHvpJ+9fFDIqa2+pW5Gl9gVG7fPr/uQZtcszCjmJSeIhO/F30JZ2b9/Q8PkrV65s0fUDAJCseDyuXbt2SZKGDx+uvLy8Fnlc4/v+Z0cgI2NMmzZNX/jCF1K9DAAAAABIe3PmzNG4ceNa5LHo3QIAAAAAIEm0eWe4U045peHjZ599Vt27d0/hajLTjh07NHnyZEnS888/r27duqV4RemJn9Ox8TNqGn5Ox5bpP6MFCxboy1/+sqREBaC0tDTFKwIA5LJt27Zp/PjxkqQuXbq02OMSpjNcYWFhw8djx45VWVlZCleTmcrLyxs+HjVqFD/DI+DndGz8jJqGn9OxZdPPqLS0NKPXDwDILo7TchGYNm8AAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJLEOdPIeWVlZfJ9P9XLSHv8nI6Nn1HT8HM6Nn5GAACkPyrTAAAAAAAkiTANAAAAAECSCNMAAAAAACSJMA0AAAAAQJII0wAAAAAAJIkwDQAAAABAkgjTAAAAAAAkiTANAAAAAECSCNMAAAAAACSJMA0AAAAAQJII0wAAAAAAJIkwDQAAAABAkgjTAAAAAAAkiTANAAAAAECSCNMAAAAAACSJMA0AAAAAQJII0wAAAAAAJIkwDQAAAABAkgjTAAAAAAAkiTANAAAAAECSCNMAAAAAACSJMA0AAAAAQJII0wAAAAAAJIkwDQAAAABAkgjTAAAAAAAkiTDdSHV1tV5++WX9/Oc/1+c//3n17t1bxhgZY3TnnXe2yNeoqKjQnXfeqeHDh6uwsFDFxcUaN26cfvvb3yoajbbI1wAAAAAAtC4n1QtIJ3PmzNHFF1/cao+/ceNGnXXWWdqwYYMkqaCgQJFIRPPmzdO8efP08MMPa+bMmerQoUOrrQEAAADIREu3xTV9YVQrdnjaesBTYcioKCSN7+3oyhFB9etsp3qJyDGE6U/p0KGDxowZ0/DnO9/5jrZv397sx3VdV5MnT9aGDRtUWlqqBx54QJMmTZLneZo+fbpuv/12LViwQFOnTtVLL73UAt8JAAAAkNlcz9eLy2P6w9u1+ufH8cPe58lFMf3rCzW64MSA/mdyvoaWEnHQNviX1sjpp5+uvXv3fuK//ehHP2qRx542bZqWLFkiSXrqqac0YcIESZJlWbr22mvleZ6uv/56vfzyy5o5c6bOPffcFvm6AAAAQCaavzmuWx6p0pJtriSpKCTdNC6kMWWOTuhqqTLia1elr9dXxfTo/KheWh7TBxviemBqWJcMCcgYk+LvANmOPdON2HbrtYbcf//9kqSzzz67IUg3NmXKFPXt21eS9MADD7TaOgAAAIB05vu+/vv1Gp38u4Nass1V306W/vC5ApX/rIP+dFVYt54S0mn9ArpwcFA3jgvpgRsKtf2/2uuyYQHtq/Y1+d5KXXt/leKun+pvBVmOMN0GqqurNWvWLEnSRRdddNj7GGN04YUXSpJee+21NlsbAAAAkC5W7nB1/l8q9JMXa+T60nfPytOyHxbrm2fmqV3ekSvNncKWpt9SqN9dUaD2+UbTF0b1rWeq5fsEarQe2rzbwIoVK+R5niRp2LBhR7xf/W3bt2/X3r171bFjxzZZHwAAAJBq766N6dJ7K3Ww1le7PKMnbg7rgsHBJn9+0DH69ll5mtDH0Vl/Pqj/ey+iroWW/v38PFkWLd9oeYTpNrB169aGj3v06HHE+zW+bevWrUmH6W3bth3zPmVlZUk9JgAAn1ZeXn7U23fs2NFGKwGQLV5aHtWV91WqNiZdOzqoP15ZoC6Fx9dEe3IfRw9OLdTV0yp15ys1soz07xfkt/CKAcJ0m6ioqGj4uKCg4Ij3a3xb489pqvHjxx/zPrS6AACaq2fPnqleAoAsMn1hVNc/UKm4J331tJD++PmCZleSrxoV1KM3hXX9g1X6z1drdPGQgMb2JPqgZbFnGgAAAEBKLCyP66aHE0H6J+fl6U9XNj9I15syJqTvnpWnuCfd+FClaqIUldCyuDzTBoqKiho+rq6uPuL9Gt/W+HOaas6cOSotLU368wAASMbmzZuPevvChQs1efLkNloNgEy1eZ+ri/5aodqYdOvJQf38kiN3cB6vn1+cr1dWxLRsu6ufvFit//1cuMW/BnIXYboNdO/eveHjLVu2aMSIEYe935YtWw77OU1VWlrKnmgAQKs71mtNU2Z4AMht1VFfZ/+5QtsP+prY19Ffrm6dkJsXMHrwhrBO/t1B/e7tiC4bFtRZAwOt8rWQe2jzbgODBw+WZSV+1EuXLj3i/epvKykpYZI3AAAAstZ/vVqjtbs9DSmx9dSthQo6rTdte3SZo/+oG0B262NVqozQ7o2WQZhuAwUFBZo4caIk6ZVXXjnsfXzf16uvvipJOv/889tsbQAAAEBb+mhzXHe9WSvLSPdfH1a3otaPJD88N09je9pav8fT95498rZLIBmE6TZy8803S5LefPNNzZ49+zO3T58+XevWrZMk3XTTTW26NgAAAKAt1ER9XT0tMXDsm2eEdFKvttl16thGD0wtVF5A+usHET27ONomXxfZjTD9Kfv27dPu3bsb/nieJykxHKzxf6+srPzE502bNk3GGBlj9NZbb33mcW+++WYNHz5cvu/ryiuv1MyZMyVJnudp+vTpuv322yVJF110kc4999zW/SaRVSojvuZvjuvFZVHdNzuia6dV6vN/r9AVf6vQlf+o0NQHKvXEgoj+uTqmFdtdHayltQkAAKTGN5+u1vo9nkaX2fqfy1p+4NjRDCmx9dvLE1/za09VKRLnPRGahwFknzJ69Ght3LjxM//9N7/5jX7zm980/O+bb75Z06ZNa/LjOo6jGTNm6Oyzz9aGDRs0adIkFRQUyPM81dbWNnzthx9+uNnfA1LnvXUx/eHtiKJxX/lBoxtOCmpYia3eHS0Z07J7gaoivl5dGdO0ORHVHx8esKX8gFFB0MgyRnHXV01cenBeVGr0elGUZ3TzuKBO6GqrdwerxY6gAAAAOJJ7P6jV3z6MqCAoTbs+rIDd9u8/7pgY0t8/jGh+uasH50Z024S8Nl8Dsgdhug316dNHixcv1l133aWnn35a69evVyAQ0NChQ3XdddfpG9/4hoLBYKqXieNUFfH16zdqFbClLoWWKiK+7pkVkSSFg0bfOjOkMT0dhVpgwMbOCk9fe7JatTFfnQuNerW31bHAKC+gw4Z21/NVGfF1oNbX3mpfOyo8/endxNqCjnTbKSGN6O6oe7Fp8dAPAACwca+rbz+T2Kv89ylhjeiemhhijNGPJuXrmmmV+s0/a3XrySGKCjhuxvd9+hsyWHl5uXr27Ckpce4nR2Olzs4KT198tEoDutga3M2WJB2s9VW+39Pm/a6i8UTl+I7T8nRyb0ft8o7vibs25uvmh6tUG/M1soejsvbJ79bwfV8Ha31tr/C15YCnqrqplu3yjL51Zp6Gl9rKD/LCAuD4zJ07V+PHj5fEaxOAhKvvq9CTi2KaOjaoh24sTOlaXM/XCb84oHV7PM24rVCTh1HMynatlZnYMw20kE5ho4AtbT/oqf4aVbs8oyEltiadENCI7rZCjtH/e7tWNz5YqeeXRrWv2kv668xcHVN11NfQUvu4grSUuCpbnG9pUFdbZw9wdHp/R306WqqO+fqvV2s05f5KPb0oqg17XHG9DQAANEf5fk9PL44p5Ei/vaJt90kfjm0ZffvMRHv3b/5Zy3sdHDfCNNBCbMvoq6flqTLia8sB/zO39e5o66wBjsb3dlScb/TX9yO6+eEqPbkwqm0Hmhaqd1d6uuf9iDoWGPXu0DK/vsYYtc+3NLy7o/NOCGhUma32BUb3zY7oG09V64YHqzRvU5whHQAA4Lj89+s18nzpqpHBNjkGqym+cHJIncNG766L66018VQvBxkqPf41A1liYl9HAVtatv3w4dMYo25Flib2dTShj6POYaP750T0pcerdP+ciFZsd+V6Rw6tb6yOyfelwSV2q+xtdmyjnu1tTewb0NkDA+rbKVGt/tkrNbp2WqKavnkf1WoAANA0q3a4+usHEdmW9NML8lO9nAaFoUPV6T+/V5vi1SBTMYAMaEH5QaOfXZSvH79Qo4VbXI3vdfjQa4xR50KjzoWWDtR4WrvH01OLonpyYVRBR7plfEgndLHVq4Ol/KBRTdTXe+vjenheVN2LLXUsaP3rYIUho2Gljk7s6mvrQU8b9nr66/uRhtu+dnpIJ3a11bmQa3IAAODwfvxitVwvMUX7hK52qpfzCbdNCOlnr9bo2SUxbT3gqXsx72mQHMI00MKGd3d06ykh/ePDiD7eZY75wlGcb2lMmaUh3Xxt3u+pfP+h0CpJtiW5Xv19jYaXtu0LkWMb9epgq1cHWwdqPG3e72nLAU+/fiNxFbcoz+jLp4Z0YjdbXQuZBg4AABLmbYrr6cUxhYPSf1yYPlXpet2KLH1+RFCPL4jq7x9G9O9pVDlHZiBMA63g0qEBPbkwqlU7XYWDRj2aMCgsL2A0sIutgV1sVUZ87ar0tL/GV8SVgrbUOWzUo9iSncLjG4rzLRXnWxrczdfuKl/bKzxtP+jprn8mgnVhyOj2CYkrzz04ZgsAgJx215uJ9wffPCMvbfZKf9pXTg3p8QVR/fWDiP51Up6cFJx9jcxFmAZaQcA2+ss1YX3x0Sot3BJXwHbUNYkXkcKQUWEovVqhGrMto25Fif3fw0sSZ1dvO+hp60FPv3sr8cKZH0xUrIeX2kl97wAAIPO9uzamxxdElR+QvnpaXqqXc0RnDnA0uJulFTs8vbQipss4JgtJ4B0u0Era5Rn9+aoC2ZY0d3NcuyuTPwYrE1hWYu/38O6OzhsU0MR+jvp3tuT70u/fqtUXH63STQ9VatGWuOIug8sAAMgFv5qZuLj+o3Pzj/soz7ZgjNFXTk2E/b+8FznGvYFPSt9/2UAW6Fpk6U9XhWUbac6muHZlaaCuZxmjjgWWhpQ4mnSCowl9HfXsYOlgra9/e7FGU+6v0jtrY6qoJVQDAJCtlm2L66XlMRUEpa+fHkr1co7ppnFB5QekV1fFtG63m+rlIIMQpoFW1r04EagtI83ZGNfWJp4pnemMMeoctjSqh6NJJwQ0pMSWZaTfzKzVDQ9WasbSaNZW6wEAyGX1e6VvOyWkjuH0jxvtCyxdOzoo35cemR9N9XKQQdL/XzeQBboXW/rLNWEFbKOPyuNavye3rnoGHaP+nW2dM9DRSb0ctc83uvf9iL7waJWmL4xqW45cYAAAINtt2e/p4Y+isi3pO2em717pT7t+bKKC/hhhGkkgTANtpFuRpXunFCgcNFq6zdXSbXF5fm61O1uWUWk7S6f2dTSxr6OuhZYemBPRlx6v0oNzI9qQYxcZAADINn94p1YxV7p6ZFB9OqXvMNVPO3uAoy6FRsu2u1qyNZ7q5SBDEKaBNtShwNI/rgurY9ho/R5PczbGFY3nVqCWEi3gHcOWTu7t6Iz+jnq0t/TEwqi+8VS1HpgTUfl+KtUAAGSag7W+7nk/McTrB+dkTlVakhzb6JpRiUneD82jOo2mIUwDbawgaPSP68L60qkh7ar09c66uA7U5G54LM63NKbM0TkDAurZwdL0RVHd8USVHp8fof0bAIAMct/siA7W+jpnoKMxPTPvBN6bxiVavR/6KCLXy71iB5JHmAZSwLaMLh0a0J0X5SsS8/XeusQ+aj/H2r4bC4eMRvVwdPaAgEraWXpoXlRferxKTyyIaGcFoRoAgHTmeb7+9G5i8Ni3MmivdGPjetka1NXS1gO+3lgVS/VykAEI00CKGGM0tqejv1wTVn4gsY969iZXNbHcDdSSVBgyGtcr0f5d0s7Sg3Oj+uJjVXp2cVR7qwjVAACko1dXxrRmt6e+nSxdMiSQ6uUcF2NMQ3X6AVq90QSEaSDFuhdbuv+GsG6bENKuCk9vr4lp077crlJLifbvcb0cndbPUeew0d8/jOiWR6r04rJoTrfFAwCQjv74bmKv9NdOC8m2TIpXc/xuOCmxb/r5pVFFcnCuDZJDmAbSQMgxunx4UHddXiDbMlq0xdX763N7L3W9DgWWJvQJaELfxJFad8+K6KaHqvTy8qgqI7zIAQCQah/vcvXyipgKgtKtJ4dSvZxm6dXB1tietioi0psf0+qNoyNMA2lkUDdbD94Q1h2nhbS3JjGcbOGWuKqjhMbOYUsT+zo6ubejojyj/3svohserNTM1TFCNQAAKVS/V/qGsSF1KMj8eHH5sER1+pnFhGkcXeb/aweyTF7A6OIhQd17bVhXjwxq8z5P//w4poVb4qqoze3QaIxR1yJLp/dzNK6Xo/yA0e/fqtUND1bq5eVRHczxnw8AAG2totbXfbMTLd5fPz2zq9L1Pjcisef76cVRxVzeW+DICNNAmippZ+mm8SH94fMF6pBvtHmfp7fWxPTBhpg273dz+sndGKOSdpbOHOBobE9HhaFEpfrGulCd6xcdAABoKw/MjagiIp01wNHw7pl3HNbhDCt1NKK7rd1Vvl5nqjeOgjANpLl+nW3df0Oh/u/qAt08PqQ9Vb4Wlrt6bWVMczbGtWlf7k4At4xR9+JEpXp84/bvhyr16gpCNQAArcn3fd3zfqIq/Y3TM/M4rCOpH0T2EFO9cRTZcfkIyAE9O9jq2cHWRYMDWrHD1YodrqYvjGpHhSS5apdn1KXQUpdCo04FRlYGT9JMljFG3YqMuhYa7az0tXKHqz+9G9H/zYrojokhTejjqDifa4cAALSkBeWulmxz1bXQaPKwzDwO60iuHR3Uv8yo0QvLoorGfQWd3HlfhaYjTAMZJhwyOqmXo5N6ObpqZFBrdrtas9vTQ/MiWrvb1drdUsCWuhZZKm1nqWuhyegjKpLROFTvqPC1eperP78b0V/ei+i2CSGd1s/JisEoAACkg2lzElXpG04KKWBn13uNXh1sjS6ztaDc1Vtr4jr/xOy6WICWQZgGMlh+0Gh4d0fDu0tXDA9o+0FfH+929df3I9qy39OW/Z4CtlTW3lLvDraK8rLrhe5IEnuqE8F6V2UiVP/1/Yj+9kFEt58a0ml9HbUnVAMAcNwicV8Pf5Rogb5lfDDFq2kdlw8LakF5jWYsjRKmcVi8mwSyhDFGpcWWzugf0EM3FuqhG8P60aQ8hRyj9XsSw8s+3BjX7ipPvp8be4nrp39P7OvolD6OivON7pkV0c0PV+ml5VHO8QYA4Di9sCymvdW+xva0s2bw2KddVte6PmNpLGfeOyE52fkvH4CK8y1N7Gfp1L6ONu/3NHujqwfmRrSrwlPHsNGJXW11CufG9TRjjLoUGnUOG+2u8rVih6u/vBfRPe9H9LXT8nRaP0cFwdyo2gMA0BLqW7xvGZ8dx2Edzqgetnp1sLRpn6cF5a7G9CQ64ZNy4500kMOMMerVwdbVo4K6//qwbpsQ0r5qX++vj2vOprgqI7lzpTURqg9N/y4MGv3xnVpNfbBS762LKRrPnZ8FAADHa/tBTy+viClgS9eNyc4WbynxvuFQdZqp3vgswjSQQzqGLV0+PKhp14d10/iQdhxMtH8v3x7PqXOrE4PKLJ3R39GYno6CttGv36jV1AertGhLnFYuAACO4uGPInK9RBt0tne5XTYscbFgxjLOm8Zn0asA5KCOYUtXjwrqlN62/vWFGq3d7WnLAV9DS2yVtjMypu1bnr/8RFWT73vPNeEW+ZrGGPUoNiopMtq0z9Oqna7+7cUaFecb/fel+erVwW6RrwMAQLbwfV/3za4fPJa9Ld71zuzvqDCUOAZsy35PPdpn98UDJId/DUAO69nB1oM3hHXnRflyPV8fbY5r7iZXNdG2rcwmE6Tr71//pyXYllHfTrbOGRhQv86WDtb6+tr0aj21iCFlAAA0tqDc1bLtrroVGV2YAxOug47R+YMS3+dLK2j1xicRpoEcZ4zR2J6OHryhUF84OaQdFZ7eXBPThr1uRrQ7t2SoDjpGQ0scnTUgoG5FlqbNTkz+nrUuJtdL/58FAACt7ZH5iUB53ZignCw7W/pILh2aaPV+gVZvfAphGoAkKRwy+vzIoH73uQIFbaMlW129vyFzBpS1ZKguDBmN750YUpbnGP3qjVrd+FCVNu1zW+TxAQDIRJ7n6/EF9WE6+1u86100OFGZfmN1TLWxzHhfhLZBmAbwCQO62HrwxrDuOC2kvdW+3lkb0/o9mVGlllo2VHcrsnTmAEcndLVVFUm0fj+/NNrmbfAAAKSD99bHVb7fU//Olsb1yp25IiXtEt9vdVR6Z2081ctBGiFMA/iMkGN08ZCg/vj5AoUco6XbXH2wIa6qVgqRLTVQrLGW2ldtW0aDuto6c0BAncJGf30/opsertKqHVSpAQC55ZGPDrV4p2JYaSqdV7dv+p8f0+qNQwjTAI6oT6fEgLKvTAxpT5Wvt9fEtGlf61SpWyNQ12uJUF0YMprQx9HIHrZcz9f3n6vWkwujqsqQNngAAJoj5vp6clHutXjXO2cgYRqfRZgGcFRBx+iSoUH9/vMFCthGi7a4mrPJzcg9Q80N1cYY9epg66wBAZW0s3T/nIhuerhSy7bR8gUAyG6vr4ppT5WvEd1tDSnJnRbveqf2cRS0pY82u9pfzUkfSCBMA2iS/p0TVepbTwlpZ4Wnt9bEtPVAy76YtGZ1urHmhuq8gNFJPW2N7elIkn70fI2eWBDJmGFtAAAk6/45h1q8c1F+0OjUvo48X3pnHRfRkUCYBtBkeQGjz40I6n8uK5Ax0keb45pfHlfMbbkQ2VaBWmpeqDbGqHuxpbMGBFRabOnBuVHd/HCllm9nLzUAILts2ufqyUVRBW3ppnG51+Jdr6HVezWt3kggTANI2uASW/dPLdR1Y4Past/TW2vi2l3ZclXqtgzUUvNCdcgxOqmn01Cl/uGMaj29KJqRbfAAABzOIx9F5fnSNaOD6l6cu/HhnIGJ1/p/fkxlGgm5+9sAoFkKQ0bXjQnqzovyFXN9fbAxrmXb43K9lgmRbR2opeaF6u7Fls7sH1C3dpbumx3RTQ9VacMeqtQAgMz36PxEi/fUsbnZ4l1vXC9H4aC0ZJurnRXsmwZhGkAzGGM0tqejf1wXVod8o3W7Pb2zNq4DNS3zApOKQC0d/7FaeQGjcT1tjehuK+r6+sbT1Xp9VUzxFmyDBwCgLS3dFtfira66FBqde0Ig1ctJqaBjdEb/xM/gTaZ6Q4RpAC2gY9jStOvD+uGkPFVHfb27Lq7Vu1x5LXCEVqoCdb1kQ7UxRr072jqjf0Dt84z+39u1uuWRqhZtgwcAoK3UV6WvGRVUwM6ts6UPh1ZvNEaYBtAiLMvotH4B3X1NWIUho1U7XM1aF2+RCdf3XBPOuFBdGDKa2NfRwC62DtT6uu2xKs3bFG+VM7oBAGgNvu/rkY8SYfr6HG/xrsd502iMMA2gRZUWW3pgaljfOCNP+2t9vbM2pg173RYJkakO1JKSCtSWZXRiN1sT+zgK2kY/e6VGD38U1cFaAjUAIP19uCGuDXs99e5gaUIfJ9XLSQsje9jqUGC0ZrenTfuYjZLrCNMAWpxjG51/YkB/urJAQcdoyVZXcza5isSzJ1AnE6o7hi2dOcBRzw6WHp8f1S0PV2rlDl6AAQDp7ZH5h86WNoYWb0myLaOzBiQuLLxJq3fOI0wDaDW9O9p6YGpYt54S0s4KT2+vibfI9Mt0CNRScqE6YBuN6uFoXK/EC/APnqvWi8uiLXpGNwAALSXu+np8AS3eh1Pf6j2T86ZzHmEaQKvKCxh9bkRQ/31pvlzf1+yNcS3ZGm/2hOt02EddL5lQXdLO0pkDAuoUNrp7VkS3PMxwMgBA+pm5OqZdlb6Gldoa3p0W78bOrqtMv72WynSuI0wDaBPDuzu677qwOoWNNuz19M66uPZVZ0+VWmr6fur8gNEpfRwN6mrrYCQxnGzxVl6QAQDp47H6qvQYqtKfNqTEVuew0aZ9njbuZdtWLiNMA2gz7Qss3Xd9WD8+L081MV+z1se1amfzj9BKt0DdlFBtGaMTutqa0MeRYxn95IUaPbM4qmgL7CsHAKA54q6v55clWpivGkWY/jRjjE7vX1edXsPF8FxGmAbQpowxmtA3oHuvDasoZLR6Z+IIrapmHqGVTm3fUtNDdee64WRdiyz948OIbn64SjtaYF85AADH6/0Nce2p8jW4m6WBXexULyctndk/sW/6nXWE6VxGmAaQEl2LLN0/Naxvnpk4QuvttTFt3t/8Vql0CtRS01q/Q47R+F62BpfYqor6+tLjnEkNAEidZxcnWrwvH0ZV+kjOaKhMM4QslxGmAaSMYxudNyig//f5AgVso4XlruaXx5s94TodA/WxQrUxRgM625rY11HISZxJ/eSimGpjBGoAQNvxfV/PLU0ExMuHE6aPZER3W8V5ifOmtx6goyxXEaYBpFzfTrbunxrW9WOD2rLf07trmz+cLN0CtdS0KnWHAktn9ndU0s7SA3MiuuUR2r4BAG1n2XZX6/Z4KmmX6JrC4dmW0cR+ier0rPW0eucqwvQRVFRU6M4779Tw4cNVWFio4uJijRs3Tr/97W8VjUab9divv/66rrnmGvXu3Vt5eXnKz89Xv379NHXqVL399tst9B0AmaUgaDRlTFD/fkG+amK+3l8f15rdbrNandNtH7XUtCp1wDY6qaetYaW2aqK+vvx4lRaW80INAGh9zy1JVKUvGxqUZZkUrya9TexbF6bX0eqdqwjTh7Fx40aNGDFCP/vZz7R06VL5vq9IJKJ58+bp+9//vk455RTt27cv6cf1fV9f+cpXdP7552v69OnatGmTjDEyxmj9+vV65JFHdNZZZ+m73/1uK3xXQPozxmh8b0f3XBtWQdBoxXZXH26MN7vVOd0CtXTsKrUxRn072ZrQ11HANvr3l2o0Y2m02S3wAAAczXNL6/ZLDw+keCXprz5Mv7+BC965ijD9Ka7ravLkydqwYYNKS0v1+uuvq6qqStXV1XrsscdUVFSkBQsWaOrUqUk/9rRp03TPPfdIkq666iqtXr1aNTU1qq6u1sqVK3X55ZdLkn73u9/pmWeeadHvC8gkJe0sPXBDWF86NaTdlb7eXRfX7qrcbPvuWNf23aXQ6N73I/rCI1Utcj43AACftvWAp7mbXIWD0jkDCdPHMq6XI8eSFpS7qo5ysTsXEaY/Zdq0aVqyZIkk6amnntKkSZMkSZZl6dprr20Iwy+//LJmzpyZ1GM/8MADkqQBAwbo0Ucf1cCBAxtuGzRokKZPn65+/fpJkp544olmfy9AJgvYRpOHBfXfl+Yr5vr6YEPzz6TO1LbvoJOo2J/Q1daBWl9ffLRKH+9q/uRzAAAam1FXlb5wcEB5AVq8j6UgaDS6zFbck+ZspDqdiwjTn3L//fdLks4++2xNmDDhM7dPmTJFffv2lXQoHDfVtm3bJEkjR46U4zifuT0QCGjUqFGSpMrKyqQeG8hWw7s7+vt1YbWrO5N6do62fVvGaFBXW+N7JZ47vvtstd76OMbxWQCAFlO/X5ojsZqOVu/cRphupLq6WrNmzZIkXXTRRYe9jzFGF154oSTptddeS+rx66vOixYtUjz+2V+4WCymhQsXSpJOOumkpB4byGadwokzqb88MdH2/c7a3G377lZk6fT+AbULGf32zVo9Nj+qSJxADQBonoO1vv75cUy2JV0yhBbvpprYN/GzYqJ3biJMN7JixQp5XuIN+rBhw454v/rbtm/frr179zb58e+44w5J0po1a3TddddpzZo1DbetWrVK11xzjdatW6f+/fvrO9/5zvF8C0DWcmyjS4cG9cvJ+Yp7ud32HQ4andbPUY/2lh75KKpbHq7SnmZeXAAA5LZXVkQVdaXT+znqGCYiNNWp9ZXp9XF5Hhe3cw2/KY1s3bq14eMePXoc8X6Nb2v8OccyefJk/e53v1MwGNSTTz6pgQMHqqCgQAUFBTrxxBP11ltv6Y477tCcOXPUrl27pNe/bds2lZeXH/UPkOmGlTr625SwivMOtX03tzKbboFaOnaV2raMRvewNaTEVmXU1+2PVWkN+6jRRo71WrNjx45ULxFAkp5bSov38ehebKlPR0v7a3yt3MmF7Vzz2Y27OayioqLh44KCgiPer/FtjT+nKb797W9r4MCBuvXWW7Vz507V1NQ03BaJRFRRUaEDBw6oY8eOST2uJI0fP/6Y92F/JbJB50JL064P69WVMd09K6J31sY1psxWp2ZcSb/nmnCT2qzb0pefqDpq0DfGqH9nW+3yjD7aHNd3nq3Wj87N06l9HRnD4Bi0np49e6Z6CQBaUMz19dLyujDNkVhJm9jX0Ya9Uc1aH9OQEjvVy0EbojLdhqqrq3Xttdfq0ksvVa9evfTaa69p9+7d2rVrl1577TUNHTpUDz30kMaPH6/FixenerlAWnNso0uGBvXzSxLTvt/fENfqXc1v+043TWn77lJo6fR+ARWFjH71Rq2eWRxTnPOoAQBN9M7auPbX+BpeaqtvJ8JgsuqHkM1ax77pXENlupGioqKGj6urq494v8a3Nf6cY/nBD36gJ554QieccILeeecd5efnN9x23nnn6bTTTtOoUaO0evVqfe1rX9O7776b1PrnzJmj0tLSpD4HyHQjezi6d0pY3366Wqt2uNpb5WlMmaOgc3yV2fpAnWlV6nDIaGJfRwu2uLpvdkTPLonq/64OqzBEhRotb/PmzUe9feHChZo8eXIbrQZAcz23JHEkFlXp4zOhTyJSzeZ4rJxDZbqR7t27N3y8ZcuWI96v8W2NP+doKioq9Ne//lWS9PWvf/0TQbpefn6+vv71r0uS3nvvPe3cubNJj12vtLRUZWVlR/0DZKMuhZamTQ3r9lND2lU37Xtfde5N+w7YRuN62hrQxda+al+3PlqlnRXs30LLO9ZrTbdu3VK9RABN5Ps++6WbaViprYKgtHKn1+z3H8gshOlGBg8eLMtK/EiWLl16xPvV31ZSUtLkvc2rV69uOA6rf//+R7zfwIEDGz5ev359kx4bQCJIXjYsqJ9dlK+I6+v99XGt3+M2a05AJgZqY4wGd7M1soet2pivLz9RpbW7GUwGADi8pdtcbdrnqUex0dietHgfD8c2GluWqE7P3cRrbi4hTDdSUFCgiRMnSpJeeeWVw97H9329+uqrkqTzzz+/yY9dH9IlaePGjUe8X+MJqMm0kANIGNPT0d1Xh5UfMFq6zdW8za6izZj2na6B+lihulcHW6f0dmQkfeeZas3bROsZAOCz6gePXTwkyPDKZjilrtX7ww283uYSwvSn3HzzzZKkN998U7Nnz/7M7dOnT9e6deskSTfddFOTH/fEE09saO3+29/+1lClbsx13YZW8A4dOmjQoEFJrx+AVFqcaPu+eXxI2w96zW77TsfzqKVjV6k7F1o6rV9A+QGjn71So1dWRDkDEwDwCS+vSITpiwazX7o5Tu5dt2+ai9c5hTD9KTfffLOGDx8u3/d15ZVXaubMmZIkz/M0ffp03X777ZKkiy66SOeee+4nPnfatGkyxsgYo7feeusTt+Xn5+u2226TJM2fP1+TJ0/WkiVL5HmePM/T4sWLdfHFF+v999+XlDhCy7ZptQGOV17A6MqRgZxv+y4MGZ3ez1HHsNGf343o8QVRxZj0DQCQdKDG03vr4wrY0rknEKaboyFMb4xzFG0OIUx/iuM4mjFjhvr06aMtW7Zo0qRJCofDCofDuuaaa3Tw4EGNHj1aDz/8cNKP/etf/1oXXnihpEQb+YgRI1RQUKCCggKNHDlSr732miTpuuuu009+8pMW/b6AXGSMyZm276MJOkan9HbUo72lRz6K6ouPVqkqwgs9AOS6N1bH5XrSaf0ctcujxbs5ytpb6l5stKfK19rdDCHLFYTpw+jTp48WL16sn/70pxo2bJiMMQoEAho7dqzuuusuffjhh+rQoUPSj5ufn6+XXnpJ06dP1+WXX66ysrKGK1c9e/bUlVdeqRdeeEGPPPIIVWmgBZUWW7r/hk+2fe+tal7bd7o5VqC2LaPRPQ5N+v7iY1Xaz8RRAMhpLy1PHIl1MS3eLWJcz0R1+qPNtHrnCuPTh5DRysvL1bNnT0mJcz85/go4Mt/3taDc1X++WiPPl07oYmtAF0vWcQ5cSbezqOsdK+yv3+Nq6XZXeY7Rn64qULcirquiZc2dO1fjx4+XxGsTkK5831eP/9ivbQd9Lf1hOw0tdVK9pIz3X6/W6Kcv1+gH5+Tpfy4rSPVy0EhrZSbeQQHIGfVt3/dOCasoZLRqp6sP1sdVHT2+a4qZOpisbydbY8scReO+vvJEldbv4RgPAMg1i7a42nbQV68OloaU0BHZEuqPFqMynTsI0wByTpdCS/dPDetrp4e0t9rXW2ti2rj3+IeTZWKg7l5s6eS6Yzy+/XS1lm8nUANALmk8xZsjsVrG2Lo27/nlzRt4isxBmAaQkxzb6MLBQf3+8wUK2EaLt7qau8lVbSx3AnXnsKWJfQMK2EY/fL6aK+kAkENeWlF/vjT7pVtKtyJLPYqN9tf4WreHuSS5gDANIKf172zrwRvCuvWUkHZUeHp7bUxbDnjHdUU5EwN1uzyjif0cFQSM7ny5RrPWxdpoZQCAVNlX7emDDXEFbemcgYTpljSmjCFkuYQwDSDn5QWMPjciqF/XDQuZvzmueZuPr0qdiYE6HDSa2NdRUZ7Rr96o1czVMdrTACCLvb4qJteTzujvqDBEi3dLatzqjexHmAaAOkNKbD0wtVA31R2hdbx7qTMxUOcFjE7t46hDgdHv36rVKysI1ACQrRrvl0bLYghZbiFMA0Aj4ZDR1aMSVWrbSuylnrU+rora5AN1uoXqLz9RddRQHXSMTu7tqFPY6P/ei+j5ZTF5HoEaALKJ5/kNYfriIcEUryb7jK1v82YIWU4gTAPAYQwpsfXQjYmJ3/tqfL29NqaVO9ykw2W6BWrp6FXqgJ0I1F2KLN37fkTPLiFQA0A2WbDF1Y4KX306WhrUlSjQ0kqLLZW2M9pX7WvDXoaQZTt+gwDgCEJOYuL3X65OnEv98S5X766L60BNci+OmRaobctoXE9b3dpZum92RE8tjsklUANAVni50RRvjsRqHfX7pj/azL7pbEeYBoBjKGtv6YEbwvre2XmqiPh6d11cy7e7SQXMTAzUJ5XZKi229MCciJ5aFCVQA0AWeGk5+6Vb29gy9k3nCsI0ADSBbRmdNTCge6eE1S7PaO1uV2+tiWt3ZdOr1JkWqC3LaEyZre7Flh6cG9WTCwnUAJDJ9lR5mr0xrpAjnT2AMN1aGirT5YTpbEeYBoAkdCuy9MDUsP71vDxF4r4+2BDXkm3xJofMjAvUxmh0XaB+aF5UTy2KsocaADLUaytj8nzprAEBhTkSq9U0Ph6LIWTZjTANAEmyLKNT+wb0j+vC6hQ22rDH0ztrm76XOpMD9YNzo3p6MUPJACATcSRW2+hebKmkndGeKl+b9jGELJsRpgHgOHUMW/rHdWH9y7l5qor6em9dXGt2N+0qdLoenXUkljEa3cNWSTtL98+J6LmlnEMNAJnE83y9svLQ8DG0roYjshhCltUI0wDQDJZldHr/gO6+JqyCoNGK7a4+3BhXbSwz276PtYd6bJmtbkWW/vFhRC8sI1ADQKb4qNzVrkpf/TtbGtjFTvVyst6Y+iFk7JvOaoRpAGgB3Yst3T81rC+dGtLuSl/vrG36cLKMC9Q9bXUpNPrr+xG9vipOoAaADFA/xftiWrzbBMdj5QbCNAC0kKBjNHlYUL+cnK+45+uDjXGt3tn0tu90csxzqHs56lhg9Md3avXuOq66A0C6e2l5VBIt3m3lUJjmonM2I0wDQAsbVuro79eF1S5ktGqnq7mbXMXc7AvU43s7apdn9JuZtZq9gUANAOlqZ4WnuZtd5QcSk7zR+noUG3UtNNpd5WvzfoaQZSvCNAC0gk5hS9OmhnXrKSHtqPD07rq4DtZmV6AO2Ean9HFUGDL6+Ws1WryVQA0A6ejVlTH5vnTuCQHlBTgSqy0YY2j1zgGEaQBoJQHb6IrhAf3b+fmqjvp6b11M5U24Op1JgTrkGJ3S21FewOjfXqzRx7t4wwAA6eZF9kunxNieiSFk8xlClrUI0wDQiowxOrmPoz9fVaCAbbSgPK6l2+LyjrF/Kt0C9dHkBxOB2rGkHzxXrS20swFA2oi7vl5dyfnSqcDxWNmPMA0AbaBnB1v3XR9Wx7DR+j2e5myMKxrPnEB9tOq0JBXlJfZQS9I3n67S3ioCNQCkg9kb49pf42twN0t9OnEkVluqr0wzhCx7EaYBoI0Uhoz+cV1Yt58a0q5KX++tj6sykj2BumOBpTFljqKu9NXp1ao6xvcGAGh9L62oa/EeEkzxSnJPWXtLXQqNdlb62nKA18RsRJgGgDZkW0aThwb07xcc2ke96xjnUWdSoC5pZ2l4qa2qqK8vP1F1zOo7AKB1vbyCFu9UMcY0avVm33Q2IkwDQBszJtES/fvPFUhKtOBt2nf0/VSZFKj7dLQ1sIutAzW+pi+MyvMI1ACQCtsOeFpQ7ioclE7r56R6OTmpcas3sg9hGgBSpF9nW/dcG1ZewGjRFlcrdrhH3VOVSYF6UFdLPTtYemx+VC+tiLFXDABS4JW6wWOTTggo5HAkVio0HI9VzhCybESYBoAU6lJo6b7rwupQYLRml6sFW9yjVnIzJVAbYzSi1FbnQqN7ZkX0/nquyANAW6PFO/XGlHE8VjYjTANAioVDRn+/Lqzrxwa1Zb+n2ZviirmZH6gty+ikno6K8ox+9Uatlm/nqjwAtJW46+u1+iOxhhCmU6VXB0udwkbbD/raeoCTLrINYRoA0kDANrp2dFBfnhjS7kpf76+PqzaW+YE6YBuN7+Uo6Eg/fqFa23gjAQBt4oMNcR2o9TW0xFavDhyJlSqJIWTsm85WhGkASBOWZXTJkIB+cG6eDtYmAnV1NPMDdUEwEah9Sd98uloVteyfBoDWRot3+mjYN02YzjqEaQBII8YYndE/oP+4MF9VMV+z1sd18CjhM50C9dF0KLA0qruj2pivO6ZXHbWNHQDQfC83nC9NmE41hpBlL8I0AKShk3o5+uWl+YrEfb2/Pqa91Uduj06XQH2sCd892ls6oWviyKxnFkeZ8A0ArWTrAU8Lt7gqDEkT+3IkVqrR5p29CNMAkKaGlTr63ysK5PnShxvi2lGR+YH6hC6WuhdbenBuVG+t4U0FALSGV1YcOhIryJFYKde7o6WOBUbbDvrafpDZIdmEMA0AaWxAF1t/vios25Lmbopry/7MDtTGGI3qYas43+h/36zVsm0EagBoaS+viEqixTtd1L/2SdKiLbzuZRPCNACkuR7tLd1zTVghx2j+lrg27j3ynqtMCNS2ZTSul6OQY/STF2u4Sg8ALSjm+nptVSKwXTQ4mOLVoN7I7nVheiv7prMJYRoAMkDnQkv3XlugcMBo8VZX6/ZkdqDODxid1MuW7ycmfFdF2D8NAC3hvXWJwZXDS22VteetfroY2SOxd50wnV34DQOADNG+wNK9U8IqDBkt2+Zqza70D9RH07HA0vDutmqivr76ZJVcj0ANAM31wrLEfunJw2jxTicNlekthOlsQpgGgAxSlGf0tylhtcszWrHD1cdpHqiPNZCsVwdbfTtZ2lvl68XlMSZ8A0AzvbAssV/6UvZLp5XBJbYcS1q501VtjNe6bEGYBoAMEw4Z3VsXqFfucLU6wwP1kBJbXQqN7n0/ojkbuWIPAMfr412uVu/y1DlsNL43R2Klk5BjNLibLdeTlm/ntS5bEKYBIAMVBBOBujjfaNUOV6t3Zm6gtozR2J6OwiGjX7xeow1H2Q8OADiyF+tavC8eEpBtcSRWuhnZgyFk2YYwDQAZqiBodO+1dYF6Z2ZXqAO20biejiwjfe+5ah2oYcI3ACTrtVWJMH3hibR4p6ND+6Y5HitbEKYBIIPl1wXqdnmJCvWa3ZkbqIvyjEaXOYrGpa89Wa2Yy54yAGiqSNzX22sTYXrSIMJ0OhrRnYne2YYwDQAZLj94aA/1iu1HPzYr3ZW2szSwi60DNb6eX8pAMgBoqg/Wx1UdlUaX2epSyFv8dNS4zZvXt+zAbxoAZIGCoNE91x46NutIgTrdq9OSNKirpZJ2lu6bHdEHG2iFA4CmqG/xPu8EqtLpqluRpW5FRvuqfZXvZztTNiBMA0CWKKyb8l0fqDfszcxAbYzRqB62CkNGv3yjVuuO0roOAEh4vT5M0+Kd1hr2TdPqnRUI0wCQRQpDRn+9Nqxw0GjJVleb92VmoA7YRuN6OXIs6QczGEgGAEezs8LTR+Wu8gPSaf04EiudjexRt296C2E6GxCmASDLFOUZ/XVKWPlBo4VbXW07ePggmu6BujBkNLpHYiDZ15+sVpyBZABwWC+viMn3pXNPCCgvwJFY6YzKdHYhTANAFmqXZ3T31QUK2tL88rj2VmVmoC5pZ2lAF1v7a3y9uJyBZABwOC8si0qSLhlCi3e6axhCxvFYWYEwDQBZqmPY0h+vDMtImrMprsrI4YNoOgTqoxnU1VKXIkt/+yCi+eVcyQeAxqJxX6+tSgQzwnT6G9TVVtCWPt7tqeoIr8vIHIRpAMhi3Yst3XV5geKe9OHGuGpj6Rmoj1adtozRmB628oNG//lKjbYeYP80ANR7b11cB2t9jehuq2cHO9XLwTEEbKOhpbZ8X1q6jQvEmY4wDQBZbkAXWz+7KF81UV9zNsUVO8Le43QO1EHHaFxPWzLSt5+pVk2Uq/kAIEkvrUhM8aYqnTkO7Zum1TvTEaYBIAeMLnP0/XPydKDG10flrjwv8wJ1cb6lYSW2aqK+nlkSZf80ACgxfEySLiZMZ4yR3esmejOELOMRpgEgR5zR39Htp4a0q8LT4m1uRobRXh0slbW39OhHUc1azxV9ALlt415Xy7e7ap9vdEpvjsTKFIeGkBGmMx1hGgByhDFGlw4JaMqYoDbv87R6V3pO+D5addoYo+GltgpDRv8zs1Yb9/JGBEDuqq9KnzcoIMfmSKxMUd/mvXhr/IidYsgMhOnDqKio0J133qnhw4ersLBQxcXFGjdunH77298qGo02+/EPHjyoX//61zr11FPVpUsXhUIhlZWV6eyzz9add96p/fv3N/+bAIDDsCyjq0cFVZxvtHqnq037Dh9G0zlQO7bRST0dWUb6/nM1TEMFkLMaWrwH0+KdSTqGE11WFRFpw16GamYywvSnbNy4USNGjNDPfvYzLV26VL7vKxKJaN68efr+97+vU045Rfv27Tvux3/zzTd1wgkn6Ec/+pE++OAD7d+/XwUFBdqyZYveeust/exnP9OGDRta7hsCgE8JOkZ3XxNWftBo8VZXuyozr0JdlGc0oruj2pivpxdHubIPIOdE4r5mrk6E6QsJ0xnn0BAyOqwyGWG6Edd1NXnyZG3YsEGlpaV6/fXXVVVVperqaj322GMqKirSggULNHXq1ON6/FmzZumSSy7Rjh07NGnSJL333nuKRCLat2+fqqurNW/ePP3kJz9RcXFxC39nAPBJhSGj/7uqQI4lzdscV0Vteg4kO5qy9pZ6dbT0xIKo3lnL/mkAueXdtXFVRaXRZbZK2vGWPtMc2jfN61cm4zevkWnTpmnJkiWSpKeeekqTJk2SJFmWpWuvvVb33HOPJOnll1/WzJkzk3rs6upq3XTTTaqpqdGVV16pV199VRMnTpRlJf4vyM/P19ixY/Xzn/9cffv2bcHvCgAOr3Nh4gxq15PmbIorEk+/QH206rQkDSuxVZxv9Nu3arVuN1f3AeQOWrwzG5Xp7ECYbuT++++XJJ199tmaMGHCZ26fMmVKQ9B94IEHknrsBx98UOvWrVN+fr7uvvvuhhANAKnUr7Ot/7gwX9VRX/M2x+Wm4ZFZRwvUtmU0tsyRbaQfPs/+aQC546XliTk+FxGmMxLHY2UHEl2d6upqzZo1S5J00UUXHfY+xhhdeOGFkqTXXnstqcevD9+XX365Onfu3IyVAkDLGtvT0ddOD2lvla+lRzkyK10DdTh0aP/0c0s5fxpA9tuwx9XKnZ7a5xudzJFYGWlAF0v5AWn9Hk8Hj7DVCumPMF1nxYoV8rzEEJ5hw4Yd8X71t23fvl179+5t0mPXDzCTpDPPPFPr1q3TF7/4RZWVlSkUCqmkpESXX365Xn755WZ+FwBwfM4fFNC1Y4LatM/Tuj2ZN1m0rP2h86c/3MD+MwDZrb7F+3yOxMpYtpU46lFKHJGFzMSlrDpbt25t+LhHjx5HvF/j27Zu3aqOHTse87E3bNjQcKRWeXm5RowYoaqqKgWDQRUUFGjHjh2aMWOGZsyYoa985Sv6y1/+clzfw7Zt2455n7KysuN6bADZzbKMrhoZ1IvLYlq+w1VRyKhr0Wevt95zTfiY+5hby5efqDpqdXx4qa19Nb5+9Uat/nKNre7FXC9uLeXl5Ue9fceOHW20EiA3vVS/X3oILd6ZbGQPR3M2uVq0xdVp/fj/MhPxTqNORUVFw8cFBQVHvF/j2xp/ztE0Pkrrl7/8pQKBgB599FFVVlZq37592rRpk6ZMmSJJuvvuu/WHP/wh2eVLksaPH6+ePXse9Q8AHElewOjPVxUoYEnzy+OqimZWu7djG40tS1zl/96z1Yq5tM21lmO91kyePDnVSwSyVm3M1z8/5kisbMAQssxHmG4D9e3j9R/ffffdmjJligKBxBNgz5499fDDD2v06NGSpJ///OeKx2n3AND2OoYt/fqyAsU8ad6muOJHCKTpemRWcb6lQV1tVUYOnb8KANnk3XVxVUelMWW2uh2mgwiZg+OxMh+/gXWKiooaPq6urj7i/Rrf1vhzmvrYPXv21LXXXvuZ+1iWpe9973uSpN27d+ujjz5q0mM3NmfOHG3evPmofwDgWAZ2sfUv5+TpYK2vhVvTbyDZsdrM+3W21LHA6M/vRvTxLq72t4ZjvdY8//zzqV4ikLVeqWvxZop35htRN9F7yTb3iKdpIL2xZ7pO9+7dGz7esmWLRowYcdj7bdmy5bCfczSN91mfeOKJR7zf4MGDGz7euHGjTj755CY9fr3S0lL2RANoEaf1c7S9IqQH5kT0cZ7RCV3sw94vVXuoj7Z/2jJGY8ocvb02ph89X637pxaqMMSAnpZ0rNeapszwAHB8XluVCNMXnEiYznTt8oz6drK0fo+nNbs8Dep2+NdapC8q03UGDx7ccPbz0qVLj3i/+ttKSkqaNHxMkjp27NgQqI058hu6xtWfo90PAFqbMUaXDwuoXZ7Rqp2udlRk1oTv/GDiuKxoXJrBcVkAssTWA56WbnNVFJJO6UNNLBsc2jdNq3cmIkzXKSgo0MSJEyVJr7zyymHv4/u+Xn31VUnS+eefn9Tj199/xYoVR3xTt2LFioaP+/btm9TjA0BLCzpGf6obSLagPK6qSGa1e3cvttSj7risjzbT7g0g8722MlGVPmdgQAGOxMoKIxhCltEI043cfPPNkqQ333xTs2fP/szt06dP17p16yRJN910U1KP/YUvfEFSYp/Z448//pnbPc/T//7v/0pKtIWPGTMmqccHgNbQocDS/9QPJNscP+KernQdSDa81FZ+wOjnr9Vob1VmVdcB4NPqW7zPp8U7a4ys2ze9aAthOhMRphu5+eabNXz4cPm+ryuvvFIzZ86UlAi606dP1+233y5Juuiii3Tuued+4nOnTZsmY4yMMXrrrbc+89inn366rrrqKknSHXfcoccff1yxWOIJcfPmzZo6daoWLFggSfrFL37R0HIOAKk2oIut752VGEi2ZFt6DSQ7VnU6YBuN6mHL9aTvPFtNuzeAjOV5vl5nv3TWaZjoTWU6I5HYGnEcRzNmzFCfPn20ZcsWTZo0SeFwWOFwWNdcc40OHjyo0aNH6+GHHz6ux582bZrOOOMM7d+/X1OmTFFRUZE6duyoXr166bHHHpMk/fSnP22okANAujhzgKMpY4LavM/Txn1HrvCmY6DuXGipT0dLe6t8fbiBPWkAMtOCLa52V/nq18lS/84MqsoWfTtaKgxJ5fs9OqgyEGH6U/r06aPFixfrpz/9qYYNGyZjjAKBgMaOHau77rpLH374oTp06HBcjx0Oh/Xmm2/q3nvv1RlnnKFwOKzKykr16NFDU6ZM0axZs/Szn/2shb8jAGg+Y4w+PyKogqDRsm2u9lWnV6A+liEltgqCRr+eWXvUtQNAuqrfL33+IKrS2cSyjIaVJC6OLN1OdTrTGJ+et4xWXl6unj17Skq0i3M0FoDWtP2gpy8/UaWQbXRGf0dB5/ADcFJxXNaxQvzuSk8fbIjr6lFB3TguyKkJrWju3LkaP368JF6bgJZy9p8O6q01cT1za6GuGBFM9XLQgr70eJXu/SCiP19VoK+elpfq5WSl1spMVKYBAE1W0s7SnRfmqybma+HW9No/fSydCy317GBp+sKoFjLoBUAGOVDj6b11cTmWdPZAjsTKNsNLE5XpJeybzjiEaQBAUkb1sHXLySHtOOhp3Z70afduSjV8aImtkGP0X6/WqPIIR30BQLp5Y3VccU86rZ+j4nzevmebYfVhehthOtPw2wgASIoxRpcODSgcMlqxI7P2Twdso2GltmKuGqbiAkC6e3FZVJJ0yRD2S2ej+sr00qOcmIH01OpheuzYsXrvvfda+8sAANpQyDH6w+cLZIw0v9xVzD3yi39bBuqmVKdL2xl1LbL0jw8jWrebKgCA9Pb/27vvMKnKg/3j9znTdmd3WZZepUuRBZEmGkUUCRg7BrFiNMbXNI0CJjGv4hu7YE2MRqPoz14jakARIQYU6U1BlM4uvW3faef3x+yOENkyy8ycmZ3v57r2YnbOmTPPHHbmOfc8LRSyNLNq8rGf9GGsdGPUIttUmyaGDlVY2n6QSTJTSdzD9PLlyzV8+HCNHz9eW7dujffTAQASpHWOqT+NylSZz9KqWsZPS8nVQm0YhvLbOmSa0u0flisYohUAQPJaXhDUziJLnZuZ6tWaTqWNVT5dvVNSwt6Rb775pnr37q077rhDZWVliXpaAEAcDero0JWD3Co8VPv608nG6zbUvYVDJZWWvtzC2tMAkteHX1W3SrtYhaARq14ei0nIUkvcw/Sbb76pzp07y7IslZeX65577tHxxx+vl19+Od5PDQCIM8MwdEH+9+tPHyy3f/x0fZfl6t7CVKbb0EOfVjAZGYCk9a+14fHS5zBeulHLbxeepZ21plNL3MP02LFj9fXXX+vPf/6zsrKyZFmWCgsLdfXVV2vYsGFatGhRvIsAAIijDJehRy/ySpKWbkue8dN1cZiGTmjjUCAozf2WycgAJJ89JSEt2hpUhksa0Z0w3ZixPFZqSkg3b4/Ho9tvv13ffPONLr/8chmGIcuytGjRIp1yyimaMGGCCgsLE1EUAEActM01dXvV+OmVSTB+ur6t021yDDXPMvT3Lyq1syh1uqkDSA+z1vplWdKZPVzKdNPFuzHr08Yhw5DW7qr9S2kkl4TOYtCuXTu99NJLmj9/vgYNGiTLshQKhfTSSy+pZ8+euueee1RZWZnIIgEAYmTwcQ5dPcSjHYdC2rQ/NYKpYYRbp2VJc9bTOg0guXz49ffjpdG4ed2GujU35QtK3+5JjToUNq0zXd29+x//+IfatGkjy7JUWlqqO+64Qz179tSbb75pR7EAAMfAMAyd39elLLehtTuTY/x0feRmmmrf1NRry3zawFJZAJJEIGjpo6olsc7pTZhOB/ntqtebZmLMVGHr/Po/+9nPtH79ek2cOFFut1uWZWnr1q0aP368hg8frhUrVthZPABAlDxOQ49eHB4/vczm8dP17eotSb1bOWQa0p0zy2vtog4AibJwS0AHyy31aeNQ5+YOu4uDBIjM6M3yWCnD9sXqsrOz9eCDD2rNmjU699xzJUmWZek///mPBg0apOuvv167d++2uZQAgPpq08TUH87OVKnP0uod9o+fro9Mt6FOzUwdKre0dhfd6wDY719f0yqdbqpn9GYSstRhe5iu1r17d82YMUOzZs1S7969JUmhUEjPPfecjj/+eE2dOlV+P+PZACAVDO3k0OUD3So4GNK2g7WH03gG6mhap3u0dMhhSnd/TOs0APtFwjTjpdNG9YzeLI+VOpImTEvhFul27drpt7/9rVq0aBGZ9buoqEi33Xab+vTpo/fff9/uYgIA6lC9/nSm29DqHUEdqmX8dLLwOA11ae5QcYWlr7mQAWCjgoMhrSwMKscjndrFaXdxkCDdW5jyOKWN+0IqreRL3VRga5jevHmz3njjDU2aNEnDhw9XkyZN1L9/f/3yl7/Uvn37JIUvyKRw0N64caMuvPBCnXvuudq8ebONJQcA1MXrNvTIhV7JSp31p7s1N+UwpXtnV9A6DcA2M9f6JEmjernkdrIkVrpwOgz1bu2QZUlf8aVuSkhYmN67d69mzpypu+66Sz/5yU/UqlUrdevWTZdddpkefvhhzZ8/X6WlpbIs64ifpk2basyYMRoxYkTkvpkzZ6p///568cUXE1V8AEADtG9q//jpaLp6u53hsdNFFZbW707+1nQAjdO/1laPl3bbXBIkWnVXbyYhSw1x7zcyfvx4LVq0SFu2bDni/qNdUJmmqT59+mjYsGGRn549e0a2z507V5MnT9bSpUtVXFwcmQ387rvvjvfLAAA0UPX46VeW+tQiy9BxeTXPSvv0uKyowm88dGvu0OZ9Id0zu1wvXJEV6SEFAIlQGbA0+5twmB7N5GNp58jlsTz2FgZ1inuYfuONNyJjn/9b06ZNdfLJJ0eC89ChQ5WTk1PjsUaMGKHFixfrr3/9qyZOnKjKykrdd999GjFihM4666x4vgwAQANVj59+d5Vfa3YE1TTTVJOMxAbUG94orXfLd4bLUIc8U1v3h7T1QEidmrEkDYDEmb8xoJJKaUAHh9rlJtX0RkgAlsdKLQmZ0cCyLBmG8YNW5169ejXoeL/61a/UrVs3nXPOOZKkxx9/nDANAEnM6zY07cJM/eqtMi3dFtDp3ZxymEcP1MnSOr11f0gLNwcI0wASiiWx0hvLY6WWuH/ddeedd+qjjz7SgQMHtHr1av3973/Xz372swYH6WqjR4/WGWecIcuytHLlyhiVFgAQLx3zHJp8VoZKKq06J1aJx/jpaAJ6tsdQ6xxTLy31aW8JY6cBJM6/vg5PPsaSWOmpfa6hppmGdpdY2l1M/ZPsEhKmzz77bDVp0iTmx87Pz5ckFRYWxvzYAIDYO6WzUy2yDW3ZH1LBIfvWn66Pbi1MyZK+3BKwtRwA0sfGvUGt2x1SM6+hoZ1YEisdGYahvtXrTdPVO+ml9EAMtzs8w2EwyB8aAKQC0zT0xNgsuZ3SqsKASn21Lz8V60AdTet0M6+h3ExDz3xRyXqfABJixprvJx6raSgMGj9m9E4dKR2mL730Uk2aNInx0gCQQrI9hh4836tAUFq2LaBQKDmDqmEY6tbCoWBIWrKN1mkA8ff2qnAX74v7sSRWOsunZTplpHSYHjRokB544AF9/PHHdhcFABCFbi0c+u3wDB0st7R+T2K7e0fTOt22iaFMt6HHP6tQIJicoR9A47DjUEgLNgWU6ZJG92K8dDr7vps3X+Qmu5QO0wCA1HVmD6dyMw19uzeovaXJOX7aNAx1aWbKF6C7HYD4mrHGJ8uSxvR2KctDF+90dkLV8lhrdgaPurwwkgdhGgBgC4dp6LGLvXIY0ortQfnraPmNZaCOpnX6uDxTDlOa+mkFFzUA4mbGV+Hx0hf0pYt3umuWZapdrqGSSmnrAWb0TmaEaQCAbZpnmfrTqEyV+616jQ2zo4Xa5TB0XJ6pogpLm/ZxUQMg9korLc1Z75dpsCQWwvpWtU7XtZQk7EWYBgDYamBHh8af5Nb2gyEV1rFcVixF0zrdpblDMqSFLJMFIA4+WudXZUAa1tmpFtlcnkPq2za8NBqTkCU33q0AAFsZhqHz+7ojy2WV+xO7XFZ9ZLkNtW1i6tWlPu0sonUaQGy9uSI8i/fY/nTxRhhrTacGwjQAwHY5GYbu/YlX/qC0srDuCVdiFaijaZ3u0SJcZX6xmdZpALFT5rP0/lfhMH1Jf7p4IywSpunmndQI0wCApNCztUM/H+bRnuKQttRjwpVEt1DnZppqnmXo+S8rdaic1mkAsTFrrV+lvnAX7455DruLgyTRp3X4b+HrnUEFQ0x+mawI0wCApDG6t0uZbkNf7QyqpLLui4dYBOpoWqe7tXDIsmidBhA7H1S1Sl/Uj1ZpfC/LY6hrc1OVAWnDXr7ATVaEaQBA0vA4DT10fqZClrRse0ChJFuKqlW2oSYZhp5aUKkyX3KVDUDqCYUsffh1eEmsc/swXhpHYtx08iNMAwCSSqdmDt10eoYOlVv1+jY+ka3ThmGoe0uHgiFp8VZapwEcm8Vbg9pdYqlbC1O9WnNZjiMRppMf71oAQNI5o7tT2R5D63cHVVSRmO7e9dW2iaFMt6HHP6uQL0DrNICG++fqcBfv805wyTAMm0uDZFO91jSTkCUvwjQAIOk4HYamXehVSNKKgvp19z7WQF3f1mnTMNS9hSlfQFq+nQscAA1jWZbeXsWSWKjZCZGWaXpCJSvCNAAgKbXLNfXbqu7em/bVb/KVRLVQd2xqyu2Ups2rUIhZVgE0wNc7g/p2T0itsg0N6+y0uzhIQj1bOeQwpfV7QqqkJ1RSIkwDAJLWGd2dynIbWrc7qNJ6zO4tHVugrm/rtMM01LmZQ+U+S+t2M8sqgOi9uzo88diF+W45TLp444c8TkPHtzQVDEnf7KYnVDIiTAMAkpbLYej+8zIVCkmrdwRk1XN270S0UHfKM2Ua0v2flNe7XABQ7Z1VLImFuvVtG+61wCRkyYkwDQBIap2bO/TzYR7tKbG0/VD9W4HjHagzXIY65pk6UGZpYz27oQOAJG3ZH9Ty7UE1yTB0Zg/CNGoWmYSMMJ2UCNMAgKR3dk+X3E7pqx3BuI8bq29Xb0nq3sIhw5Dmb2RyGAD1915VF++f9HHJ7aSLN2rG8ljJjTANAEh6Xreh/xuTKX8wPGlPfcW7ddrrNtShqam3Vvi0eR8XOgDq559rwl28L8ynVRq1i4RplsdKSoRpAEBK6NPGofEnubX9YEh7S5Knu3ePFg7JkP69gdZpAHXbVxrSZxsCcjuk0b1ZEgu169bClMcpbdoXUkk9J+JE4hCmAQApwTAMndPHJYcprSoMKhjFklTRBupounpneQx1rGqd3nqAlgMAtfvwK7+CIems411qkkEXb9TOYRrqUzVuOpqeWUgMwjQAIGXkeU1NPDNDpb7oJ/16elxW3Fqpu1e1Tn9G6zSAOtDFG9FiErLkRZgGAKSUkzs5leUx9O2eoCr80Xd5i0egzvYYatfE1OvLfNpZxMzeAI6uzGdp1lq/DEM6vy9dvFE/1eOmv6JlOukQpgEAKcXpMHTX6EwFQ9K63Q27sIhHoO7WIlylfr6J1mkAR/fJer/K/eEvBds04TIc9fP9jN7UL8mGdzEAIOUc38rUT090a9uBkA6UNawluLZA3ZCw3TTTVItsQ9MXVepQOa3TAH7o3VV08Ub0mNE7eRGmAQApxzAMje7tkmGEu71ZVsNmOI31OOquzR2yLGnRVi54ABypMmDpn1XrS1/Ujy7eqL+OTU3leKTCQ5b2l/JlbTJx2l0AAAAaolWOqRtP9ejJ+ZUqLLLUPrfhs+LGKlC3yjaUk2Hor/+p0GldncpwMVMvgLCP1vl1sNzSwI4O9WjpsLs4SCGGYahvW6e+2BzQVzuDOq0b7aHJgv8JAEDKGt7dJZdDWrszuqWy4sUwDHVrbioYkpZvZ2wbgO+9tizcxXv8AFqlEb3vx03T8ymZEKYBACnL6zY06cwMlfstbYpyqax4aZdryu2UHplXmRQBH4D9SistvVe1JNalhGk0QGR5LMZNJxXCNAAgpQ0+zqkst6Fv9wZVGbA/vDpMQ12aO1Tut/TN7uQI+ADs9f5XPpX5pB91dapjHl28ET1appMTYfooiouLNWXKFOXn5ys7O1u5ubkaPHiwpk2bJp/PF9Pnuv/++2UYRuQHABAdp8PQHaMzFQhK3+5JjouM45qaMg3pgTnldhcFQBJ4fXn4+vGyk2iVRsNUh+nVOxo+6SZijwnI/suWLVt0xhlnaPPmzZIkr9eryspKLVmyREuWLNHLL7+sOXPmKC8v75if65tvvtFdd911zMcBgHTXu7Wp3ExDWw6E1K25pUy3vV9OZrgMtcs1tf1gSLuKQ2qdw3fXQLoqqbQ0a51fhiGN7U+YRsO0yjHVKtvQ7hJLBYcsdWhKI1wyoHY/TDAY1HnnnafNmzerbdu2mj17tkpLS1VWVqbXXntNOTk5Wr58ua644opjfq5QKKTrrrtOFRUVGjZsWAxKDwDpyzAMTRmdqVBI+iZJWqc7NQtXscu2MREZkM5mrvWrwi+d1tXJF2s4Jv3bh1unVxZQryQL3tGHmT59ulavXi1JevvttzVy5EhJkmmauvTSS/X0009LkmbOnKk5c+Yc03M98cQTWrBgga644gqNGjXq2AoOAFC3FqZ+eqJb2w6GVFxhfxe4vMzwMlnPfFEpf9D+8gCwxzsrw128L2ZtaRyj/u3CnYpXFibHl8YgTB/hhRdekCSNGDHiqK3F48ePV5cuXSRJL774YoOfZ9OmTbr99tvVvHlzPfLIIw0+DgDge4Zh6MzjXZKVHK3ThmHouDxT/qC0bpf95QGQeBV+Sx98HQ7TF/Vz2VwapLrvW6apU5IFYbpKWVmZFixYIEkaM2bMUfcxDEOjR4+WJH388ccNfq7rr79epaWlevjhh9WyZcsGHwcAcKQOTU1dPtCtHYdCOlRu/0zaHXLDE5Gt4MIHSEuzv/GrpFIa1NGh45jFG8eof7uqMF1IN+9kQZiusnbtWoVC4Quvvn371rhf9badO3dq//79UT/PM888ozlz5mjkyJG6+uqrG1ZYAECNzjzeJRnSuiRYlsrtNNQqx9QbK3wqSoKu5wAS640V4VbpcawtjRjo2cohl0P6dk9IZT7qlGTAbN5VCgsLI7fbt29f436HbyssLFSzZs3q/RwFBQWaNGmSMjMzI+OvY2nHjh117tOhQ4eYPy8AJJPWOaYmDPbohUWVOlAWUp7X3u+N2+ea2lkU0tc7gzq5c+Oodrdv317r9l27diWoJEDyqvBbem91OEz/lFm8EQNup6E+rR1aWRjUVzuDGnxc46hTUhn/A1WKi4sjt71eb437Hb7t8MfUxw033KBDhw7pgQceUNeuXaMvZB2GDBlS5z6sSwcgHQzv7tQLiyu1dldQp3SxN0y3zjHkMKUn51fo5M7ZtpYlVjp27Gh3EYCk99E6v4orpcHHOdS5OV28ERv924fD9MqCAGE6CdDNO0FeeuklffjhhzrxxBN1yy232F0cAGjUWmabum6oR/tKLe0tsbe7t8M01DrH1IEyi67eQBp5s7qL94m0SiN2vh83zVwcyYCvM6rk5OREbpeVldW43+HbDn9MbXbv3q2bb75ZDodDzzzzjJzO+Jz2RYsWqW3btnE5NgCkmuHdnXr+y0qt2x3UqVmGDMOwrSxtmpgqPBTS+t1BDWoELQnbtm2rdfuKFSt03nnnJag0QPKp8Ft6b01VF2/CNGKof/uq5bGY2DIppH6NHiPt2rWL3C4oKFC/fv2Oul9BQcFRH1Ob2267Tfv27dONN96oXr16qaSk5IjtPp8vcrt6m9vtltsd3Ydv27ZtGRMNAFXyvKauP8WjpxdUak+JpVY59oXpVtmGTENa10jCdF11TX3m8AAas/e/Cs/iPbSTQ52a0cUbsdOvqmV6VWFQlmXZ+kUx6OYd0bt3b5lm+HSsWbOmxv2qt7Vp06bek49t2rRJkvS3v/1NOTk5P/i57777IvtW3zd58uSGvhQAQJXTujplGtI3u4O2zhnhchhqlmXorRU+BYJ09QYau+mLKiVJVw/22FwSNDYts021bWLoUIWlrQfsX7Ui3RGmq3i9Xp166qmSpFmzZh11H8uy9NFHH0mSRo0albCyAQAaJjfT1C9O8ehguaW9pfaG2FbZpoIhadN+Ln6AxmzHoZBmrfXL7ZDGsyQW4qB/u6qu3oybth1h+jATJkyQJM2dO1dffvnlD7a/+eab2rhxoyRFtUb0vHnzZFlWjT933nlnZN/q+x599NFjezEAAEnSqV2cMpKgdbpldrjK/W4PFz9AY/by0kqFLOn8vi41y+JSG7HXv33VJGSMm7Yd7/DDTJgwQfn5+bIsS2PHjtWcOXMkSaFQSG+++aauv/56SdKYMWN01llnHfHY6dOnyzDCE9zMmzcv0UUHANSgqTfcOn2gzNKeEvvCdI5HynQZemGxr+6dAaQky7I0fVH4PX7NELp4Iz6+n9E7YHNJQJg+jNPp1IwZM9S5c2cVFBRo5MiRysrKUlZWlsaNG6eioiINGDBAL7/8st1FBQBE4bSu4dbpb/fY1zptGIZaZBsqrWSJLKCxWrY9qK92BtU6x9CPe7nsLg4aqepJyGiZth9h+r907txZq1at0h133KG+ffvKMAy5XC4NHDhQU6dO1cKFC5WXl2d3MQEAUcjNNHX9MI/229w63TwrPOvq5v1cAAGN0QtVE49dOcgjp4NZlhEfPVs55HFKG/aFVFLJl7N2IkwfRU5Oju666y6tXr1aJSUlKioq0pIlS3TrrbfWuFzVNddcExnvfMYZZ0T1fFOmTIk8FgAQH6d3c8o0pbW77GudblE1fnLDXiYhAxobX8DSq8vCXbyvHszEY4gfp8PQCW0csixpzQ6+nLUTYRoAkBZyM0396kcZKqqwVHjInjCd6TLUJMPQS0sq+QIVaGRmrfNrb6ml/u0c6tcu9deTR3KLjJsuYNy0nQjTAIC08aOuTjkd4Zm9QyF7wmzLbEO+gLTbxu7mAGLv/y2uXluaVmnEX//2LI+VDAjTAIC04XUbuuWMDJX6LG05YE9X6+ZVXb037aOrN9BYFFdY+uBrvwxDuuwkZvFG/EUmISNM24owDQBIKyd3dirDZWj9nqD8wcS3Djf3GjJNaf1uLoCAxmLGGp8q/NIZ3Z1qm8vlNeKvupv36sKAbT2tQJgGAKQZl8PQbWdlyBewZyIwp8NQiyxTb630qcLPBRDQGLy8NDzx2PgBdPFGYjTLMtWhqaniSmnzfno62YUwDQBIOyd1cCgnw9CGfUGV+xIfaFvnGLIsaSNdvYGUt7MopI/W+eV2SJf0J0wjcfrT1dt2hGkAQNoxTUN/PidToZD09a7EX4S0zA5Xv9/t5QIISHUvL61UyJLO7+tSsywurZE4/dszo7fdeMcDANJStxYOXTHIrcJDIe0tSWwLcZbbUJbb0EtLfCyRBaQwy7L0wqJwF+8JQ5h4DInVry0t03YjTAMA0taPe7lkmtKanYlfKqtltqFyn6V9pYRpIFWtKAhq9Y6gWmUb+nEvl93FQZqpXh5rFWHaNoRpAEDayvOa+t3wDBVXWNqQ4PHL1V297ZgEDUBsvLAovLb0FQPdcjkMm0uDdNOjpalMV3j+jaIKvpi1A2EaAJDWftTVqWxPeKmsksrEXYw0zzIkQ9q0nxYFIBX5g5ZeWRbu4n31YLp4I/EcpqG+bb9fIguJR5gGAKQ1p8PQ/eeFJyNbVRhI2Bhml8NQ0wxDbyz3sUYokIJmrfVrT4mlfu0cOrGD0+7iIE31bxf+22PctD0I0wCAtNepmUM3nOrRvlJLWw4krtt1i2xT/qC0s5gwDaSaFxaHu3hPGMxyWLBPv+rlsQoI03YgTAMAIOnsni5lug19vTOo0gR1927uDY+x3EJXbyCl7C8N6f01fjlM6YpBdPGGfaqXx1pFN29bEKYBAJDkcRq6/9xMBUPSioKAQgno7p3nDY+b3naQSciAVPL6cp98QWl0L5da53A5DftUL4+1ekfiV6UAYRoAgIiuLRz65Y882l9maWMCZtl2OQw18Rh6a4U/7s8FIHaqu3hfTRdv2Kyp11SnPFOlPiV8VQoQpgEAOMLIni5lewx9szuYkKVGcjMNlfstlftoUQBSwTe7gvpyS1BNMw2d35cwDftVd/Vm3HTiEaYBADiMy2Fo6gVeWZKWbw8oGOduc1nu8LjpfWWEaSAVVLdKXzrArQwXa0vDfpFJyBg3nXCEaQAA/kv7pqZuOSNDRRWW1u6K7zf93qowfaic7nlAsguGLP2/JeG1pZnFG8kisjwWLdMJR5gGAOAoTu/mVDOvoU37QtpVHL+g6wk3KKiMYdNA0pv3XUDbD4bUo6WpkzuztjSSw/czehOmE40wDQDAUZimocfHeuVyhGf3LvfHpxu2UdVLNETDNJD0XlhUPfGYR4ZBF28kh27NTWW5pS0HQjpYRmWSSIRpAABqkJtp6t5zvfIFpGXbAnFZdoSR0kBqKK6w9PaqcBfvqwbRxRvJwzQN5VctkbVqB63TiUSYBgCgFr1aO/Tr08LLZa3bHfuLlOoWaacj5ocGEENvr/SpzCeN6OFUp2a8YZFc+jFu2haEaQAA6jCyp0t5XkMb9oa0oyi2XeicVTWxL0AbNZDMXlwS7uI9YbDH5pIAPxRZHotx0wlFmAYAoA4O09ATY71yO8Pjp0sqYxd83c7wuMtSX8wOCSDGNu0Lau63AXnd0tj+dPFG8unfrnoSMpbHSiTCNAAA9ZCbaWrqBV4FQtLirQH5g7EJ1BlVEwIXVdAyDSSrZ74It0qPH+BWtoeJx5B8qrt5r9kRVDAO83vg6AjTAADUU5fmDv1hZIZKKi2tKAjKso79gsXpMORxGtrHDKxAUvIFLP1jYThM33BKhs2lAY4uJ8NQ1+amyv3St3uoTxKFMA0AQBSGdXbqmqEe7SwKaX2MLliyPNLMr/0xCecAYuu9NX7tLrF0YnuHBh/HxGNIXv2qunqvKKCrd6IQpgEAiIJhGDq/r0u5mYbW7w6q8NCxB+rcDEP+oLS/jDANJJunP6+QJN1wCmtLI7kN7Bju6r10G5OQJQphGgCAKLkchv56iVcep6EVBQEdOMYu2s284ep4y3665gHJ5Ls9Qc1ZH1CWW7p8ILN4I7kN6hhumV6yjZbpRCFMAwDQALmZph672CtJWrQ1oFJfw1uVm3nDrV1bDhCmgWTy96qJxy4f6FGTDFqlkdy+b5kOKMQkZAlBmAYAoIHaNzV1/3le+YLSoi2BBq8VneEylO0x9PISH+OmgSRRGbD0/JfVE4/RKo3k1zLbVKc8U8WVTEKWKIRpAACOQa/WDv3p7EyVVFpavDXQ4CVJ2jQxVRmwtJXWaSApvLvKp72llgZ2dERa/IBkN+g4unonEmEaAIBjNLSzU785PUP7yywt2x5UqAGty22bhLuQrt3FxDFAMnj6c1qlkXoGV33xs3grYToRCNMAAMTA2T2d+lnVklmrCqNfgzo3w1CW29A/FvoY6wbYbN2uoOZ9F1COR7rsJMI0Useg48JhegkzeicEYRoAgBgwDEMX5Ls0/iS3th0Iac3O6AK1YRhqm2uqwm9p20G6egN2+nvVclhXDvIo28PEY0gdJ3UId/NeXhBQIMgXs/FGmAYAIEYcpqFxA9xqkW1o877oA3WbnPBFOxPHAPap8Ft6YbFPEl28kXryvKa6tzBV5pPW7aZ1Ot4I0wAAxJDLYejvl2Z9H6h31D9QN800lOEy9I+FlXEuJYCavLXSp/1lloZ2cqh/eyYeQ+oZVDVueslWwnS8EaYBAIixIwL1/pBWFATrNQ7aMAy1yjFUUmlpfymt04Advp94LMPmkgANM7hqRu/FzOgdd4RpAADiwOUw9MylWbpsoFvbD4a0eFuwXuPXWmaFq+ZN+wnTQKJ9tSOg+RsDys0wdOkAt93FARokMgkZM3rHHWEaAIA4cTrCF+Q/H+bR7uKQFmwKqNxXe6DO84bHTe8oIkwDifb3L8Kt0lcNdsvrZuIxpKYB7Z0yDGllYVC+AJOQxRNhGgCAOHKYhs7v69KkszJUVGHps41+7SmpOShnOCWXQ9pbwgUQkEhlPksvMvEYGoGcDEO9WpmqDEhf7WTcdDwRpgEAiDPDMHR6N5ceusCrkCUt3BzQmh0BBY8yjtowDBmGRJQGEuuN5T4dLLd0Shen+rZl4jGktsgkZIybjivCNAAACdKrtUPPXZalPK+hTftCmvtdQIWHQkfM9h0KWQoEwy3UABLDsiz9ZX54ben/oVUajQAzeicGVTUAAAmU5zU1/fIsLdoa1ANzyrV0W0DZHkNtmpjKdkt7Si2FLKl9U77vBhLli80BLd0WVKvs8FrxQKpjRu/EIEwDAJBgpmno5M5OvXRltj7fHNDf5lfouz3ftx6MG+DW0E5U0UCiPP5ZeOKx/znVI4+TiceQ+vq3d8phSqsLg6rwW8pw8XcdD9TUAADYJMtj6OyeLp3Vw6ndJZYOllvK8Rhql2vIMLjwARJh+8GQ3lrpk8sh/Q9rS6OR8LoNndDGoVWFQa0qDGoIX9DGBX3IAACwmWmGu3n3au1Q+6YmQRpIoL8tqFAwJI070a22uVwao/EY1DHc1ZtJyOKHTwwAAACkpXKfpac/D3fx/u3ptEqjcfl+EjLCdLwQpgEAAJCWXlvu075SS0M7OegGi0Zn0HHhv+nF25jRO14I0wAAAEg7lmXp8c/Cy2HRKo3GqF87h1wO6eudQZVWWnU/AFEjTAMAACDtzN8Y0IqCoNo0MXRJf5bDQuPjcRrq186hkCWtKKCrdzwQpgEAAJB2qlulbzwlQ26Ww0IjFRk3TVfvuCBMAwAAIK1s3BvUO6v8cjukG0712F0cIG6Y0Tu+CNN1KC4u1pQpU5Sfn6/s7Gzl5uZq8ODBmjZtmnw+X9TH27dvn55//nldeeWV6tOnj7KysuTxeNShQwddeOGFevfdd+PwKgAAAFBt2rwKhSzpqsEetc7hchiNV3XL9GJm9I4Lpi2sxZYtW3TGGWdo8+bNkiSv16vKykotWbJES5Ys0csvv6w5c+YoLy+v3sds06aNAoHv/5gzMjLkcrlUUFCggoICvffeexozZozeeusteb3eWL8kAACAtLa7OKTnvqyUYUiTRjDxGBq3E9o6lOGSvtkdUlGFpSYZDGmIJb6Kq0EwGNR5552nzZs3q23btpo9e7ZKS0tVVlam1157TTk5OVq+fLmuuOKKqI4bCAQ0ZMgQPfnkk9qwYYPKy8tVUlKiTZs26brrrpMkzZw5UzfccEM8XhYAAEBae/yzClX4pYvyXerZ2mF3cYC4cjkMndgu/He+jK7eMUeYrsH06dO1evVqSdLbb7+tkSNHSpJM09Sll16qp59+WlI4+M6ZM6fex/3000/15Zdf6sYbb1TXrl0j93fu3FnPPvtsJES/9NJL2rZtW6xeDgAAQNorrrD01/mVkqTbzsq0uTRAYlSvN8246dgjTNfghRdekCSNGDFCw4YN+8H28ePHq0uXLpKkF198sd7HHTFiRK3bq1unJWnJkiX1Pi4AAABq98wXFTpYbumM7k4N6cRoR6QHZvSOH8L0UZSVlWnBggWSpDFjxhx1H8MwNHr0aEnSxx9/HLPnzsj4fuxOMMgfPAAAQCz4ApYenhdeDuu2sxgrjfQxuKpleuFmWqZjjTB9FGvXrlUoFJIk9e3bt8b9qrft3LlT+/fvj8lzz5s3L3I7Pz8/JscEAABIdy8v9angkKX+7Rz6cS+X3cUBEqZXK1N5XkNbDoS0/WDI7uI0KvRvOYrCwsLI7fbt29e43+HbCgsL1axZs2N63oMHD+q+++6TJJ122mnq2bNnVI/fsWNHnft06NChQWUDAKDa9u3ba92+a9euBJUEqJ9QyNKDc8olSZPPypBhMKMx0odpGjq1i1MffOXXgo1+XXoSa6vHCmH6KIqLiyO3a1ue6vBthz+mIUKhkK666irt2LFDHo9HTzzxRNTHGDJkSJ37WJbVkOIBABDRsWNHu4sAROWfq/1atzukzs1MjTvRbXdxgISrDtPzNwUI0zFEN+8kcdNNN+mDDz6QJD355JPq37+/zSUCAABIfZZl6d5Pwq3St52VIaeDVmmknx91Dbehzt/IuOlYomX6KHJyciK3y8rKatzv8G2HPyZaEydO1F/+8hdJ0iOPPKJrr722QcdZtGiR2rZt2+ByAABQH3Ut3bhixQqdd955CSoNULtP1ge0dFtQbZsYumYILXJIT4M6OuV2SKsKgyqqsNQkgy+VYoEwfRTt2rWL3C4oKFC/fv2Oul9BQcFRHxONyZMna9q0aZKkhx56SDfffHODjiNJbdu2ZUw0ACDu6qpr6jOHB5Ao984Ot0rfOiJDGS4CBNJThsvQ4OOcWrApoIWbAxrFJHwxQTfvo+jdu7dMM3xq1qxZU+N+1dvatGnToMnHJk2apIceekiS9OCDD2rixIkNKC0AAACO5vNNfs37LqA8r6EbTmE5LKS3U7tUd/X221ySxoMwfRRer1ennnqqJGnWrFlH3ceyLH300UeSpFGjRkX9HBMnTtTUqVMlhYP0pEmTGlhaAAAAHM3/fRReV/qm0zOU7aFVGumtetz0gk2Mm44VwnQNJkyYIEmaO3euvvzyyx9sf/PNN7Vx40ZJ0tVXXx3VsSdOnBjp2j116lSCNAAAQIwt3BzQR+v8ys0wdNPpjJUGTqlqmV64JSB/kBV+YoEwXYMJEyYoPz9flmVp7NixmjNnjqTwElZvvvmmrr/+eknSmDFjdNZZZx3x2OnTp8swDBmGoXnz5h2x7bbbbosE6Ycffli33npr/F8MAABAmrlrVnis9E3DPWrq5ZIXaJ5lqk8bh8p80oqCoN3FaRT4ZKmB0+nUjBkz1LlzZxUUFGjkyJHKyspSVlaWxo0bp6KiIg0YMEAvv/xyvY+5detWPfjgg5Ik0zT1wAMPqE2bNjX+VHcDBwAAQP19scmvWev8apJh6ObhjJUGqlWPm17AElkxQZiuRefOnbVq1Srdcccd6tu3rwzDkMvl0sCBAzV16lQtXLhQeXl59T5eKBQ64vauXbtq/SkpKYnHywIAAGjUplS1St883KM8WqWBiB9VT0K2iUnIYoGlseqQk5Oju+66S3fddVe9H3PNNdfommuu+cH9nTt3lmUxPgEAACBePt/k18ffBJSbYeh3tEoDR6iehGz+xoAsy5JhMDHfseCrOgAAADQad84Mt0r/7owMxkoD/6VLc1NtmxjaVWxpw95Q3Q9ArfiEAQAAQKMwf6Nfn6wPqGmmoZuHM4M38N8Mw9CPuroksURWLBCmAQAAkPIsy9IdVa3St5yRodxMLnOBo6mehGw+k5AdMz5lAAAAkPJmrPFr7rcBNc8y9FvWlQZqdFrVuOl53zEJ2bEiTAMAACClVQYs3fLPMknS3edk0ioN1KJ/e4fyvIa+2xvS1gOsN30s+KQBAABASnt0XoU27gupXzuHrh9GqzRQG4dpaET3cOv0nPV09T4WhGkAAACkrB2HQrp7dnis9KMXeeUwWeoHqMtZx4cnIZuznq7ex4IwDQAAgJR1+4dlKqmUxvZ3aUQPl93FAVLCyOow/a1flmXZXJrURZgGAABASlq8NaDnF/nkcUoPne+1uzhAyujR0lSHpqZ2Fln6eifjphuKMA0AAICUY1mWbnonPOnYrSMy1KW5w+YSAanDMAyd1YNx08eKMA0AAICU8+oyn77YHFDbJob+MDLT7uIAKeesw7p6o2EI0wAAAEgppZWWJs8It0o/cJ5X2R4mHQOiVR2m530XUCDIuOmGIEwDAAAgpTz4abkKDlkacpxDVwx0210cICW1yzXVu7WpogpLS7YxbrohCNMAAABIGd/uCeqBORWSpMcuzpLJUlhAg7FE1rEhTAMAACAlWJalX71VqsqA9POTPTq5s9PuIgEprXqJrNnfEKYbgjANAACAlPDGcp9mfxNQiyxD95/HpGPAsRrRwyWnKS3YFFBRBeOmo0WYBgAAQNI7VB7Szf8MTzo29QKvmmdxGQscqyYZhk7t4lQgJH1KV++o8SkEAACApHf7h+XaWWRpeDenrh7MpGNArIzuHe7qPXMtYTpahGkAAAAktfkb/XpyQaVcDunJn2bJMJh0DIiVMVVhetY6vyyLrt7RIEwDAAAgaZX7LF33aqksS/rT2Znq08Zhd5GARqVfO4faNjG09UBIa3eF7C5OSiFMAwAAIGn98cMyrd8TUr92Dv1+ZIbdxQEaHcMwNLpXVev0Wp/NpUkthGkAAAAkpbnf+vXov8Pdu1+4PEtuJ927gXgY0yc8DwHjpqNDmAYAAEDSOVQe0jWvlEqSpozO1IkdWFMaiJeRxztlGtJnGwIqrWTcdH0RpgEAAJB0bn63TFsPhHRyJ4cmn0n3biCe8rymhnV2yheU5nxL63R9EaYBAACQVF5fVqnpi3zyuqUXr8yW00H3biDezj0hPG76/TWE6foiTAMAACBprN0Z1HWvhbt3P3qRVz1aMns3kAjnVYXpD772KRSiq3d9EKYBAACQFEoqLY19vlilPunqwW79/GSP3UUC0kafNg51aW5qZ5GlZduDdhcnJRCmAQAAYDvLsnT9a6Vauyuk/LYO/e2nWTIMuncDiWIYhs7tU9XV+yuWyKoPwjQAAABs95f/VOq15T41yTD09rXZ8roJ0kCind83vETWu6sYN10fhGkAAADY6otNft36XpkkafrlWYyTBmwyvLtTzbyGVu8I6ptddPWuC2EaAAAAttlTEtK4F0rlD0oTR2Toon5uu4sEpC2Xw4i8B99aSVfvuhCmAQAAYItgyNLlL5Zo+8GQTu/m1H3nZtpdJCDt/bR/eNz0mysI03UhTAMAAMAWU2aV65P1AbXOMfTa1awnDSSDM493qZnX0MrCoL7dQ1fv2hCmAQAAkHAffuXT3R9XyGFKb1yTrba5XJYCycDlMHRhPq3T9cGnFgAAABLq651BXf7/SiVJ952bqdO7uWwuEYDD/fTE8LjpV5f5ZFmWzaVJXoRpAAAAJMzekpDOe6ZYRRWWxg9wa+KIDLuLBOC/jDzepVbZhtbsCGplAV29a0KYBgAAQEL4ApbGPl+ijftCGnKcQ89dliXDYJw0kGycDkOXDwy3Tr+4uNLm0iQvwjQAAADizrIs/c8bpfpsQ0Admpr653U5ynQTpIFkddUgjyTplWU+BYJ09T4awjQAAADi7oE5FXp+kU9etzTj50w4BiS7AR0c6tPGoV3Flj5Z77e7OEmJTzEAAADE1StLK/WHD8plGNKrV2drQAen3UUCUAfDMHTVoHBX7/+3hFm9j4YwDQAAgLh5bmGlrnwpPHP3wxd4dX5ft80lAlBfVwx0yzCkd1f5VFRBV+//RpgGAABAXLy1wqfrXy+VZUkPnp+pm89g5m4glXTMc+jMHk6V+8M9THAkwjQAAABi7tWllbr0hRKFLOnuczI16cxMu4sEoAFuOCX8JdhTCypZc/q/EKYBAAAQUx9+5dNVL5cqZEn/NyZTfzybFmkgVV2Y71KbJoZWFga1cHPA7uIkFcI0AAAAYuaDr3y6ZHqJgiHpf0dl6H9/nMla0kAKczkMXTc0vEzWU5/T1ftwhGkAAADExIuLKnXRP0pU4ZduOt2ju8bQtRtoDK4f5pFhSK8v92lfacju4iQNwjQAAACOSShk6ffvl2nCK6UKhKQ/jcrQIxd5aZEGGolOzRz6SR+XKgPS9EW0TlcjTAMAAKDBSiotXfxciR6YUyGXQ3p2fJb+fA5BGmhsfnlquKv3459Vyh9kIjKJMA0AAIAG2rI/qFMfK9J7a/xqnmXok1/m6LqTPXYXC0Ac/LiXSye0cWjrgZBeX+6zuzhJgTANAACAqH201qeTphZpVWFQfdo4tOh3TXR6N5fdxQIQJ6ZpaPKZ4Zn5H5xTwTJZIkwDAAAgCqGQpf+bVa4xfy/R/jJLF+a79PlNOerawmF30QDE2WUD3erY1NTqHUHNXOu3uzi2I0wDAACgXvaXhnTuMyW6c1a5DEn3n5upd67NVm4ml5RAOnA5DN1yRrh1+oE5FTaXxn588gEAAKBOy7YFNHBakWau9atltqHZN+botpGsIQ2km58P86iZ19BnGwL6dH16t04TpgEAAFCjYMjSQ5+Wa9ijRdq8P6STOzm0bGKuzjye8dFAOsr2GLrtrHDr9G3vl6X12GnCNAAAAI5qy/6gzvxrsSbPKJcvKP32dI/+/Zsm6tCUS0ggnf3mtAy1zzW0ZFtQb61M39ZpPgkBAABwhGDI0mP/rlDfBw7psw0Btcs19NH/5Oixi7PkdtKtG0h3mW5D/zfGK0n64wdl8gXSs3WaMA0AAICIpdsCGvpIkW5+t0wlldL4AW6tnpyrUb3o1g3gexOGuNW3rUPf7Q3poU/TczIywjQAAABUUmnpd++WasjDRVq6LajOzUz96xfZenVCtpplcckI4EgO09CTl4Rbp//8cbm+2xO0uUSJxycjAABAGvMFLD06r0K97zukR/9dKcOQJp+Zoa9+n6sxfdx2Fw9AEjutm0vXDnWrMiD98q3StJuMzGl3AQAAAGCPj9f59dt3SvXN7pAkaWgnh/5+aZb6teMSEUD9PHi+VzPW+DX7m4BeXurTlYM8dhcpYWiZBgAASDMb9wZ10T+K9eOnivXN7pB6tzY14+fZ+vymJgRpAFFpnmVq2gXh7t6/ebtM2w6kT3dvwjQAAEAasCxLK7YHdNVLJTr+3kP652q/cjzStAsytXJyrs7r65ZpMlM3gOhdNditi/q5dLDc0tUvlyoYSo/u3nz1CAAA0IgdKAvp+S8r9df5ldq4L9yd22lK153s0d3nZKpNE9pWABwbwzD0zKVZWrj5kOZ9F9D9n1To9lGZdhcr7gjTAAAAjdCaHQE98VmlXlpaqTJf+L7WOYYuO8mtW87IUMc8h70FBNCoNM8y9eIV2Rr1VLH+d2a5+rd36NwTGvckhoRpAACARmJfaUjvrPLp5SU+/XtDIHL/mN4u/fo0j0b3ctGVG0DcjOzp0v3nZuq298t1+YslWvi7XPVp03i/uCNMAwAApLCiCkvvrvLp1WU+fbLer2C4J7eaZBj62RC3fnVahnq0bLwXswCSy6QzM7SyMKhXlvo0+qlifX5zE3Vo2jiHkxCmAQAAUszOopBmrvXr/a98+tfXflVWNUK7HNKYE1wad6JbF/VzK9tDKzSAxDIMQ89emqVtB0L6z8aARv2tSJ/8sona5Ta+QE2YBgAASHIVfksLNgX08Tq/Zq/3a/n275eeMQ1p5PFOXXaSRxf1cynP2/guWAGklky3oRk/z9bwvxRrVWFQpz1epNk35qhri8bVS4ZPW6S97du3yzAMGYah7du3212cpMV5qhvnqH44T3XjHKGk0tK/v/PrvtnlOvvJIuX98YBGPlmsBz+tCAdpf6nO7hbQ0+O82j6lqWb/somuPdlDkBbvn7pwfurGOapdfc9PU6+pub/K0dBODm3cF9KPHi/Ssm2BGvdPRbRMAwAA2KiowtKaHQGtKAhq8daAFm8N6utdQVmHLdNqGNLAjg6d3K5Cf500Vtq5WM9t3qAOHTLsKzgA1KFZlqlPftlEF/2jWJ+sD+iUx4r0+MVeXT/MI8NI/WEohGkAAIAE8AUsfbM7qDU7glpd/VMY1JYDoR/s63ZIJ7Z3aGgnp0b0cGl4N6eaZZnavr1Yfy1YYEPpAaBhsj2GPvhFjia+V6a//KdSN7xRpn9vCOgvY70p35uGMA0AABAjwZClbQdC+m5vSBv2BsP/7gtq/e6QvtkdVOCHuVmZLumENg71a+fQoI5ODenkVH5bh9zO1G+1AQBJ8jgNPTE2Sz/q4tTPXy/VK0t9mvutX/ed69WVg9xypOiSfYRpAACAerIsS3tLLW3ZH9LWAyFtORDSpn3BSHjetD8kf/DojzUNqWcrU/ltHcpv61R+O4fy2zrUpbmZsheSABCNS0/yaNBxTl37aqk+2xDQNa+U6oE55frzOV5dlO+SmWKfhYRpAACAKr6Ape0Hvw/KWw9U3w5Gbpf7a368YUjH5Znq1sJU9xYOdWtuqntLh7q3MNWzlUMZrtS6UASAWOvWwqG5v8rR68t9umNmudbuCumS50vUrYWp/znFo/EneVJmXWrCNAAAaNRCoXBr8s7ikHYVW9pZFNLO4pB2FlnaVXzk7b2lVp3Ha5FlqFMzU8flmeqUZ6pTnkPdW4bDc+dmJoEZAOpgmoYuG+jRJSe6NX1Rpe7/pEIb9oY0aUa5Js0o1+DjHLow363RvVzq184hpyM5P1cJ0wAAIGVYlqVSn3SgLKQD5ZYOlFX9lFs6UBbS7pKqgFxUHZ7D9wWPMlb5aJym1DHP1HFNzUhgPq4qMFff9rqT86IOAFKNy2Ho+mEZum6oRx+t82v6Ip/+tdanxVuDWry1XLd/WC6vWxrc0amhnZzq2cqhHi1N9WjpUOscw/YZwQnTKS4Q+H6tth07dthYktR1+HnjHNaM81Q3zlH9cJ7qlurnqLCwMHL7lc+2KKdZpfxByR+S/AEr/G/QUqD6vqDkC0rlfktllZbK/FKZ31KZzwrf55dKKy2V+S2VVIYfFxVDau411DLbVKvs8L/hH1XdZ6pFlqFWOaZyM4yax+z5pf27pf0NPzXHLNX/NhKBc1Q7zk/dOEe1i9f5yW8iTRsp3TPc0oKNfn30TUCLtvi1YW9I/94n/XvFkftnuKUW3vDnd7MsU829hvK8prxuKdMpZboNZboMZbqkioM7I487PD8dK8OyrLr7MyFpLV68WEOGDLG7GAAAAACQ9BYtWqTBgwfH5FipMbIbAAAAAIAkQst0iquoqNDq1aslSS1btpTTSc99AIC9SkpKtHDhQknSsGHDlJWVZXOJAADpLBAIaM+ePZKk/Px8ZWRkxOS4hGkAAAAAAKJEN28AAAAAAKJEmAYAAAAAIEqEaQAAAAAAokSYBgAAAAAgSoRpAAAAAACiRJgGAAAAACBKhGkAAAAAAKJEmAYAAAAAIEqEaTQ6xcXFmjJlivLz85Wdna3c3FwNHjxY06ZNk8/ni/p4+/bt0/PPP68rr7xSffr0UVZWljwejzp06KALL7xQ7777bhxeRXzF+hzV5v7775dhGJGfVBHvc1RUVKQHHnhAp5xyilq2bBn5mxoxYoSmTJmigwcPHvuLSIB4nqfZs2dr3Lhx6tSpkzIyMpSZmamuXbvqiiuu0L///e8YvYL4KSsr08yZM3X33Xfr4osvVqdOnSLvgylTpsTkORL5XkZ8UGfVjvqqdtRVdaOeOjrqqBixgEZk8+bNVufOnS1JliTL6/VaHo8n8vuAAQOs/fv3R3VMp9MZebwkKyMjw8rKyjrivjFjxlilpaVxelWxFY9zVJN169ZZGRkZR5yrVBDvc/Tpp59arVu3jhzP6XRaTZs2PeI8LV++PHYvKE7idZ5CoZB1ww03/OB9l5mZecR9v/vd7+LwqmJn7ty5R5T38J8777zzmI+fyPcy4oM6q3bUV7Wjrqob9VTNqKNiI/k/KYB6CgQCVn5+viXJatu2rTV79mzLsiwrGAxar732mpWTkxO5iIiGJGvIkCHWk08+aW3YsCFy/6ZNm6zrrrsu8qFw5ZVXxvT1xEO8ztHRBINB69RTT7UkWcOGDUuZi5N4n6P58+dHKtuRI0da8+fPt4LBoGVZllVWVmYtWbLEuv32262NGzfG7DXFQzzP03PPPRf5e7nkkkus9evXR7atW7fOuuCCCyLb33nnnZi9plibO3eulZeXZ5111lnWpEmTrFdffdVq06ZNTC5UEvleRnxQZ9WO+qp21FV1o56qHXVUbCT3JwUQhWeffTbywfX555//YPsrr7wS2f7JJ5/U+7iffvpprdsP/2Zy69atUZc7keJ1jo7m0UcftSRZV1xxhXXnnXemxMWJZcX3HJWWllpdu3a1JFljx46NXJikoniepzPOOMOSZHXv3t3y+/0/2O7z+SLncfz48Q1+DfEWCAR+cF+nTp1icqGSyPcy4oM6q3bUV7Wjrqob9VTtqKNiI7k/KYAonHbaaZYka8SIEUfdHgqFrC5duliSrKuvvjpmz7to0aKk//axWqLO0caNG62srCyrefPm1u7du1Pm4sSy4nuOnnrqKUuSlZmZae3ZsycWxbVNPM9Tz549IxdxNbn44ostSda5554b1bHtFqsLFbs+7xA71Fm1o76qHXVV3ainokcdFT0mIEOjUFZWpgULFkiSxowZc9R9DMPQ6NGjJUkff/xxzJ47IyMjcjsYDMbsuLGWyHN0/fXXq7S0VA8//LBatmzZ4OMkWrzP0YsvvihJuuCCC9SiRYtjKKm94n2eunbtKklauXKlAoHAD7b7/X6tWLFCkjRo0KCojt0Y2Pl5h9igzqod9VXtqKvqRj1ln3SrowjTaBTWrl2rUCgkSerbt2+N+1Vv27lzp/bv3x+T5543b17kdn5+fkyOGQ+JOkfPPPOM5syZo5EjR+rqq69uWGFtEs9zVFlZqSVLlkiShg8fro0bN+q6665Thw4d5PF41KZNG11wwQWaOXPmMb6K+Iv339KNN94oSfruu+902WWX6bvvvots++abbzRu3Dht3LhR3bp10+9+97uGvISUZufnHWKDOqt21Fe1o66qG/WUfdKtjiJMo1EoLCyM3G7fvn2N+x2+7fDHNNTBgwd13333SZJOO+009ezZ85iPGS+JOEcFBQWaNGmSMjMz9fTTT0dfSJvF8xxt3rw5sgzE9u3b1a9fPz333HPas2ePvF6vdu3apRkzZuicc86JVNLJKt5/S+edd54eeeQRud1uvfXWW+rRo4e8Xq+8Xq969eqlefPm6cYbb9SiRYvUpEmThr2IFGbX5x1ihzqrdtRXtaOuqhv1lH3SrY4iTKNRKC4ujtz2er017nf4tsMf0xChUEhXXXWVduzYIY/HoyeeeOKYjhdviThHN9xwgw4dOqQpU6ZEukClknieowMHDkRu33fffXK5XHr11VdVUlKiAwcOaOvWrRo/frwk6amnntJjjz0WbfETJhF/SzfffLPeeecdtWrVSpJUXl6u8vJySeGWk+LiYh06dCiqYzYWdnzeIbaos2pHfVU76qq6UU/ZJ93qKMI00EA33XSTPvjgA0nSk08+qf79+9tcInu99NJL+vDDD3XiiSfqlltusbs4Sae6y1P17aeeekrjx4+Xy+WSJHXs2FEvv/yyBgwYIEm6++67jzoOKx2UlZXp0ksv1bnnnqvjjjtOH3/8sfbu3as9e/bo448/1gknnKCXXnpJQ4YM0apVq+wuLpASqLO+R31VM+qq+qGeQjXCNBqFnJycyO2ysrIa9zt82+GPidbEiRP1l7/8RZL0yCOP6Nprr23wsRIlnudo9+7duvnmm+VwOPTMM8/I6XQ2vKA2iuc5Ony/jh076tJLL/3BPqZp6tZbb5Uk7d27V0uXLq3XsRMt3u+3SZMm6Y033tDxxx+vzz77TGeffbaaN2+uFi1a6Oyzz9Znn32m448/Xnv37tWvfvWrhr2IFJbozzvEHnVW7aivakddVTfqKfukWx1FmEaj0K5du8jtgoKCGvc7fNvhj4nG5MmTNW3aNEnSQw89pJtvvrlBx0m0eJ6j2267Tfv27dMvfvEL9erVSyUlJUf8VI+/knTU+5JFPM/R4WODevXqVeN+vXv3jtzesmVLvY6daPE8T8XFxfr73/8uSfr1r3+tzMzMH+yTmZmpX//615Kk+fPna/fu3fU6dmORyM87xAd1Vu2or2pHXVU36in7pFsdRZhGo9C7d2+ZZvjPec2aNTXuV72tTZs2atasWdTPM2nSJD300EOSpAcffFATJ05sQGntEc9ztGnTJknS3/72N+Xk5Pzgp3rCG0mR+yZPntzQlxI38TxHzZo1i1ykGIZR436WZUVu17afneJ5ntavXx/pMtitW7ca9+vRo0fkdvXfX7pI1Ocd4oc6q3bUV7Wjrqob9ZR90q2OIkyjUfB6vTr11FMlSbNmzTrqPpZl6aOPPpIkjRo1KurnmDhxoqZOnSopfFEyadKkBpbWHok4R6ku3ueoev+1a9cecSFyuLVr10Zud+nSJarjJ0o8z1N1BSzV3tqxa9euyO1U7h7WELyXUx91Vu34G68ddVXdqKfsk3bvXwtoJJ599llLkmUYhrVw4cIfbH/99dctSZYk65NPPonq2LfeemvksVOnTo1VkRMunueoNnfeeWfkuMkunufos88+izz21Vdf/cH2YDBoDRgwwJJktW/f3goGgw1+HfEWr/NUVlZmZWZmWpKsk046yfL7/T/YJxAIWKeccoolycrLy7MCgcAxvZZE6tSpkyXJuvPOO4/pOHa9lxE71Fm1o76qHXVV3ainokcdFb3k/qQAouD3+638/PzIh3v1mzMYDFpvvPGG1aRJE0uSNWbMmB889vnnn4+8qefOnXvEtsmTJ0e2Pfzww4l4KXETr3NUl1S5OLGs+J+jSy65xJJkNW3a1Hrttdcsn89nWZZlbd261Ro/fnzk8dOnT4/ba4yFeJ6n3/zmN5Hto0ePtlatWmUFg0ErGAxaK1eutEaNGhXZftddd8X7pR6T/fv3W3v27In8dOzY0ZJkTZo06Yj7i4uLj3hcXefoWM4/kgN1Vu2or2pHXVU36qm6UUcdu+T+pACitGnTJqtz586RN7jX67UyMjIivw8YMMDav3//Dx5X04fCli1bIvebpmm1bt261p+HHnooga+2YWJ9juojVS5OqsXzHJWUlFinn356ZD+Px2Pl5eVFfpdk3XHHHXF+hbERr/NUVlZmjR49+ohz4vF4LI/Hc8R9l112WdJ/21/9LX9dPxMmTDjicfX5W2ro+UfyoM6qHfVV7air6kY9VTvqqGPHmGk0Kp07d9aqVat0xx13qG/fvjIMQy6XSwMHDtTUqVO1cOFC5eXl1ft4/73e4q5du2r9KSkpicfLiqlYn6PGKJ7nKCsrS3PnztUzzzyj008/XVlZWSopKVH79u01fvx4LViwQHfddVeMX1F8xOs8ZWZm6l//+pfefPNNXXDBBerQoUNk3F7Hjh01duxYffDBB3rllVfkcDhi/bJSBu/l1EedVTv+xmtHXVU36in7pMv717Cq/+cBAAAAAEC90DINAAAAAECUCNMAAAAAAESJMA0AAAAAQJQI0wAAAAAARIkwDQAAAABAlAjTAAAAAABEiTANAAAAAECUCNMAAAAAAESJMA0AAAAAQJQI0wAAAAAARIkwDQAAAABAlAjTAAAAAABEiTANAAAAAECUCNMAAAAAAESJMA0AAAAAQJQI0wAAAAAARIkwDQAAAABAlAjTAAAAAABEiTANAAAAAECUCNMAAAAAAESJMA0AAAAAQJQI0wAAAAAARIkwDQAAAABAlAjTAAAAAABEiTANAAAAAECUCNMAAAAAAESJMA0AAAAAQJQI0wAAAAAARIkwDQAAAABAlAjTANLOM888I8MwZBiGcnNztX///lr3//bbb9WyZcvIYyZOnJigkgIAGrspU6ZE6pdx48bV+3Fvv/125HEDBgxQKBSKYykBHA1hGkDaufbaa9WzZ09JUlFRkR566KEa9921a5dGjx6tvXv3SpIuv/zyWvcHACAap5xySuT2okWL6vWYiooKTZo0KfL7o48+KtPksh5INN51ANKOw+HQn//858jvTzzxhPbs2fOD/UpKSvSTn/xEGzdulCSNHDlS06dPl2EYCSsrAKBxGzp0aKRe2bJli3bv3l3nY6ZNm6ZNmzZJki655BINHz48rmUEcHSEaQBp6ZJLLtGgQYMkSaWlpXrwwQeP2B4IBPTTn/5US5culSQNGDBA77zzjlwuV8LLCgBovHJzc9WnT5/I74sXL651/4KCAt13332SpIyMDHpLATYiTANIS4Zh6N577438/te//lU7d+6M/P6LX/xCs2bNkiR16dJFM2fOVE5OTsLLCQBo/IYNGxa5XVdX79tuu02lpaWSpFtuuUWdO3eOZ9EA1IIwDSBtnX322TrzzDMlSeXl5br//vslSf/7v/+r559/XpLUsmVLffTRR2rdurVt5QQANG71DdNffPGFXnnlFUlS27Zt9Yc//CHuZQNQM6fdBQAAO913330aOnSoJOnpp59WXl6e7r77bklSVlaWPvzwQ/Xo0cPOIgIAGrnDw3RN3bwty9JNN90ky7IkSffff7+ys7MTUj4AR2dY1e9IAEhTF198sd59990j7nM6nZoxY4bGjBljU6kAAOnCsiw1a9ZMBw8elCRt2LBBXbt2PWKf559/Xtdee60kafDgwfryyy+ZEBOwGd28AaS9e+65Rw6H44j7nn32WYI0ACAhDMPQySefHPn9v7t6FxcX649//GNk38cee4wgDSQBwjSAtLds2TKFQqHI7yeeeKImTJhgY4kAAOmmtq7ef/7znyOTZF5++eVH7AvAPoRpAGlt9uzZ+tnPfqbDR7ysWLFCn3zyiY2lAgCkm5omIfvuu+/02GOPSZK8Xm9kskwA9iNMA0hby5cv19ixY+X3+yVJJ510UmQbM6QCABJp6NChMs3wpfmyZcsUDAYlhZe/8vl8ksLLYnXo0MG2MgI4EmEaQFratGmTxowZo+LiYknSrbfeqk8++US5ubmSpCVLlujtt9+2s4gAgDTSpEkT9enTR5JUVlamr776Sh9//LHef/99SVLHjh01adIkO4sI4L8QpgGknb179+rHP/6xdu3aJUkaP368HnroIeXl5emWW26J7PenP/0p0jIAAEC8Hd7V+/PPP9fNN98c+f3BBx9UZmamDaUCUBOWxgKQVkpLS3XmmWdGxqONGDFCs2bNktvtlhSeMbVLly7at2+fJOkf//hHZCkSAADi6fDlr1q0aKG9e/dKkk499VTNnz/fzqIBOApapgGkjUAgoHHjxkWCdH5+vt59991IkJaknJycI7rRTZkyRZWVlQkvKwAg/RzeMl0dpKuXwgKQfAjTANLGL37xC/3rX/+SFB57NnPmzMgY6cP9+te/VqtWrSRJ27Zt05NPPpnQcgIA0lPPnj2Vl5d3xH3XXHONBg4caFOJANSGMA0gLfzpT3/S888/L0nKy8vTrFmz1L59+6Pum5WVpd///veR3++9997IRGUAAMSLYRg6+eSTI7/n5OTo3nvvtbFEAGpDmAbQ6P3tb3/TPffcI0nyeDx67733IjOm1uTGG29Uu3btJIW72k2bNi3u5QQAoKKiInL79ttvV5s2bWwsDYDaMAEZAAAAkATmzp2rM888U5LUtWtXff311/J4PDaXCkBNaJkGAAAAbGZZliZPnhz5fdq0aQRpIMkRpgEAAACbPfnkk1qyZIkkafTo0brwwgvtLRCAOtHNGwAAALDRrFmzdOGFF6qyslLZ2dlauXKlunbtanexANTBaXcBAAAAgHSyePFivf766/L5fFq9erXmzZsX2fb4448TpIEUQZgGAAAAEuitt9466ioRv//97/Wzn/3MhhIBaAjCNAAAAJBAy5YtkyQ5HA61a9dO/fv3129+8xuNGjXK5pIBiAZjpgEAAAAAiBKzeQMAAAAAECXCNAAAAAAAUSJMAwAAAAAQJcI0AAAAAABRIkwDAAAAABAlwjQAAAAAAFEiTAMAAAAAECXCNAAAAAAAUSJMAwAAAAAQJcI0AAAAAABRIkwDAAAAABAlwjQAAAAAAFEiTAMAAAAAECXCNAAAAAAAUSJMAwAAAAAQJcI0AAAAAABRIkwDAAAAABAlwjQAAAAAAFEiTAMAAAAAECXCNAAAAAAAUSJMAwAAAAAQpf8Pbu/HtOkYYNUAAAAASUVORK5CYII=", "text/plain": [ "
" ] @@ -3241,7 +3532,7 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9MAAAPTCAYAAACzDVpFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAB7CAAAewgFu0HU+AADSXklEQVR4nOzdd3hc5Zn+8fs9Z2bULctyk3vvxhgXDMZgg+k4DUIIJZBKSEg2jZTNbwNks0sCpOymEBKylNBCC6GDbWyMMWDcsA24995tSSNNO+f3x2hk2bho7Jk5U76f69KFsKTRa1kzc+55n+d5jeu6rgAAAAAAQKtZXi8AAAAAAIBcQ5gGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkuTzegE4OY2NjVqyZIkkqUOHDvL5+CcFAHgrGo1q586dkqThw4eruLjY4xUBAJB6JK8ct2TJEo0dO9brZQAAcERz587VmDFjvF4GAAApR5k3AAAAAABJYmc6x3Xo0KH5/blz56qmpsbD1QAAIG3durW5aqrl8xQAAPmEMJ3jWvZI19TUqFu3bh6uBgCAQzHLAwCQryjzBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJLk83oBAOAV13W1eEtML3wQUSTmqkeVrYsH+1VTyeuMAAAAODbCNICC47qunl0S0U9fDOqj7c4hH7Mt6eLBfp3e06frRgfUs53t0SoBAACQzQjTAArKnnpH33gqqH8sDEuSSvzSp08JqE+1pQ+3xfT8BxG90PT281cb9KsppfrOOUUyxni8cgAAAGQTwjSAgvGn2Y36wb+CaohIFUXSXZ8o1ZfHFclnHwzKO2odPfV+WLNWR/WPhWF979mgZqyM6P6ry1RdRvk3AAAA4rgyBJD3XNfVLf8K6ptPxYP0BQN9WnRLpW4cX3xIkJakjhWWvnFWsR6/vlzPf7Vc7UqNnv8gotN/e0Crd8U8+hsAAAAg2xCmAeS938xs1N0zGhWwpfs/X6ZXb2qjPu2P3wt92dCA3v9hpcb0sLV6l6Mzf3dACzZGM7BiAAAAZDvCNIC89uDckH7wrwZJ0sPXleuG04uS+vpubS29/s02uniwXzvqXJ1/T62Wb2eHGgAAoNARpgHkrfV7YvrW0/WSpD9dUarPnho4odspLzL611fK9anhfu0JuvrM/XUKR91ULhUAAAA5hjANIC+5rquvPF6v2pB05akBfX18cjvSh/PbRo9eV65BHeNTv385rTFFKwUAAEAuIkwDyEtPvx/RtBVRtS8z+tNnS1NytFVJwOgvnyuTJP38tQa9uTpy0rcJAACA3ESYBpB3gmFX3/9XUJJ0x2UlKT3SakJfv/79/GLFHOlzD9ZpZ52TstsGAABA7iBMA8g7d05v0Ia9jkZ3t/WlJAeOtcbPLy7RpP4+bT0QP3ILAAAAhYcwDSCvrNsd069ej/cz//7yMlnWyZd3H862jP52VZmK/dKD74Up9wYAAChAhGkAeeX7/wqqMSJdPyagcb18afs+vatt/b/zSyRJ33gqqEiM6d4AAACFhDANIG9MWx7RM4sjqiiSfjmlNO3f7wfnFmtAB0tLt8b0P28w3RsAAKCQEKYB5IVozNV3/hnvX771ohJ1bpP+h7cin9Efr4hP977tlQZt3scwMgAAgEJBmAaQFx56L6wPtsXUr72lb00oztj3nTzQrytPDag+LN36SkPGvi8AAAC8RZgGkPOiMVe3vxoPsv99WakCvtQPHTuWOy4rkd+W7n83pA+3xTL6vQEAAOANwjSAnPfCBxFt2OtoUEdLV4zwZ/z792lv6+tnFslxpZ++yFFZAAAAhYAwDSDn/emt+PCvb5xVLGMyuyud8P8uKFF5kfTskojeXstRWQAAAPmOMA0gp63YEdPU5VGVBaQvjAl4to6OFZa+PzHeq/3vL9I7DQAAkO8I0wBy2j1Nu9LXjS5SZYm3D2nfn1SiqlKjmauimrs+6ulaAAAAkF6EaQA5qzHi6oG5YUnSTWcVebwaqaLY6Kbx8XX8egbnTgMAAOQzwjSAnPXskrD2Nbg6vaetU7r4vF6OJOnms4oVsKWn3g9r7W4mewMAAOQrwjSAnHX/uyFJ0hfHer8rnVBTaena0QE5rvS/s9idBgAAyFeEaQA5acPemKauiKrYL111mneDx47ku02DyO57J6T9DY7HqwEAAEA6EKYB5KSH3gvLdaXLTwl4PnjscMNqfLpgoE91oXigBgAAQP7JritQAGgFx3EPlnifnj0l3i19r2l3+n9nhRSNuR6vBgAAAKlGmAaQc95cE9Wa3Y56Vlma1C87Bo8d7oJBfg3pbGvDXkfPLA57vRwAAACkGGEaQM5J7EpfPzYgyzIer+bIjDHNu9O/ntEo12V3GgAAIJ8QpgHklNpGV0++H9/pvSGLpngfyTWjAupQbjR3Q0xvr4t6vRwAAACkEGEaQE55clFYwbA0qb9Pvattr5dzTMV+o6+fGQ/8985hEBkAAEA+IUwDyCn3z82+s6WP5cvjimRM/EWAfUGOyQIAAMgXhGkAOWPj3phmr4mqLCB95pTsOlv6aHq2s3XBQL8aItJjCxhEBgAAkC8I0wByxlPvRyRJU4YGVFaUnYPHjuQr4+K76H/lzGkAAIC8QZgGkDOeXBTf2f3sqbmxK53wiWF+dSg3WrgppgUbGUQGAACQDwjTAHLCxr3xidhlAeniwX6vl5OUgM/o+jHx3en72J0GAADIC4RpADnhmcXxEu/LhgZUEsidEu+ELzeVej8yP6z6EGdOAwAA5DrCNICc8OySeIn35SNyq8Q7YVAnW2f18elAo6tnFjOIDAAAINcRpgFkvd31jmatjipgSxcNyq0S75auHxN/IYCp3gAAALmPMA0g673wQUSOK00e6FdFce6VeCd85pSA/LY0dXlEu+o4cxoAACCXEaYBZL1EifenhufurrQktSuzdOEgv6KOKPUGAADIcYRpAFktGHb16rKIjImfL53rrhoZ/zs8vpAwDQAAkMsI0wCy2tTlETVEpHE9fercJvcfsj4xLKBivzRzVVRb91PqDQAAkKty/8oUQF57/oP4kVifHJbbJd4JFcVGlw3xy3WlJxexOw0AAJCrCNMAspbjuHrxw3jgnDIs90u8E646LX7mNKXeAAAAuYswDSBrLdwc07YDrnq1szS4U/48XF0y2K/yIuntdVGt2x3zejkAAAA4AflzdQog77zQVOJ92VC/jMndI7EOVxIw+tTw+E77E5R6AwAA5CTCNICs9cIH8aB52ZD86JduKTHV+7EFhGkAAIBcRJgGkJW27nc0b2NMpQHpnH75F6bPH+hXVanRos0xLdtOqTcAAECuIUwDyEovfRQv8T5/gF/F/vwp8U4I+IyuGBHfnf4Hg8gAAAByDmEaQFZKTPG+bGj+7UonJEq9H18Qkuu6Hq8GAAAAySBMA8g6oair15bFd6YvGZI/R2Id7px+PnWqMFq2w9HiLZR6AwAA5BLCNICs88aqqOrD0mndbHWpzN+HKdsyuvLUxO40pd4AAAC5JH+vUgHkrJeaSrwvzcMp3oe76rSmvulFYUq9AQAAcghhGkDWeW15vMT7osH5H6bH9fSpS6XR2t2Olmyl1BsAACBXEKYBZJWNe2P6aLujymKjsT18Xi8n7SzL6BND47vTzy2NeLwaAAAAtBZhGkBWeW15VJJ03gCffHb+HYl1JJ8YFt+Bf24pfdMAAAC5gjANIKskpnhfMDD/S7wTJvX3qywgvbchpi37Ha+XAwAAgFYgTAPIGjHH1dQVTWF6UOGE6WK/ae4Pf57daQAAgJxAmAaQNeZvjGlv0FX/DpZ6V9teLyejmvumP6BvGgAAIBcQpgFkjcQU7wsLaFc64ZIhfllGmr4ioroQR2QBAABkO8I0gKzxagH2Sye0L7c0vrdPoejBvnEAAABkL8I0gKxwoNHV2+ui8lnSxH6FF6alg1O9X/iAvmkAAIBsR5gGkBVeXxFRzJHG9/aporgwjsQ63JSmvukXP4zIcSj1BgAAyGaEaQBZIdEvXUhTvA83oKOlfu0t7ahzNW9jzOvlAAAA4BgI0wCyQiEPH0swxuiyoU1HZFHqDQAAkNUI0wA8t3pXTKt3OWpfZjSya2EdiXW4y5pKvV/giCwAAICsRpgG4LnE9OrzB/plWYXZL50woY9PFUXSos0xbdrneL0cAAAAHAVhGoDnZqyKSpLOG1C4Jd4JAZ9pLnVnqjcAAED2IkwD8JTrunpjVXxnemI/n8eryQ6XDomXer/8EaXeAAAA2YowDcBTy7Y72lHnqmulUZ9qHpIk6aLB8Z3p6SsiCkU5IgsAACAbceUKwFNvrE7sSvtlTGH3Syd0bmNpZDdb9WFp9pqo18sBAADAERCmAXjqjaZ+6XMo8T7ExU270y99SKk3AABANiJMA/CM67qa2bQzfU5fho+1lAjTL3/EEDIAAIBsRJgG4JmVOx1tO+Cqpo1R/w48HLU0rqdPbYqNPtruaMt+jsgCAADINly9AvDMGy12pemXPpTPNjq7b7z0fcZKSr0BAACyDWEagGfolz62SU0/l8Q53AAAAMgehGkAnnBdVzNX0S99LJP6x38u7EwDAABkH8I0AE+s2e1o835XHcuNBnXioehIRnSxVVVqtGa3ow17Y14vBwAAAC1wBQvAEwdLvOmXPhrLMjqnuW+aUm8AAIBsQpgG4ImDw8folz6WSf0o9QYAAMhGhGkAnpjZtDM9keFjxzSp/8EhZK7rerwaAAAAJBCmAWTcut0xbdjrqH2Z0ZDOttfLyWpDO9tqX2a0Ya+jtbs5bxoAACBbEKYBZNwbq+O70mf39dEvfRyWZZp37zkiCwAAIHsQpgFkHEdiJYcjsgAAALIPYRpAxiV2pif2p1+6NZqHkK2K0DcNAACQJQjTADJq8754729VqdEw+qVbZVAnS53bGG3Z72rlTvqmAQAAsgFhGkBGzV4TL1U+s5dPlkW/dGsYYzSxL6XeAAAA2YQwDSCj3lobL/E+qw8l3sloeUQWAAAAvEeYBpBRiTA9vjdhOhmJIWQz6ZsGAADICoRpABlTF3L1/paY/LY0ujthOhn92lvqWmm0vdbVR9vpmwYAAPAaYRpAxry7PqqYI43qZqskQL90MowxHJEFAACQRQjTADJm9pqmEu8+nC99IhJHZCXO6QYAAIB3CNMAMuattfEQeBb90ickMYRs5qqoHIe+aQAAAC8RpgFkRMxx9c66+M70mYTpE9K72lbPKku76l19sC3m9XIAAAAKGmEaQEYs2RJTbUjq38FSxwoeek5U8xFZKzkiCwAAwEtc0QLICI7ESo1E3/QM+qYBAAA8RZgGkBGJ4WNnMXzspCR2pt9YTd80AACAlwjTADKCnenU6F5lq297S3uD8TO7AQAA4A3CNIC027g3po37HFWXGQ3syMPOyWou9ea8aQAAAM9wVQsg7RK70mf28skY4/Fqcl/zELJVDCEDAADwCmEaQNpR4p1aiZ3pWaujisbomwYAAPACYRpA2r3N+dIpVVNpaWBHSwcaXS3cTN80AACAFwjTANIqGHb1/uaYfJY0ujthOlXomwYAAPAWYRpAWs3bGFXUkU7taqskQL90qjT3Ta+kbxoAAMALhGkAaTWnqV/6jF7sSqfSxKad6TfXRBShbxoAACDjCNMA0irRL02YTq2OFZaGdrZVH5bmbWB3GgAAINMI0wDSxnVdho+lEUdkAQAAeIcwDSBt1ux2tLPOVec2Rj2qeLhJNYaQAQAAeIerWwBp806LEm9jGD6Wauf088mY+DneoSh90wAAAJlEmAaQNvRLp1d1maVTamw1RKS56yn1BgAAyCTCNIC0eacp4I3rSZhOF47IAgAA8AZhGkBaBMOu3t8ck8+SRnUnTKfLpP5NfdOr6JsGAADIJMI0gLSYvzGqqCON6GqrNEC/dLqc3ccny8RL6hsj9E0DAABkCmEaQFokSrzpl06vtqWWRnazFYoe7FEHAABA+hGmAaRFItjRL51+HJEFAACQeYRpACnnui6TvDOoeQjZKnamAQAAMoUwDSDlNux1tO2Aqw7lRr2reZhJtwl9/LIt6d31UQXD9E0DAABkAle5AFLunRa70sYwfCzdKoqNRne3FYlJb61ldxoAACATCNMAUo5+6cyjbxoAACCzCNMAUi4xyXsc/dIZ03zeNGEaAAAgIwjTAFKqMeJqwaaYLCON6UGYzpTxvX3y29J7G2OqC9E3DQAAkG6EaQAptXBTVJGYNLzGVnkR/dKZUlZkNLq7TzHnYM86AAAA0ocwDSClOBLLO2f1if/MZ6+h1BsAACDdCNMAUuqd9TFJ9Et7YUJTmH5zDTvTAAAA6UaYBpBS7Ex758ze8Z/5O+ujisTomwYAAEgnwjSAlNm0z9GmfY7alRr178DDS6ZVl1ka2tlWMCwt3BTzejkAAAB5jatdACmTGHw1rpdPxjB8zAtnNZd60zcNAACQToRpACnTfL50T0q8vTKheQgZfdMAAADpRJgGkDL0S3vvrBZh2nXpmwYAAEgXwjSAlAhHXc3fGJUx0lh2pj3Ts52t7m0t7ap3tXyH4/VyAAAA8hZhGkBKvL8lplBUGtrZVpti+qW9NL5pqvdbayn1BgAASBfCNICUeHst/dLZYnxTqfdbDCEDAABIG8I0gJRIDB+jX9p77EwDAACkH2EaQEq83eJYLHhreI2t8iJpxU5HO2rpmwYAAEgHwjSAk7btgKN1exxVFhsN6sjDitd8tmkut5/D7jQAAEBacNUL4KS907QrfXpPW5bF8LFscFYfvyRKvQEAANKFMA3gpCX6pSnxzh70TQMAAKQXYRrASUv0SzN8LHuc3tMny0jzNkbVEHa9Xg4AAEDeIUwDOCnRmKv3NiTKvAnT2aKi2GhEV1uRWDxQAwAAILUI0wBOyuItMTVEpEEdLVWV8pCSTSj1BgAASB+ufAGcFPqlsxdhGgAAIH0I0wBOSvP50pR4Z53ERO85a6NyHPqmAQAAUokwDeCkJM4xHt+HMJ1turW11KPK0p6gq+U7HK+XAwAAkFcI0wBO2LYDjtbsdtSm2GhIJ9vr5eAIDpZ6RzxeCQAAQH4hTAM4YYld6TN6+WRZxuPV4EgSYXr2GvqmAQAAUokwDeCEzWnqlz6T4WNZiyFkAAAA6UGYBnDC6JfOfsO72KooklbtcrS9lr5pAACAVCFMAzghjRFX8zdGZRlpbA/CdLayLaMzmioH5rA7DQAAkDKEaQAnZP7GqMIx6ZQutiqK6ZfOZuN7x4/IotQbAAAgdQjTAE5Ic790b3als12iDP+tNUz0BgAASBXCNIATkigZZvhY9ju9p0+2Jc3fFFND2PV6OQAAAHmBMA0gaa7r6u11B4/FQnYrLzIa0cVWJCa9t5FSbwAAgFQgTANI2ro9jrbXuupYbtS7moeRXNB8RBbnTQMAAKQEV8EAktZyV9oYho/lgrP6cN40AABAKhGmASSNEu/ck5joPWddVI5D3zQAAMDJIkwDSFpzmGaSd87o2tZSzypLe4Oulu1wvF4OAABAziNMA0hKMOzq/c0x+SxpdHfCdC5JHJE1myOyAAAAThphGkBS5m+MKupII7raKg3QL51LmoeQ0TcNAABw0gjTAJJCv3TuIkwDAACkDmEaQFISYXpcT8J0rhlWY6tNsdHqXY6219I3DQAAcDII0wBazXVddqZzmG2Z5n83zpsGAAA4OYRpAK22bo+j7bWuOpYb9a7m4SMXJUq9GUIGAABwcrgaBtBqLXeljWH4WC6ibxoAACA1CNMAWo0S79x3ek+fbEtasCmmYNj1ejkAAAA5izANoNXeSYTp3oTpXFVWZHRqV1tRR3pvA7vTAAAAJ4owDaBVGsKuFm2OyWdJo7sTpnPZWZR6AwAAnDTCNIBWmbcxqqgjjehqqzRAv3QuG9/HL4kwDQAAcDII0wBaZU5T8DqTfumclxhCNmdtVI5D3zQAAMCJIEwDaJU5Tf3SZ9IvnfO6VFrq1c7SvgZXH26Peb0cAACAnESYBnBcrus270wzyTs/NB+RtYZSbwAAgBNBmAZwXKt3OdpV76pLpVGPKh428gHnTQMAAJwcrooBHFfLfmljGD6WD87qQ5gGAAA4GYRpAMd1sF/a7/FKkCpDO9uqLDZas9vR1v2O18sBAADIOYRpAMfFJO/8Y1mmuf+d3WkAAIDkEaYBHNP+BkdLt8VU5JNGdrO9Xg5SaHxzqXfE45UAAADkHsI0gGOauyEm15VGd/cp4KNfOp8whAwAAODEEaYBHFNziTfnS+edsT18si1p4aaYgmHX6+UAAADkFMI0gGOa01QCTL90/ikrMjqtm62oI81dz+40AABAMgjTAI4q5rh6Z31MknQGO9N5iVJvAACAE0OYBnBUH26L6UCjq77tLXWq4OEiH41vOu5s9hrCNAAAQDK4OgZwVG+v40isfJfYmX57XVSOQ980AABAaxGmARxVYvjYGYTpvFVTaal3taX9ja4+2BbzejkAAAA5gzAN4KjmrGOSdyGgbxoAACB5hGkAR7SzztHKnY7Ki6RhNbbXy0EanUWYBgAASBphGsARJfqlx/X0ybaMx6tBOo3vEw/TDCEDAABoPcI0gCN6ey0l3oViSCdbbUuM1u1xtGW/4/VyAAAAcgJhGsARJfqlGT6W/yzLNP87U+oNAADQOoRpAB8Tibmau+FgmTfyX/MQsjURj1cCAACQGwjTAD5m0eaYGiPS0M622pbyMFEImOgNAACQHK6SAXwM/dKFZ2xPn3yWtHBzTPUh1+vlAAAAZD3CNICPaT5fmn7pglEaMDqtm62YI727nt1pAACA4yFMA/iYRKkvw8cKC6XeAAAArUeYBnCIjXtj2rTPUXWZ0YCOPEQUkvF9/JII0wAAAK3BlTKAQ8xZe7DE2xjj8WqQSYmd6bfXRRVz6JsGAAA4FsI0gEMkdiXHM3ys4HRuY6lPtaUDja4+2BbzejkAAABZjTAN4BDNw8cI0wXprD6J86Yp9QYAADgWwjSAZvUhV4s2x+SzpNHdCdOFKFGRMJu+aQAAgGMiTANo9t6GqGKOdFo3WyUB+qULUfNEb3amAQAAjokwDaAZ/dIY3MlW2xKj9Xsdbd7neL0cAACArEWYBtDsYL+03+OVwCuWZZr75d9aG/F4NQAAANmLMA1AkuQ4rt5m+BjUotSbvmkAAICjIkwDkCQt2+Fob9BVzypLXSp5aChkzRO9CdMAAABHxRUzAEnSnKaS3vF92JUudGO6++S3pUWbY6oLuV4vBwAAICsRpgFIkuY07UKe2YswXehKAkandbMVc6R317M7DQAAcCSEaQCSDpb00i8NSRrfNISOI7IAAACOjDANQLvqHK3Y6agsIA2vsb1eDrJAom96NhO9AQAAjogwDaB5V3pcL598tvF4NcgGiYneb6+LKhqjbxoAAOBwhGkAmrU6HqbP7sP50ojrWGFpUEdLdaH4IDIAAAAcijANQLNWx0t5z+5LvzQOmtA3/uJK4vcDAAAABxGmgQJX2+hqwaaY/LZ0ek/CNA5KvLjyJkPIAAAAPoYwDRS4OWsjclxpTA+fSgL0S+OglmHaceibBgAAaIkwDRS4WWsS/dLsSuNQPaps9aiytLve1bIdjtfLAQAAyCqEaaDANQ8fo18aR5D4vaBvGgAA4FCEaaCANUZczV0flWWkM3sTpvFxE/rQNw0AAHAkhGmggM3dEFU4Jp3a1VZlCQ8H+Lizmyd6R+W69E0DAAAkcPUMFDBKvHE8Azta6lButGmfo/V76JsGAABIIEwDBezg+dJ+j1eCbGWMaS71Trz4AgAAAMI0ULAiMVdz1sbD0VlM8sYxTOgTf7GFvmkAAICDCNNAgVq4Kab6sDSks60O5TwU4OiY6A0AAPBxXEEDBaq5xJtdaRzHiK62KoqkFTsdba+lbxoAAEAiTAMFi+FjaC3bMhrfu6nUm75pAAAASYRpoCA5jtvc/zqB4WNohQl9E+dNU+oNAAAgEaaBgrR0W0z7Glz1qbbUrS0PAzi+s5vDNDvTAAAAEmEaKEiUeCNZY3r4VOSTFm2OaX8DfdMAAACEaaAAHQzTlHijdYp8Rqf39Ml11XykGgAAQCEjTAMFxnXdg5O82ZlGEg4ekUWYBgAAIEwDBWblTkfba111qTTqU81DAFpvQp94JcMs+qYBAAAI00ChOXi+tF/GGI9Xg1xyRi+fbEt6b0NUwbDr9XIAAAA8RZgGCgzDx3CiKoqNRnWzFYlJb69jdxoAABQ2wjRQYAjTOBnn9Gsq9V7NedMAAKCwEaaBArJ+T0zr9zqqLjMa3Mn2ejnIQec0vQjzxip2pgEAQGEjTAMF5M2mwVET+vhkWfRLI3ln9fHJGOmd9VE1RuibBgAAhYswDRQQzpfGyaossXRqV1uhqDR3A7vTAACgcBGmgQLC+dJIBUq9AQAACNNAwdhe62j5DkcVRdKILvRL48Sd01TZ8AZDyAAAQAEjTAMF4s2mEu/xvf3y2fRL48RNaNqZnrM2qnCUvmkAAFCYCNNAgaDEG6lSXWZpeI2thog0fyOl3gAAoDARpoECwfnSSKXE79EbqwnTAACgMBGmgQKwN+ho8daYiv3S6B6EaZy8c/o19U0zhAwAABQowjRQAN5aG5XrSuN6+lTko18aJy8x0Xv22oiiMfqmAQBA4SFMAwWAEm+kWscKS0M626oLSQs2xbxeDgAAQMYRpoECcHD4mN/jlSCfTOwXf3Fm5iqOyAIAAIWHMA3kubqQq/kbY/JZ8TJvIFUmNvVNz6RvGgAAFCDCNJDn3l4XVdSRRnW3VVZEvzRSJ9E3/eYa+qYBAEDhIUwDeW7q8ngJ7nkDKPFGatE3DQAAChlhGshzry2Lh+kLBhKmkXr0TQMAgEJFmAby2PZaR+9viaksIJ3Ri35ppB590wAAoFARpoE8Nq2pxHtSf78CnC+NNKBvGgAAFCrCNJDHXmsK0+fTL400adk3PX8jfdMAAKBwEKaBPOW6bnOYvmAQYRrpM6mpb3r6SvqmAQBA4SBMA3lq6daYth1w1b2tpYEduasjfSY3VT5MW0GYBgAAhYMrbCBPtdyVNoZ+aaTPxH4+WUZ6a01UwTB90wAAoDAQpoE8dfBILKZ4I73alloa08NWOCbNXsNUbwAAUBgI00Aeaoy4mrUmKmOk8xg+hgyg1BsAABQawjSQh2aviaoxIo3ubqu6jLs50o8wDQAACg1X2UAeau6XHsiuNDLjjN4+lfilhZti2lXneL0cAACAtCNMA3ko0S99PmEaGVLkMzq7b/z37XWOyAIAAAWAMA3kmW0HHL2/JaaygHRGL4aPIXMmD4j/vk1bwRAyAACQ/wjTQJ5J9KxO6u9XwMeRWMgc+qYBAEAhIUwDeebgkViUeCOzTuliq32Z0drdjtbsinm9HAAAgLQiTAN5xHVdTV1BmIY3LMs0H8XG7jQAAMh3hGkgjyzdGtO2A656VFka0JG7NzKPvmkAAFAouNoG8kjLI7GMoV8amZfom56+IiLHcT1eDQAAQPoQpoE80twvPYgSb3ijV7Wtvu0t7Qm6WrSZvmkAAJC/CNNAnmgIu5q1JipjpPP6cyQWvMNUbwAAUAgI00CemL02qsaINLq7rXZl3LXhnUSYnrqcMA0AAPIXV9xAnuBILGSLc/v7ZBlp1uqo6kL0TQMAgPxEmAbyRPPwMfql4bF2ZZbO6OVTOBYfRAYAAJCPCNNAHth2wNHiLTGVF0njetIvDe9dOiT+os6LHxKmAQBAfiJMA3kgMehpUj+/Aj6OxIL3Lh0aD9MvfRiW61LqDQAA8g9hGsgDHImFbDO8xla3tpY273f1PkdkAQCAPESYBnKc67oH+6UZPoYsYYxpLvV+6SNKvQEAQP4hTAM5bsnWmLbXuupZZal/B+7SyB70TQMAgHzGlTeQ415uCirnD/TLGPqlkT3O7e9XkU96Z11Uu+sdr5cDAACQUoRpIMc9/0E8TF82lBJvZJeyIqNJ/fxyXOkVSr0BAECeIUwDOWxXnaO310VV5JMmDyBMI/skpnpT6g0AAPINYRrIYa8si8hx4+W0ZUWUeCP7JPqmX1kWUTTGEVkAACB/EKaBHPb8Ukq8kd16V9sa3MnS3qCrd9ZHvV4OAABAyhCmgRwVibl6ZRlhGtnvsqEBSdJzSyn1BgAA+YMwDeSo2WuiOtDo6pQutnpU2V4vBziqTw2Pv9jz7JKwXJdSbwAAkB8I00COen5pWJI0hV1pZLnTe/rUsdxo5U5Hy7ZzRBYAAMgPhGkgR73wYaLEO+DxSoBjsy2jTwyL/54+uyTs8WoAAABSgzAN5KAVO2JaudNRh3KjMT0o8Ub2+2RTqfe/lhKmAQBAfiBMAzno+Q/igeTSIX7ZFkdiIfud19+vsoD07vqYtuyn1BsAAOQ+wjSQg174gBJv5JaSgNGFg+K708+zOw0AAPIAYRrIMXuDjt5cE5Xfls4fyPAx5I5PDU/0TXNEFgAAyH2EaSDHvLosopgjTeznU5tiSryRO+JtCdLrKyM60MgRWQAAILcRpoEc89zSphLvIZR4I7e0K7M0oY9P4Zj0ykeUegMAgNxGmAZySCTm6qWmI7E+MYwSb+SeRKn3v5ZS6g0AAHIbYRrIIW+ujmp/o6vhNbZ6VXMkFnLPJ4cdHELWEKbUGwAA5C7CNJBDEkdisSuNXNWr2ta4nrZqQ9KLH7I7DQAAchdhGsgRrus2l8Z+Yhj90shd14wukiQ9Mj/k8UoAAABOHGEayBHzNsa0drejbm0tje5OiTdy15WnBmRb0ksfRrQ36Hi9HAAAgBNCmAZyxBML4yXenx3hl2VxJBZyV8cKS+cP8Csck55+n6neAAAgNxGmgRzguq6eWBQPHZ8bWeTxaoCTd/WoeKvCowsI0wAAIDcRpoEc8M66qDbsddSzytLYnpR4I/d9anhAJX5p5qqoNu+j1BsAAOQewjSQAxK70leODMgYSryR+yqKjT4xLCDXlR5bwCAyAACQewjTQJZzHFdPNpd4M8Ub+eMaSr0BAEAOI0wDWW7Ouqg273fVp9rSad0o8Ub+uHCQX+1KjRZuiumjbTGvlwMAAJAUwjSQ5f6x8OCuNCXeyCcBn9FnT03sTlPqDQAAcgthGshiMcfVU+9T4o381VzqPT8s13U9Xg0AAEDrEaaBLPbm6qi2HXA1oIOlU7pQ4o38M763T93bWlqz29G76yn1BgAAuYMwDWQxSryR7yzLNJ85/ch8Sr0BAEDuIEwDWSoac/X04oNHYgH5KhGm/7EwrGiMUm8AAJAbCNNAlpq5Kqqdda6GdLY1rMbn9XKAtDmli0/DamztrHM1bUXE6+UAAAC0CmEayFIPzI2XvF7FrjQKwNWnHRxEBgAAkAsI00AW2hd09PTisIyRrh9LmEb++3xTmP7nkrCCYUq9AQBA9iNMA1nosQVhNUakCwf61aOKKd7If72qbY3v7VNdSHp+KbvTAAAg+xGmgSx0f1OJ9xdPZ1caheOa5qnehGkAAJD9CNNAlvlga1TvbYipqtToE8MI0ygcnz01IJ8lvfxRRLvrHa+XAwAAcEyEaSDL3D83vit39WkBFfs5WxqFo325pQsH+RV1pKcWsTsNAACyG2EayCKRmKuH5yVKvIs8Xg2QeYkzpx9dQJgGAADZjTANZJFXPopoe62rYTW2TuvG4DEUnk8OC6g0IM1aHdWGvTGvlwMAAHBUhGkgizQPHhsbkDGUeKPwlBUZfWp4fHf6cXanAQBAFiNMA1liZ52j55dG5LOka0dT4o3ClZjq/fC8sFyXM6cBAEB2IkwDWeLheWFFHemSIX51rOCuicJ1/kC/OlUYLdka03sbKPUGAADZiSt2IAu4rqs/v9UoSfrKOHalUdj8ttEXx8bvB/fOafR4NQAAAEdGmAaywOsro1qx01GPKkuXDPF7vRzAc189Ix6mH18Y1v4GzpwGAADZhzANZIF7mnalv3ZGkWyLwWNAn/a2LhjoUzAcb4EAAADINoRpwGNb9jt6dkl88NiXKfEGmt14ZrEk6d45IQaRAQCArEOYBjx239shxRzp8hEBdW7DXRJImDLMr85t4oPI3lkX9Xo5AAAAh+DKHfBQNObqL2/HS7xvGs+uNNCS3zbNA/n+Z1bI49UAAAAcijANeOiFDyLavN/VkM62zu7r83o5QNa5aXyx/Lb01Pthrd/DMVkAACB7EKYBDyUGj339zCIZw+Ax4HBdKi1dfVpAMUf6X3anAQBAFiFMAx5ZtTOm15ZHVRqQvjAm4PVygKz13YnxQWR/fbtRBxoZRAYAALIDYRrwyL1z4rtsV59WpMoS7orA0Yzo6tPkAT7VhqT7mmYMAAAAeI0reMADjRFX98+Nh2kGjwHH9/1J8d3p/5kVUjTG7jQAAPAeYRrwwBMLw9pd72psD1undWfwGHA8Fw7ya0hnWxv2Onr6/bDXywEAACBMA5kWc1zdMa1BkvSts4s9Xg2QG4wx+l5T7/SvZzbKddmdBgAA3iJMAxn25KKwlu1w1Le9patGMngMaK1rRgXUsdzovQ0xzV4T9Xo5AACgwBGmgQxyHFf/+Vp8gNJPzy+Rz+Y4LKC1iv1GN0+I707/cjqDyAAAgLcI00AGPb04og+3xdS72tK1o9mVBpJ184QiVRRJL30Y0fyN7E4DAADvEKaBDHEcVz9/Nd4r/dPzS+RnVxpIWlWp1bw7/V9TGzxeDQAAKGSEaSBD/rkkoqVbY+pZZek6dqWBE/bdicUq8Uv/XBzR0q3sTgMAAG8QpoEMaLkr/e/nFyvgY1caOFEdyi19/cz4+ew/e5ndaQAA4A3CNJABzy2NaPGWmLq3tXTD2CKvlwPkvB9PLlF5UXx3+p117E4DAIDMI0wDaea6rn7+Wnz37CeT2ZUGUqFjhaUfTCqRJP3wuSDnTgMAgIwjTANp9sIHES3cFFPXSqMvjWNXGkiV700sVodyozfXRPXShxGvlwMAAAoMYRpII9d1dXtTr/SPJ5eoiF1pIGUqio1+dmF8d/rHLzQo5rA7DQAAMocwDaTRI/PDmr8xpi6VRl9hVxpIua+dUaQ+1ZaWbo3p4Xlhr5cDAAAKCGEaSJPaRlc/fC4oSbpzSqmK/exKA6kW8Bn916Xx3en/eKlBjRF2pwEAQGYQpoE0+e+pDdp6wNWZvX26ehTnSgPpcuWpAY3sZmvjPkd/mt3o9XIAAECBIEwDabBqZ0y/mdkoY6T//UypjGFXGkgXyzL61ZRSSdJ/TW3UvqDj8YoAAEAhIEwDafC9Z4MKx6SvjCvSqO4+r5cD5L3zB/o1eYBPe4Ku7nyd3WkAAJB+hGkgxV75KKznP4iosvhgLyeA9Ptl0+70b99o1OpdMY9XAwAA8h1hGkihcNTVd/4ZHzp2+8Ul6lDOXQzIlFHdfbphbECNEelr/6iX6zKMDAAApA9X+kAK/f7NRi3f4WhIZ1vfOIujsIBM+/UnS9Wx3Oj1lVHd/y5HZQEAgPQhTAMpsmmfo9tfaZAk/c+nS+W3GToGZFq7Mkt/uKJMUnx2wZb9DCMDAADpQZgGUsB1XX3l8XrVhqTPnhrQ5IF+r5cEFKwrRvj1qeF+7W909c2nKPcGAADpQZgGUuD+d8N6dVlE7cuM/nB5qdfLAQqaMUZ/vKJMlcVGzy6J6On3I14vCQAA5CHCNHCSNu6N6bvPxoeO/fGKUnWs4G4FeK1LpaVffyr+wtY3n6rXnnrKvQEAQGpx1Q+cBNd19dV/BHWg0dUVI/y6ciRDx4Bs8aXTAzq3v0876lzd/HTQ6+UAAIA8Q5gGTsKd0xuby7v/2DT0CEB2MMbovqvKVFEkPbYgrEfmhbxeEgAAyCOEaeAEzVgZ0b+/2CBjpL9fW0Z5N5CFelfb+sPl8Re6vvFUUOt2xzxeEQAAyBdc/QMnYPM+R1c9WCfHlW69sEQXDQ54vSQAR3HdmICuPDWgA42urnm4XuEo070BAMDJI0wDSQpHXV35YJ121Lm6aJBf/3FBsddLAnAMxhj9+cpSdW9rac7aqL70GMdlAQCAk0eYBpL0w+eCmrM2qp5Vlh6+rkyWZbxeEoDjqCq19PxXy9Wm2OiR+WH9cTb90wAA4OQQpoEk/GNBSP8zK6SALT31xXJVl3EXAnLFiK4+PXh1vH/6e88GNXMl508DAIATRxIAWunDbTF9+fF6SdLvLy/V6B4+j1cEIFmfOiWgW84tViQmfepvdVqyJer1kgAAQI4iTAOtsHW/o0v/Uqv6sHT9mIC+egbnSQO56peXlehzIwPa3+jqontrtX4PE74BAEDyCNPAcRxodHXJX2q1bo+jM3r5dM9ny2QMfdJArrIsowevKdO5/X3ast/VBffUamed4/WyAABAjiFMA8cQjrq6/P9qtWhzTAM7xgcYlQQI0kCuK/IZPfvlCo3qbmvFTkeX3Fur2kYmfAMAgNYjTANHEY25uu7hek1bEVXnNkav3FjBwDEgj1QUG730tQr172Bp3saYPv1/tWoIE6gBAEDrkAyAI0gE6ScWhVVZbPTy1yrUq9r2elkAUqxjhaXXvl6hLpVG01dEdelfa1UXIlADAIDjI0wDh4nGXF3/aL0eXxgP0lO/UaFTuzG5G8hXvaptvf6NNupaaTRjZVQX/rlW+xvooQYAAMdGmAZaaIy4+uwDdXp0flhtio1eu6lCYzgCC8h7AzvZevPbbdSrnaU5a6M694+12naAQA0AAI6OMA002V7r6Px7avXskojalRpN+0aFxvYkSAOFond1PFAP6GBpwaaYxv7mgBZzDjUAADgKwjQgacHGqMb8+oBmr4mqe1tLs77FjjRQiLq1tfTWv7XRhD4+bdzn6MzfHdBzS8NeLwsAAGQhwjQK3mPzQxr/vwe0cZ+jCX18mvf9NhpaQ5AGClX7cktTv1Gh68cEVB+WPvW3Ov3itQY5DoPJAADAQYRpFKyY4+rHzwd19d/r1RiRvn5mkaZ9o0IdK7hbAIWuyGd0/9VluusTJZKk/3ipQRffW6sdtfRRAwCAOFIDCtLW/Y4u+0udfjW9UT5LuuezpbrnyjIFfMbrpQHIEsYY/eDcEk29qUKdKoxeWx7VyLv3a9bqiNdLAwAAWYAwjYLiOK7+MqdRg+/Yr1eWRdSh3Oj1b1bo6+OLvV4agCx13gC/Ft1SqUn9fdqy39XEP9Tq356pV20jZd8AABQywjQKxkfbYjrnD7W68Ymg9je6+uQwvxb+oFIT+vq9XhqALNe5jaWpN1XototKZBvpf2eFNOSX+xlOBgBAASNMI++Foq5uf6VBp961X7PXRFXTxuiZL5Xr2a9UqGtb7gIAWse2jG69qEQLb2mjcT1tbdrn6JP31emK+2u1ZT+91AAAFBqSBPJWKOrq3rcadcqv9uu2VxoUjkk3jS/SRz+p1KdPCXi9PAA5aliNT2/9Wxv98YpSVRRJT78f0eA79utPsxsVjVH6DQBAoeD8H+SdfUFH984J6X/fbNSW/fEL28GdLP31c2Ua34eSbgAnz7KMvnFWsT45LKBvP1OvZxZH9M2ngvrtzEbdelGJPn9aQLbFQEMAAPIZYRp5Y+PemH73Rkh/fbtRtaH4n43sZuvfJ5fok8P98ttc2AJIra5tLT39pQr9a0lYP34+qGU7HF33cL3+e2qDbruoRFeMCMgiVAMAkJcI08hprutq9pqo7p0T0j8WhhVtalu8cJBft5xbrHP7+2QMF7IA0uuTwwO6bKhfjy0I6/ZXGvTRdkefe7Bew2oa9e2zi3T1aUUqK+KxCACAfELPNHLSpn2O/ntqgwb8136d/ftaPTI/PlH3utEBvf/DNnrl6xU6b4D/uEF606ZNMsbIGKNNmzZlYunIcvxO4HCt/Z2wLaNrR8fnMvztqjL1rLK0dGtMX/tHUF1u3advPV2vD7fFMrhyAACQTuxMI2fUh1y99FFE978b0qvLInKa5vz0qLL0xbEBfXlckbpX2d4uEkDB89lGXxpXpGtHB/TPxWH96a2QZq2O6g9vhvSHN0M6vaetz40s0hUj/DxmAQCQwwjTyGqb9zmaviKiZ5eG9cpHETVE4n9e5JMuHxHQF8cW6dz+PnoSAWSdgM/oc6cV6XOnFemDrVH9eU5ID70X1rvrY3p3fVDfe1Y6vaetKUMDmjLMr+E1Nm0pAADkEMI0sobrutq4z9Gbq6OauSqqmasiWrXr4Nmtxkjje/t0zaiArjotoKpSuhQA5IahNT79/nKf7pxSqleWRfTEwrCe/yARrBv0/15qUM8qS1OG+XXhIL/O7OVTuzIe4wAAyGaEaXhmZ52jD7bGNG9jVG+vi79tPXDoGa1tS4wm9PHp0iF+fXJ4QJ3bcHEJIHeVBIw+fUpAnz4loIawq9dXRvTc0ohe+DCs9Xud5lJwSRpWY2tCH1/8ra9f3dry+AcAQDYhTCPtDjS6+mBrVEu3xbR068G3HXXuxz63Q7nRmb18mtjPr3P6+XRKF5uzWgHkpZKA0aVDA7p0aECuW6oFm2J64YOIZq6K6J310ebHynveiofrrpVGI7v5NLKrrZHdfDqtm60eVRal4QAAeIQwjZNWF3K1ca+jjfua3vbGtHGfow17Ha3YGf/vkVQWGw2rsTWiq60zevl0Ri+f+lRzYQig8BhjNKq7T6O6+3SrShSOulqwKaY310T05uqoZq+NavN+V5v3R/TCB5Hmr6sqNRrW2Va/9pb6dWj6b3tb/TrYalPMYykAAOlEmIakeL9yQ0SqDbmqbXRVF3JVG3J1oNHV7npXu4Oudtc7h7y/sy7e47yv4eM7zC2V+KWhnW0Nq2l662xrWI1PXSoNwRkAjiDgMxrXy6dxvXy65dz4Y/Ta3Y4Wbo5p4aaoFm6OacGmqLYdcPXmmqjeXPPx2+hQbtS32lK3tpa6VFrq0ib+304VlqpKjdo1vVUWG4Y4AgBwAgjTOa6urq75/d+8sFbFlQ0KRaWwI4WjUjjmxv8/6iock0IRV3VhV/XheGCuj0j1jfE/c4+diY+qJKDmi7Tm/za99aiy1KOt9bELNbdO2lx3lBvMoK1btx7xfRQufidwuGz5nQhIOr06/qYR8T/bUeto9a6Y1u11tG63o3V7Ylq3x9HaPTHtrJV2tma5Jl4p1LbEHPxviaVSv1TsNyrxGxX7pRKfVOwzKg40/ddvZBvJb8fP2Pbbavp/o+1b1jbf/LJly9Lx4wAAoNWi0ah27twpSRo+fLiKi4tTcrvGdU80QiEbPPDAA/riF7/o9TIAAAAAIOvNnTtXY8aMScltMRoUAAAAAIAkUead48aNG9f8/rPPPqsuXbp4uJrcs337dk2ZMkWS9Pzzz6tTp04erwhe43cCh+N3InkLFy7UjTfeKCm+A1BTU+PxigAAhWzr1q0aO3asJKlDhw4pu13CdI4rLy9vfn/UqFHq1q2bh6vJPZs2bWp+/9RTT+XnB34n8DH8TpycmpoafmYAgKzh86UuAlPmDQAAAABAkgjTAAAAAAAkiTANAAAAAECSCNMAAAAAACSJMA0AAAAAQJII0wAAAAAAJIkwDQAAAABAkjhnGgWtW7ducl3X62Ugi/A7gcPxOwEAAI6EnWkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEmEaQAAAAAAkkSYBgAAAAAgSYRpAAAAAACSRJgGAAAAACBJhGkAAAAAAJJEmAYAAAAAIEk+rxcAAAAA4Oi2HXD06rKIdtQ52lnnqj7sqrLYqG2J0Tl9/Rrb05YxxutlAgWHMN1CMBjUG2+8ofnz52vBggWaP3++NmzYIEm69dZbddttt6X8e/7yl7/UT37yk+b/d1035d8DAAAAucVxXP15TkgPzA3pvQ2xY3xmg07vaeuPV5RpVHcu7YFM4h7Xwty5c3XJJZdk7PstX75ct99+e8a+HwAAALLfki1R3fhEUG+vi0qSSgPS5AF+9e9gq1OFUYnfaE+9q037Hb3wQVjvro/p9N8e0LcmFOlnF5aoqpROTiATCNOHqaqq0mmnndb89t3vflfbtm1L+fdxHEdf/vKX1djYqDPOOENvv/12yr8HAAAAcofrunrovbC+/UxQBxpd1bQx+t2nSzVlaEAlgSOXcdeFSvXzVxv06xmN+t0bIT2zOKLfX16qTwwLZHj1QOEhTLcwYcIE7dmz55A/+/GPf5yW7/X73/9eb731lq655hr169ePMA0AAFDAgmFXNz1Zr4feC0uSPn9aQH/9XJnKio7dC11eZHTnJ0r1hTEBfeXxer27PqZP/61O93y2VF87szgTSwcKFjUgLdi2nZHvs3btWv30pz9VdXW1fvvb32bkewIAACA7ua6rLzxSp4feC6u8SHromjI9ct3xg3RLw2p8evPbbfTT84vluNKNTwT16xkNaVw1AHamPfDVr35V9fX1+tOf/qQOHTp4vRwAAAB46D9fbdTT70fUtsRo9rcrNLTmxC7R/bbRLy4tVd/2tr78eL1+8K8GdSiz9IWxRSleMQCJMJ1xf/3rXzV9+nRNnjxZX/jCF1J621u3bj3u53Tr1i2l3xMAUHg2bdp0zI9v3749QysBct/f3gnp1lcaZFvSI9eVnXCQbumLpxcpHHX19SeD+vqT9Tq1m61TunDZD6Qa96oM2rx5s2655RaVlJTo3nvvTfntjx079rifw9FbAICT1b17d6+XAOSFVz4K68Yn6iVJf/5sqS4ZkrqhYTeOL9a8jTHd905IF9xTq/d/WKlOFXR4AqnEPSqDbrzxRu3fv1+33Xab+vTp4/VyAAAA4JElW6K68oE6xRzp1guL9ZUzUj8s7PeXl+rc/j5tr3X1pcfq2VQBUoyd6Qx5+OGH9eKLL+rUU0/V9773vbR8j7lz56qmpiYttw0AQMLGjRuP+fFFixZpypQpGVoNkHu2HXB06V/qVBuSrh0d0K0XlaTl+xT7jR65rlzDfrVfL30Y0d/fC9M/DaQQYToDduzYoe985zuybVt//etf5fOl58deU1NDTzQAIO2O91zTmhkeQKFyXVdf/Ue9Nu5zdFYfn+67qkzGtH5qd7I6t7H0P58u1bUP1+vbzwR17gC/urWlOBVIBe5JGfCjH/1Iu3fv1te+9jUNGjRIdXV1h7yFw+Hmzz3SnwEAACA/PLEwrBc+iE/ufuL6chX50hekE64eFdCnhvu1vzF+BBfl3kBqEKYzYO3atZKke+65RxUVFR97u+OOO5o/N/FnP/zhD71aLgAAANJgb9DRt58JSpJ+/clS1VRm5lLcGKP7ripTh3KjGSujev6DSEa+L5DvCNMAAABABvzs5QbtqHM1sZ9PXzw9dZO7W6O6zNJtTb3ZP3ouqGiM3WngZBGmM2DmzJlyXfeob7feemvz5yb+7He/+513CwYAAEBKrdoZ05/fCsky0p+uSG+f9NF89YwiDehgadkOR/e9E8r49wfyDWH6MHv37tWuXbua3xzHkSQFg8FD/ryuru6Qr3vggQdkjJExRjNnzvRg5QAAAMhW3/9XUFFH+vK4Ig3ubHuyBr9t9KsppZKkW19uUG0ju9PAySBMH2bkyJHq0KFD81vi+I+77rrrkD+/+eabPV4pAAAAcsHU5RE9tzSiNsVG/3lJeo7Baq1PDvfrrD4+7ahzddfrDZ6uBch1hGkAAAAgTRzH1S3/ig8d+9mFxepU4e3ltzFGd38ivjt994xGbdnveLoeIJcRpg+zbt26Y/Y3J94eeOCBQ77uhhtuaP7YxIkTk/qet912W/PXAgAAIH88uSis97fE1KPK0s0Tir1ejiTp9F4+XXlqQA2ReLk3gBNDmAYAAADSIBJz9R9NYfW2i0oycqZ0a/33ZSXy29L/vRvSB1ujXi8HyEmEaQAAACANHpwb0sqdjgZ1tHTd6MwehXU8fdvbuml8kRxX+vEL7E4DJ8Ln9QIAIN0cx1VdWDrQ6Kou5Ko+7CoYdtUQkRqjrsJRKRxzFYlJUUeKOa4cV3JdyRjJSLItybaMinxSke/gf0v8UqnfqLzIqCxgVFliVBaQJ0eeAACyR2PE1e2vNkqS/vOSUvns7Hte+H8XlOiBuWG98EFEM1dGNLG/3+slATmFMA0gb9SFXO2odbSzztWueke76129+GFEDRFXxxtJYFmSbeKB2Zimsh0jyZVcSY4rOa6rqCM5x5nVYllSid/o0iF+VZcZtS+z1KHcqFOFpdJA9l1MAQBS7563Qtq0z9Fp3Wx95pTsDKkdyi39+Lxi/fuLDbr91QbCNJAkwjSAnOO6rvYGXW054GjbAVfbah09uziiUPTQxGxbUlkgHmJL/FKxz6jILxXZRgFf/LxNvyX5bMlKYic55sRDdTQW39EOx6RQ1FUoGt+JaIhI9WFXT78fVuyw4F3sN/rEML86t7HUpY1R10pLlSWGnWwAyCO1ja7+e2q8dPq/Ly2RZWXvY/zNE4r1q+mNmrkqqnkbohrdg3gAtBb3FgBZL+a42rzf0fo9jjbsdfT0+2FFYgc/bllSRVE8NLcpjpdcVxTFS7HTEVJty8i2pCKfVKaj377rumqMSHVhVwcaXdWG4v89PGQX+42uHBlQr3aW+lRbaldKuAaAXPbXtxu1q97V2X19umBQdu/2VhQb3TS+SL+c3qi7ZzTq8evLvV4SkDMI0wCy0p56Ryt3Olq1K6YnFx0Mn5aRKkuM2pYYtS2xVFlsVFaU3M5yphhjVBKQSgJGHVpcmziuq/qQtK/B0d6G+C77Q++F4vXkkkoDRtePDWhgR1u92lmys3hHAwBwqJjj6o+zQ5KkH51XnBMvjn7r7GL9emajnlwU1h2XxdS72vZ6SUBOIEwDyAqu62rbAVcfbo/pvrdDqgvFk6VlpLal8b7j6jKjqhKT8+HSMkYVxVJFsa3uVfE/C0dd7Q662lXnaEedq3uaLsT8tnT92CIN7WyrdzXBGgCy3UsfRrRmt6N+7S1dlOW70gldKi1dOyqg++eG9T9vNOp3nynzeklATiBMA/DUnnpHi7fE9Oc5IdU3Begin1GPKkudKuIB2p+FE1BTLeAzqmljVNPGkuu6qgtJ22odbdnv6L63Q02fI33p9CKN6OpT10pKwQEgG/3vrPgE729NKM7qXunDfXdise6fG9aD74V1x2WlKmFgJnBchGkAGReOuvpgW0y/m9moPcGDAbp3taUubSxVFXjPsGmxc92/g626UHzY2uZ9jv78VkhSSOVFRv92TrFGdLG54AGALPHhtpimrYiqvEi64fQir5eTlOFdfBrX09Y762N6enFY147OrfUDXiBMA8iYHbWO5m6I6r63Q4o58cFhXdta6lZpqX25ycq+52xQXmQ0oIOt/u0tHWh0tXGfo037HP3Xaw2yLelL44o0todPndtYXi8VAAraH96M70rfMLZIbYpz7znta2cW65319frr2yHCNNAKhGkAaeW6rtbudjR7TVRPLgpLig8Q61FlqWulVRAl3KlijFFliVFliaXBnVxt2e9o/V5Hf50T0l/nhNSu1OhHk0s0qKOVU6WFAJAP9gUdPfhevC3n5rOKPV7Nibny1IC+88+gZq2Oavn2mAZ2YhAZcCyEaQBp4Tiulu9w9F9TG7S/wZUx8QEnfaotteVc5ZNmW0bdq2x1r7K1r8HRuj3xMvAfPRdUWcDoB+cW69Sutny8WAEAGXH/3LCCYenCQf6cDaFlRUafPy2ge+eE9NC8kP7r0lKvlwRkNcI0gJRyHFdLtsb0y2mNqgu5siypV7WlPtW2yujtTYu2JZZO7RrfrV6729G6PTHd/kqDinxG35tUpNHdfQr4+NkDQLrEHLe5xPvbZ+d2efR1o+Nh+pF5Yf3nxSVUOgHHQJgGkBKu62rp1pj+a2qj6kOufLbUr4Ot3u0sFft5Is6EIp/RoE62+rW3tH6vozW7Hd0xtVEBn/Sdc4o1tqdPRYRqAEi5XDwO62jO7O1T72pLa3c7mrU6qon9c/vvA6QTYRrASXFdV6t2Obrt5QYdaIyH6AEd4yGa3VBv+Gyjvu1t9WpnaeM+R6t2ObpzejxUf39Sscb28FH+DQAp9Ps3c/M4rCMxxugLowO6/dVG3T83RJgGjoHRrwBO2Ia9Md3wSL2+98+g6sKu+ra3dV5/vwZ2tAnSWcC2jHq1s3VuP59O6WLLMkZ3TG3U5x+q14KNUTmO6/USASDnfbgtpqnLc/M4rKO5fmz87/HU+2HVNvJcARwNYRpA0nbXO3psfkjffDKovQ2uerazdG5/v4Z0JkRnI8sy6tnO1rn9fRrS2Zbjurr15QZd+3C9Vu2Meb08AMhpuX4c1pH0rrY1qb9PwbCaT+IA8HGEaQCtFo66mr4ioi89Wq9H54fVpdLSpH5+ndLFpxL6orOebcXLv8/t71f/DraCYVff/WdQjy8IaW/Q8Xp5AJBz9gUdPZTjx2EdzRebdqfvnxvyeCVA9iJMAzgu13X1wdaorv17vX43s1HlRUZn9vZpVHefyooI0bnGb8cHlU3q51eXSkuPzAvrhkfr9caqiKIxyvkAoLXunxtWfY4fh3U0l48IqKJImr0mqpVUMQFHRJgGcEx7g44emR/Wj59vUNRxNazG1oS+PlWX8fCR60oCRqO6+3RGL59K/UZ3v96oa/9er9W7uGgCgOPJp+OwjqQ0YPS5kfG/1wPsTgNHxNUwgCOKxFzNXBnRDY/W6x8LwupeZWlSf796V8cHWSF/tC+3dE5fnwZ3ttUYdfWdZ4J6+v2w6kPsUgPA0by6LH+OwzqaL54ekCQ99F5YMYZWAh9DmAbwMat2xnTt3+v16xmNKvUbndHLp1O7ckZxPrMso37tbU3s51enCksPvBvSdQ/XadGmqFyXCygAONyDc+ODuW48syjnj8M6mjN6+dS/g6VN+xzNWBn1ejlA1iFMA2jWGHH1z8VhffefQYWiroZ0tnVOX5/al/NQUShKA0Zjetga1d0nyxj9x0sNenR+WAc4GgUAmu1vcPSvpWFZRrp6VP6VeCcYY3T9GEq9gaPhChmAJGn1rpiue7he//dOSB0q4lO6+7a38/bVdhydMUZdKi1N7OdTt7aWHl8Q1g2P1On9zexSA4AkPfV+RKGodN4An7pU5vfl9HVjAjJGemZxWPs4+QE4RH7f+wEcVyTm6qUPw/rOM0FFHVcjuto6vYetkgAhutAFfEYju/k0pkd8l/r/vdigxxaEVcsuNYAC9/em47CuG52/u9IJPapsnT/Ap4aI9MziiNfLAbIKYRooYPsbHN3wSL3umR1SdZnROX396lFlyzBgDC10bhPfpe7a1tJj88O64dE6Ld/OxG8AhWn9npjeWB1VaUD69CkBr5eTEVc2TfX+55KwxysBsgthGihQG/fG9OXHgjrQ6GpAR1vjevlUym40jiLgMzqtm0+je/jkutIP/hXUyx8y3RVA4XlkfjxQfuaUgMqLCuN585PD/LKM9NqyiPY3UOoNJBCmgQITc1zNWBnRN54KKuK4GtXdp4EdOe4KrVPTxtI5/fyqLjP60+yQHp0fVihKoAZQGFzXLagS74T25ZbO7e9TOBbvFwcQR5gGCsjeoKMvPlqv38xoVJsiowl9/Hk/OAWpV+I3GtfLpx5Vlp5YGNaXHq3X7np2KgDkv/kbY1q2w1FNG6PzBvi8Xk5GfaFpqveDTPUGmnEVDRSIxVui+tJj9dobdNW3vaWz+vjUppjdaJwYyxid0sXWoE62DoRcffmxes6kBpD3/j4vHiSvHlUku8BOu/jMKQGVBqQ310S1cS9zMwCJMA3kvWjM1fNLw/rpCw2yTXxHcUhnX8FdBCD1jDHq38HWGb188tvxM6mfXRJRJEagBpB/IjFXjy2I90tfO6owBo+1VFZkNGVo/O/95CIGkQESYRrIawcaXX3x0Xr9ZU5IHcqNzunnU4dy7vZIrfZlls7p61OnCkv/905INzxSrz2UfQPIM1OXR7SzztWwGlsjutpeL8cTnxsZD9P/WEiYBiTCNJC3tux39OXH6rWvwVW/DrbG9vSpyMduNNIj4DMa08PWwI5NZd+P12sZx2cByCN/fy8eIK8bHSjYIyQvHuxXRZE0d0NMa3fzGA8QpoE89MHWqL7xZL3CMVendfdpcCemdSP9jDHxY9Z6+mQZ6ZZ/BTVteUQOx2cByHEHGl09uzQsY+L90oWq2G/0yeGUegMJhOmjCAaDevnll/WLX/xCn/nMZ9SzZ08ZY2SM0W233XbCt7t7927df//9uvbaazVkyBCVlZWpqKhI3bp106c+9Sn985//TN1fAgUn5rh69aOwfvx8g3yW0Rm9fOrKtG5kWIdyS2f38auq1Oh/3mjUowvCaggTqAHkrqffD6sxIp3b36dubQv7eZVSb+Cgwprpn4S5c+fqkksuSfntdu7cWdFotPn/i4uL5ff7tXnzZm3evFn/+te/dPHFF+upp55SaWlpyr8/8lcw7OqbT9VrV52rdqVGo7r7VOxnNxreKAnEX8xZui2mfywI64WlEf3xs6WqLivsi1AAuSkxxbuQzpY+mgsG+lVZbLRgU0wb9sbUo6ow+8cBiZ3pY6qqqtJ5552nW265RY899pg6d+580rcZjUY1duxY/elPf9Lq1avV0NCguro6rV27Vl/+8pclSS+//LJuvPHGk/5eKBw7ah3d8Gg8SPeqtnRGL4I0vGdbRiO6+DS8i61gxNVXH6/XOnrsAOSYjXtjmrkqqhJ//HioQhfwGV002C9JevGDiMerAbxFmD6KCRMmaM+ePZo2bZruvPNOXXXVVSoqOvlXI19//XW9++67uummm9SnT5/mP+/Vq5fuu+++5hD98MMPa+PGjSf9/ZD/NuyN6etP1Ksx4uqULraG1/hkcewVskivdrbG9vDJlfRvzwS1eEv0uF8DANni0flhua706VMCqijm+VWSLh0SD9MvfEiYRmEjTB+FbaenZGXSpEnH/Hhid1qS5s2bl5Y1IH+s2BHTt58OynGl03v61LMdpVbITh0rLI3v7VfAZ/TTFxo0e01ErksfNYDs9+T78d7gq09jVzrh4sF+GSO9vjKiIDMxUMAI01mmuLi4+f1YjHJIHN38jVF9/19B2ZZ0Ri/Oj0b2a1NsdFYfn9oUG/1qWqNeXUagBpDd1u2Oaf7GmNoUG00e6Pd6OVmjfbmlcT19aozEAzVQqLj6zjIzZ85sfn/48OHeLQRZy3VdvbEqotteblCp32h8b7+qSrkrIzeU+I3O7O1TuzKjP74Z0qvLuAgDkL2eXhzflZ4y1K8iHyXeLV02tKnUm75pFDCmeWeRffv26Y477pAU79keOHBgUl+/devW435Ot27dTmhtyA6u6+qlDyP681shVZYYje3BoDHkHr9tNK6nT++sj+qPb4ZUXWZpTA+ejnLJpk2bjvnx7du3Z2glQHo9/X48KF4+ghLvw1021K+fvtigFz+MVxkZw/UICg9XL1nCcRxdd9112rp1q4qKivT73/8+6dsYO3bscT+Hksrc5bqunv8gor/OCal9udHo7j75bZ64kJtsy2hMd59mr4nq56826D8vLtGp3XhKyhXdu3f3eglA2m3e5+jtdVGVBaSLBlHifbjhNba6tbW0aZ+jxVtiGtGVx3AUHmpDs8S//du/6YUXXpAk/elPf9KIESM8XhGyieu6eqEpSHeosDS2B0EauS/gMxrXy6cSn9F/vNzAlG8AWeWZphLvS4YEVBLgOfdwxhhdNoRSbxQ2XkLKAj/4wQ/0hz/8QZL029/+Vl/60pdO6Hbmzp2rmpqaVC4NWcBx4jvS970dUodyozHdbdkcfYU8URowOqO3T3PWRvX/XmzQnZ8o1aBOTKXPdsc7unHRokWaMmVKhlYDpMfTTVO8Lz+FXemjuWyoX3+eE9KLH0b00wtKvF4OkHGEaY/98Ic/1K9//WtJ0l133aXvfOc7J3xbNTU19ETnGcdx9dzSiP72TkgdKyyNJkgjD5UFjM7o5dNbayP60XNB/c9nStWrmkCdzY73XNOaGR5ANtte6+jNNVEV++M70ziySf39KvZL76yPamedw8kiKDj8xnvolltu0V133SVJuvPOO/WDH/zA4xUhmyR6pAnSKATlRfGhZJYlfe/ZoHbUOl4vCUABe3ZxWI4rXTjQr4pinnuPpjRgdF5/v1xXevlDSr1ReAjTHvnBD36gu+++W1I8SN9yyy0erwjZZvqKqO57myCNwlFZYml0d59ijnTz00EFwwxMBOCNxJFYTPE+vkub+qZfJEyjABGmPfCDH/ygubT77rvvJkjjY+auj+p/3mhUVakhSKOgdCi3NKzGVkPY1TeerJfjEKgBZNb+BkczVkbls+LnS+PYLm36Gb2yLKJIjMdsFBbC9DHs3btXu3btan5znHjZYTAYPOTP6+rqDvm6Bx54QMYYGWM0c+bMQz72ox/9qDlI/+Y3v9H3v//9jPxdkDs+2BrVf77WoLIiozE9fARpFJye7Wz1qra0u97V9JVM+AaQWa98FFHUkc7p51PbUi6Vj6dHla1Tutg60Ohq9hoes1FYeIQ4hpEjR6pDhw7Nb4nppXfdddchf37zzTe36vY2bNigO++8U5JkWZZ+9atfqXPnzkd9S5SBo3Bs2BvTv7/QoGJfvH+0yEeQRmEa2slWVanR/77RqGXbY14vB0ABea7pmKdPDKXEu7Uo9UahIkxnUGJnO/H+9u3bj/l2+I438tuBRlff+2eDjJHG9rBVypmWKGCWZTSqu08Bn/TvLwS1N8hAMgDpF4m5eqkpEE4ZRol3a102NHHedNjjlQCZxdFYx7Bu3boT+robbrhBN9xww8f+vFevXnJdeknwceFovD80FHU1qrtPlSW8zgWU+I1O7erT3PVRfeefQT1wdZmM4UUmAOkze01U+xpcDa+x1Zsj+lrt9J4+VZcZLd/haNXOmPp14GeHwsAVO+Axx3H1zOKw9je4GtDRVpdK7pZAQqcKS72qLe2pdzV3PeXeANLruaXxndVPsCudFNsyungwpd4oPFy1Ax5yXVevLovokXlhdW1raUAH7pLA4QZ3jLc93DGtgXJvAGnjuq6eW9rULz2MfulkXTaEUm8UHq7cAQ+9vS6qP80OqbrMaEQXmxJW4Ah8ttEpXWzFHOm5pRHaZQCkxbLtjtbsdtSpIn4sJZJz4SC/bEt6Y3VUtY08TqMwEKYBjyzZEtUdUxtVUWw0ujtHYAHH0qHcUo92lp5aFNbCTZR7A0i9Fz+M76heMtgvi+fkpLUttXRWb58iMWnqckq9URgI04AH9gUd/cdLDSr2G53ew6cAR2ABxzWkk61iv9EvXmtQfYhdDwCp9ULTkViXcSTWCWOqNwoNYRrIMNd19fJHEcUc6dSutko4AgtoFb9tNLSzrUhMmraCXQ8AqbMv6Gj22qj8tnT+QIaPnahLh8RfiHjpo4gchxc9kf8I00CGvbMuqkfnxweOdSjnLggko6aNUftyo/veCWl7LcPIAKTGa8vjL3Kf3denimJe5D5RgzpZ6lNtaXutq/m05KAAcM40kEGb9jm6Y1qjyouMhtcw3MRrNz5Rn9Lbu/fKspTeHj7OGKMhnWzNWh3Va8sium5MkddLApAHEsc5JXZWcWKMMbp0iF+/fzOklz+MaEwPogbyG7/hQIaEoq6+/2xQxkijuvvkt3nlO5NSHZxb+z0I2KlXWWKpa1tLTywM68zePvVtzwtTAE6c47h65aN4mL5kMCXeJ+uCQfEwPW1FRD+7qMTr5QBpRZgGMmTq8oiCYVfDamy1oYQs7TIRnlvj8HUQrlNjUEdbWw84uvXlBv392jKOlQNwwhZujmlHnas+1ZYGdKT96mSd09cvnxU//rMu5Kq8iMdn5C/CNJABa3fHdO9b8fOke7XjiTodsiU8H0/LdRKsT1xpwKhHlaV1ux2t3uWoXwd2pwGcmJeaSrwvHuznhbkUqCg2GtfLp9lropq1OqJLKJ1HHiNMA2kWirr60fMNsi1pRBcfT9QplCsB+mgI1ienX7WtDXsc/fzVBj14DbvTAE7Myx8dDNNIjckD/Jq9JqqpywnTyG9skQFpNn1FRA1hV4M72Sqj1CklbnyiPueD9OHy8e+UbiUBo25VlvYGXa3fw2RvAMnbU+/o3fVRFfmkif0I06kyeUB8v27aiqjHKwHSi51pII227nd0z1shtSulvPtkFUrQZLc6OX2rbW3Y6+id9VH1qqbUG0Bypq2IyHHjR2LxgnfqjO3pU0WRtHRrTNsOOOrchmsg5Cd+s4E0cV1XP3o+KCNpOOXdJ6yQd2wL+e/eWuVFRp3KLT0yP6x9QXanASTn1WXxEu8LB7ErnUp+2zTv9E9fEfF4NUD6EKaBNFm6Naa9QVc921lM7z4BBMmD+DkcW+9qS3KluRtiXi8FQA5xXVevNIXpiwjTKTd5QPxnOnU5YRr5izANpEE05uo/X2uU35YGMmU4aYTHj+PFhaNrX2ZUXmT057caFY66Xi8HQI5YujWmLftddWtraUhnnqtTbfLAeJietiIi1+WxGfmJMA2kwXsbomoIu+rXwVbAx650axEYj4+f0ccZY9S72lIkJi3ewu40gNZ5tcWuNK1YqTe4k6WaNkab97tavoM2HOQnwjSQYvUhV3e93qgSv1Fvho61CgExefy8DtWt0pLfln49o5EdEACt0lzizZFYaWGMaS71nkbfNPIUV/pAis1eE1EkJg3pbMu2eKX7WAjRJ4ef3UE+26hHlaW6kKu1u9kBAXBsdSFXb66Oyrak8/pzuE26nD+QvmnkN8I0kEL1IVf3vBVSm2KjmjYE6WMhCKYGL0gc1KMq3vO4YBOl3gCObeaqiMIxaVxPn9qWcjmcLuc17UzPWBlRNEbVEPIPjx5ACs1ZF1XMkfp3sOm/OgrCX3rwM40fk9W+3Oih90KqbeSiDcDRvUqJd0Z0qYwPd6sNxefJAPmGMA2kSH3I1R/fbGRX+hgIfOnFCxVSr3a2XFdasImLNgBH98pHHImVKZMHxMvop63gcRn5hzANpMjcDexKHw0hL7MK+WfdqcKoyGd0z1shBpEBOKJVO2NatctR+zKj07pxJFa6MYQM+YwwDaRAJObq97MaVRYw6syu9CEKOdh5qVB/7pYx6lJpVB9ytWU/YRrAxyVKvC8Y5JfFoNC0m9jPL9uS3l4XVV2Ix2XkF8I0kAKLt8QUiUm9qy1Z7Eo3K9RAly0K9effvW38qY1SbwBH0vJ8aaRfRbHRuJ4+RWLSrNXsTiO/EKaBk+S6ru5+vVF+W+rWlrtUQqEGuWxTiP8OlSWWKkuM/u/dkCJMjwXQQijq6vWVB3emkRnNfdMckYU8w5U/cJJW73JUF3LVo8qS32ZXmv7o7FOI/x7d21qKxqSPtnNMFoCD3loTVX1YOq2brU4VXAZnyuSBib5pKoaQX3gUAU7SexuikolPES50hRjackWh/dt0qbRkjLR0K2EawEGvcCSWJ07v6VN5kbRka0zbDjheLwdIGcI0cBIONLp6dEFYncotlQYKe1e60MJaLiqkf6Min1H7ckuPLwirIUypN4C4RL/0hZR4Z5TfNprYL/4zn85Ub+QRwjRwEhZuikqu1KOqsO9KhRTScl0h/Vt1aWPkutLyHexOA5C27He0eEtMFUXSGb18Xi+n4HBEFvJRYScA4CS4rqs/zg6p2G/UsaJwd6ULKZzli0L5N+tcES/1XkKpNwAd3JWePNDPjBMPNA8hWxGV61IxhPxAmAZO0Po9jhrCrrpWFu5xWIUSyvJRIfzbBXxGnSosPbEorAONXLgBhe6Vj5pKvAdS4u2FIZ1t1bQx2rTP0Yod9E0jPxCmgRP0/pb4blehHodVCGEs3xXCv2HXSktypY+2sTsNFLJozNVryxk+5iVjTHOp91RKvZEnCjMFACcpHHV1/7shVZYYtSkuvF3pQghhhSLf/y07lhvZlvQhR2QBBe3d9VHta3A1uJOlnpy+4ZnmvmnOm0aeIEwDJ2D5jphiTvws20KT7+EL+cVnG3Uot/TM4rBqKfUGClbzkVhM8fbUeU1hesaqqKIxHpOR+wovCRxDMBjUyy+/rF/84hf6zGc+o549e8oYI2OMbrvttpR8j9raWt12220aPny4ysvLVVlZqTFjxujXv/61wuFwSr4H0u+DbTHJSDVtCusuRJDOT/n+79qlqdT7Q0q9gYL1avP50gGPV1LYura1NLiTpQONruZt5DEZuY9zAVqYO3euLrnkkrTd/vr16zVx4kStW7dOklRaWqpQKKR58+Zp3rx5euSRRzR9+nRVVVWlbQ04eeGoq8cWhNW+zKjYXzgl3vkeuArdjU/U694ry7xeRlp0roiXev/+zUad3qvc6+UAyLBddY7mbYyp2C9N6MOlr9cmD/Dro+0hTV0e0TiOKEOOK6xttVaoqqrSeeedp1tuuUWPPfaYOnfunJLbjcVimjJlitatW6eamhpNnTpV9fX1CgaDevzxx1VRUaGFCxfqmmuuScn3Q/qs2BmT4xTerjTyX76+YGJbRjVtLO1vcLWnngmyQKGZtiIi15XO6etXSaBwXgTPVucP5Lxp5A9eDmphwoQJ2rNnzyF/9uMf/zglt/3AAw9oyZIlkqSnn35aZ5xxhiTJsix97nOfk+M4uvrqq/Xyyy9r+vTpOu+881LyfZF6H213JBM/w7ZQ5GvIQuHoUmlp0z5Hy3bEdGbvwrnvAjhY4n0h/dJZ4Zx+ftmW9Pa6qOpCrsqLeIEDuYsrihZsO33THR988EFJ0qRJk5qDdEtXXXWVevfuLUl66KGH0rYOnJxIzNUj80JqV1I4Jd4E6cKSr//e7UqNZKTN+xl4AxQS1z14JBZhOju0KTY6vadPkZj05mp2p5HbCNMZEAwG9dZbb0mSLr744iN+jjFGF110kSTptddey9jakJy1ux3FHKlzgZR452uwwrHl47+73zaqKDL6x4KwXJdADRSKpVtj2rLfVbemwVfIDpMHxItjp62IerwS4ORQ5p0BH330kRwn3qc3bNiwo35e4mPbtm3Tnj171K5du6S+z9atW4/7Od26dUvqNnGo5Tvikyc7FUCJdz4GKhS2juWWVu+KX1h3bVsYlSXpsmnTpmN+fPv27RlaCXBsLUu8jeF+ny0mD/Dr5682airnTSPHEaYzYMuWLc3vd+3a9aif1/JjW7ZsSTpMjx079rifw47MiXNdVw+9F1Z5kaG/B3kvH6d7d25jtHqX9NH2mLoW4BnxqdS9e3evlwC0Cv3S2WlcL5/Ki6QlW2PadsApmIo/5B9+czOgtra2+f3S0tKjfl7Lj7X8GmSHHXWuGiOuOpTnf5BmVxpS/v0eVDXNOrjv7RAvLAIFoCHs6s01UVlGOq8/+0fZxG8bndM3/gLH6yvZnUbu4pElj8ydO1c1NTVeLyNvrSiQEu98C1BAgjFGnSuM1u1xtLveVfsCeGEsXTZu3HjMjy9atEhTpkzJ0GqAI3trbVShqDSmh612Zfn93J2Lzu3v04sfRjRjZVRXjyryejnACSFMZ0BFRUXz+8Fg8Kif1/JjLb+mtWpqauiJTqPVuxzZllRdygU4Cke+lXt3qrC0bo+jpVtjmtifi+sTdbznmtbM8ADSLXGO8eQBlHhno4n9/JIaNHMVO9PIXVxJZECXLl2a39+8efNRP6/lx1p+DbwXc1w9uySsqlIjy8rfMM2uNI4kn34v2pfHS73/9Bal3kC+I0xntxFdbbUtMVq1y9GmfY7XywFOCGE6AwYPHizLiv+oly5detTPS3ysc+fOSQ8fQ3pt2Bs/Eqs6j8vE8ikwAUdjGaOaNkYNYVe76gnTQL7aU+9owaaYiv3Smb0pxMxGtmV0Tt/4v80M+qaRo/I3GWSR0tJSjR8/XpL0yiuvHPFzXNfVq6++Kkm64IILMrY2tM7KnfFXTDvRY4kClU8vtrRvelFs9S52QoB8NWNVVK4rTejjU7Gf5+5sNal/vGqAUm/kKsJ0hlx//fWSpBkzZujdd9/92MeffPJJrVmzRpL0hS98IaNrw/Hd/25IxX6jNsX5+YScT0EJOJ72ZUa2Jf1pdqPXSwGQJtOWU+KdCyY1TVmfsSrq8UqAE0OYPszevXu1a9eu5jfHie9cBIPBQ/68rq7ukK974IEHZIyRMUYzZ8782O1ef/31Gj58uFzX1eWXX67p06dLkhzH0ZNPPqmvfvWrkqSLL75Y5513Xnr/kkjK/gZHdaH4kVjG5GeYBlojX1508dlGndtY2ht0tTfI7jSQj+iXzg3DOtuqLjNau9vR+j0xr5cDJI0wfZiRI0eqQ4cOzW+J4z/uuuuuQ/785ptvTup2fT6fnnvuOfXq1UubN2/W5MmTVVZWprKyMl155ZU6cOCARo4cqUceeSQdfy2chHV74hfb7fO0XzpfAhKQjK6V8fvzkq1cvAH5Zt3umFbtctSu1OjUrrbXy8ExWIf0TbM7jdyTn+kgS/Xq1UuLFy/Wz372Mw0bNkzGGPn9fo0aNUp333233nnnHVVVVXm9TBxm/d54mK4uY1cayJcXXzqUG/lt6S9zQl4vBUCKTW8KZecN8Of1CRz5ItE3/cZq+qaRexhveJh169ad0NfdcMMNuuGGG477eRUVFbr99tt1++23n9D3QeY9Nj+skoBRSR4OMMmXYAQkyzJGHcotbdkfb+MoL8q/+zdQqA72S3OZmwsm9ov/O82kbxo5iJ1p4BiCYVd1IVftSrnQBhLy5UWYjhXx+/XqXZR6A/nCcVxNX0m/dC4Z0slW+zKjdXscrdvN4zFyC2EaOIYNTSXe+Rim8yUQASeqY7klGWn5Di7egHyxZGtMO+tc9WpnqU97+qVzgWUZndO0O/3GanankVsI08AxHAzT3FWAlvLhxZgin1FVidGj88OKxlyvlwMgBaYzxTsnTezHedPITSQE4Bj+/l5IfluqKPJ6JamVD0EISIVOFZZijrRxH0dkAfng4PAx+qVzCX3TyFWEaeAoojFX+xtdtSu1OF8aOIJ8eFGmY3mib5owDeS6SMzVrKaJ0IkJ0cgN9E0jVxGmgaPYE3TlulJFcX4F6XwIQECqtCmOH5H14FyOyAJy3bwNUdWFpGE1tjpVcImbS+ibRq7ikQY4ip118Z2q8jwr8QZwkDFG1WWW9jW6aozQNw3kstebSrzP7U+Jdy6ibxq5iDANHEWih7KqJH/uJuxKI9Xy4XeqfZmRXGndHkq9gVz2etORWJP6UeKdi+ibRi7Kn5QApNjWA678tlQW8HolANKpQ3n8qXDdHvr0gFzVGHE1Z11Uxqi5XBi5hb5p5CLCNHAEruvq+aVhVRQbho8Bea4sED8m6+/vhb1eCoAT9M66qBoj0mndbFVxnGVOsiyjs/vSN43cwqMNcAR1ISkSkyqK8idI50M5LrJTrv9uGWNUVWpUG3IV4bxpICclSrzPZYp3TqNvGrmGMA0cQWL4WD6FaQBH17bEyHWlLfvpmwZyEcPH8gN908g1hGngCHbVx3enygnTQEFoX8Z500Cuqgu5end9VD5LOqsPO9O5bGhnW9X0TSOHEKaBI9jRfCxWfoTpXC/DRfbL9d+xyhIjny2tZaI3kHPeWhNR1JHG9vTlzfN2obIso3P6sjuN3EGYBo5g2wFHAZ9UTLUYUBAsY1RdaulfS8IKR+mbBnIJJd75ZVJT3/sM+qaRAwjTwGFc19ULH0TUhkneQEFpVxbvm95M3zSQUxg+ll8mteibdl1e3ER2I0wDh9nf4Crm5M/wsVwvv0XuyPXftXal8fv8hr2EaSBX7A06mr8ppiKfdEYvdqbzwZDOtjqUG23Y62jtbh6Pkd0I08BhttXGXwVtU5wfYRpA61QWG1kWYRrIJTNWRuW60oQ+PhX7ed7OB8aYFkdk0TeN7EaYBg6z7UD8QpowDRQW2zJqW2L09PthOQ6lhUAumLYiXuI9eQAl3vkkcUQWfdPIdoRp4DA76hzJ5EeZd66X3SL35PrvXFWJpZgj7awnTAO5oDlMDyRM55NJLXam6ZtGNiNMA4f55+KIyvxGtpX7YRpAcuibBnLH+j0xrdzpqF2p0aldba+XgxQa1MlSpwqjTfscrd7F4zGyF2EaaMFxXAXDLudUAgUqEabXc940kPWmrTh4JBYvgOeXQ/umKfVG9iJMAy3sa3DluFJZkdcrAeCFgM+ovMjoHwvDXi8FwHFMW06/dD6b1Nw3zRAyZC/CNNDCnmC8L6cskPuvcOd67ypyV67/7rUrM2qMuNoXZHcayFaO42r6Svql89mkpnPDZ6yM0DeNrEWYBlrY25A/YRrAiakujT81bthHmAay1ZKtMe2sc9WrnaU+1VzO5qP+HSzVtDHaesDVyp08HiM78egDtLCvKUyXcFYlULCqmvqmNxGmgazV8kgsY3jOzkfGmEN2p4FsRJgGWjjQmAjTHi8EgGdK/VLAJz06n75pIFtNWx7vo508wOfxSpBO9E0j2xGmgRYONLoK+MRUUKCAGWPUrtTSgUZX4Sh9ekC2CUVdzVoT36k8l+FjeS2xMz1zFX3TyE6EaaCFlz+KqNhHkAZOVq4PIasqMXJdaesBSr2BbPPOuqiCYWlEF1sdyrmUzWd9qi11a2tpe62rZdt5PEb24REIaBJzXDVEXJXmwfCxXA8ygNfalsQfB7bs5+INyDZTm47EOp8p3nnPGNOi1Ju+aWQfwjTQpCEiyZWKeW4GCl5lIkwfoKwQyDaE6cLCEDJkM8I00KQ+HL9oDti5vzMN4OT4baOygNFTixhCBmSTvUFH8zbGFLCls/owfKwQTGzamZ65KkrfNLIOYRpowiRvAC1VFBvVh13FHC7egGwxY2VUjhsP0vnQloXj611tq2eVpV31rj7YFvN6OcAhCNNAk0SYZgAZkBq53rtfXhQfQrYnSJgGsgUl3oVpUv+mvumVHJGF7EKYBprUheIXzEVUjQGQVF4U/++uOsI0kC2mrYiH6ckciVVQJvY7eEQWkE0I00dRW1ur2267TcOHD1d5ebkqKys1ZswY/frXv1Y4fHI9dFOnTtWVV16pnj17qri4WCUlJerTp4+uueYavfHGGyn6GyBZiZ7pInamAUgqayoh3dvARG8gG6zbHdOqXY7alRqN7GZ7vRxkUGJneuaqqBxab5BF2IM7gvXr12vixIlat26dJKm0tFShUEjz5s3TvHnz9Mgjj2j69OmqqqpK6nZd19VNN92ke++9t/nPiouLZYzR2rVrtXbtWj366KP67ne/q9/85jep/CuhFYJNr5EEcvz5OddLa4FsUepvCtOUeQNZIbErfW5/n2yLF74LSY8qW32qLa3Z7WjJ1phGdCXCIDuwM32YWCymKVOmaN26daqpqdHUqVNVX1+vYDCoxx9/XBUVFVq4cKGuueaapG/7gQceaA7SV1xxhVasWKGGhgYFg0EtW7ZMn/zkJyVJv/3tb/XPf/4zpX8vHF845sqyJIsnaACKt3zYlrS7njANZIOpy+P9svRLF6bEEVkzV9E3jeyRsTBdV1enYDAox8nucrkHHnhAS5YskSQ9/fTTmjx5siTJsix97nOfaw7DL7/8sqZPn57UbT/00EOSpH79+umxxx5T//79mz82cOBAPfnkk+rTp48k6YknnjjpvwuSE45KnIoFIMEYo4oio38tDXMcC+Axx3E1fSXDxwpZ4ogszptGNklbmF6yZIl+8pOfaOTIkaqsrFRlZaUqKirk9/vVrl07nX766br22mv1m9/8Ru+9917WhOwHH3xQkjRp0iSdccYZH/v4VVddpd69e0s6GI5ba+vWrZKkESNGyOf7eHmK3+/XqaeeKin+4gMyKxR15SNNA2ihXZlROCrtYnca8NSizTHtrnfVp9pS7+oc78fCCZnUNITsjdVRjixE1kh5mG5oaNCNN96okSNH6s4779TixYtVW1sr13Wb3/bt26f33ntPjz32mG655RaNGzdObdu21dVXX62XXnpJsZg3Z8gFg0G99dZbkqSLL774iJ9jjNFFF10kSXrttdeSuv3ErvP777+vaPTjJSqRSESLFi2SJI0ePTqp28bJm7EqKj+NDwBaqCyOPyjsqM2OF3yBQsWRWOja1lL/Dpb2Nbh6fzPnTSM7pDQ6xGIxXX755brvvvvkOE5zeD6algG7rq5O//jHPzRlyhT16tVL99577xEDZzp99NFHzTvkw4YNO+rnJT62bds27dmzp9W3f9NNN0mSVq1apc9//vNatWpV88eWL1+uK6+8UmvWrFHfvn313e9+N+n1b926VZs2bTrmG44u5sT7IwGkTq4PxGs+Houd6UMc77lm+/btXi8ReWYqR2JBHJGF7JPSUXiPPvqoXnnlFRkTL5WdPHmyvvSlL2nAgAGqqKhQNBpVbW2tdu3apcsuu0zGGJWVlam+vv6Q4L1582Z94xvf0N13362HHnroiOXW6bBly5bm97t27XrUz2v5sS1btqhdu3atuv0pU6bot7/9rX70ox/pqaee0lNPPaWSkhJJ8R39tm3b6qabbtIvfvELtWnTJun1jx079rifQ9/f0TmuK8tQ5g3goMTxWHuY6H2I7t27e70EFJDGiKvZa6IyJj7JG4VrUj+f/vp2SDNWRfW9SV6vBkjxzvT//d//Nb9/11136bXXXtNVV12l0047Tf3799fgwYM1duxYXXLJJc2fd8MNN2jXrl16+umndf3116tt27aS4qFv9erVOuecc3TnnXemcplHVVtb2/x+aWnpUT+v5cdafk1rfOc739Ezzzyjjh07SoqH6IaGBklSKBRSbW2t9u/fn9RtIjUcV2KQN4CW/LZRkc9oZx1l3oBX3l4XVSgqjexqq10ZJWSFLLEzPWt1VNEYL3LCeyl9RFq+fLmMMRo2bJi+//3vt/rrqqqq9OlPf1r333+/tm/froceeqh5EFc0GtVPfvIT3XrrralcqieCwaA+97nP6bLLLlOPHj302muvadeuXdq5c6dee+01DR06VA8//LDGjh2rxYsXJ337c+fO1caNG4/5hqOLl3mTpgEcqqJYeunDCJU9LRzvueb555/3eonII4npzef2p8S70NVUWhrU0dKBRleL6JtGFkhprUxiR/Vkhmf5/X5de+21uuaaa3Tffffpu9/9roLBoH7xi19o4MCBuvrqq1O13I+pqKhofj8YDB7181p+rOXXHM8tt9yiJ554QgMGDNCsWbOaS7wl6fzzz9dZZ52lU089VStWrNA3v/lNvfnmm0mtv6amRt26dUvqaxDnOK5cdqYBHEF5kdGuOlcHGl1VlvAgIem4zzWJ0yuAVHh9ZXyGDiXekOK708t2hDRjVUSje/A7AW+ldGc60UucbOnzkRhj9NWvflWzZs1S27Zt5bquvv3tbyc18CtZXbp0aX5/8+bNR/28lh9r+TXHUltbq7/85S+SpJtvvvmQIJ1QUlKim2++WZI0e/Zs7dixo1W3jZMXbargZAAZgMOV0zcNeKY+5Ord9VH5LOmsPuxMQ5rUP3HedGYHFQNHktLoMHr0aLmuq5kzZyoSSc2UvdNOO01/+9vfJEl79+7V3//+95Tc7pEMHjxYlhX/kSxduvSon5f4WOfOnVs9fGzFihXN08n79u171M/r379/8/tr165t1W3j5NWH4xfJAY6uBHCYgC8ephsYHgtk3KzVEUUdaUwPnyqKqQzBwb7pN9dE6JuG51Iapq+77jpJ0u7du/WrX/0qZbf76U9/WkOHDpUkPfXUUym73cOVlpZq/PjxkqRXXnnliJ/juq5effVVSdIFF1zQ6ttOhHRJWr9+/VE/r+VxIsmUkOPkNDZdJPtsnqgBHCrxIlswzEUbkGkHz5emnBdxHSssDelsqy4kzd9I3zS8ldIwffHFF+vss8+W67r6+c9/rgceeCBltz1q1KjmCd/pdP3110uSZsyYoXffffdjH3/yySe1Zs0aSdIXvvCFVt/uoEGDmku777vvviOeoR2LxZpLwauqqjRw4MCk148T4zRdI9MzDeBwzWE6QpgGMu215fHrpfMHUuKNgyb1ayr15rxpeCzlHaJ/+9vfVFVVpVgspi9/+cv6yle+op07d5707dbX10tS2o+Nuv766zV8+HC5rqvLL79c06dPlyQ5jqMnn3xSX/3qVyXFXzg477zzDvnaBx54QMYYGWM0c+bMQz5WUlKir3zlK5KkBQsWaMqUKVqyZIkcx5HjOFq8eLEuueQSzZkzR1L8CC3bpuY4U2JNU3oJ0wAOF2iqWGlgZxrIqC37HX2wLabyIun0nuxM46BJTZPdE5PeAa+kPEz37dtX06ZNax4adv/996tXr166+eabtXDhwhO6zcTRUcaY5iFn6eLz+fTcc8+pV69e2rx5syZPnqyysjKVlZXpyiuv1IEDBzRy5Eg98sgjSd/2r371K1100UWS4mXkp5xyikpLS1VaWqoRI0botddekyR9/vOf109/+tOU/r1wbLGmAWSEaQCH8zW9rknPNJBZ05pKvCf188tPGxZaOKdv/MWV2WujitA3DQ+lZXbxyJEjNWfOHI0cOVKu66qhoUH33HOPRo8ercGDB+tHP/pRq2/rww8/1AUXXNA8IXzChAnpWPIhevXqpcWLF+tnP/uZhg0bJmOM/H6/Ro0apbvvvlvvvPOOqqqqkr7dkpISvfTSS3ryySf1yU9+Ut26dWs+t7R79+66/PLL9cILL+jRRx9lVxoAsoTPkmSkMBdsQEZNXZHol6bEG4dqX25peI2tYFh6bwNTveGdtNXMDBw4UO+++67uuOMO/eY3v2kuz16xYoXuvvtuGRN/hTHRg9ytWzdVV1ertLRU0WhUe/bs0fz58/XOO+80B06fz6fvfOc76VryISoqKnT77bfr9ttvb/XX3HDDDbrhhhuO+TnGGF1xxRW64oorTnKFSKXE690u18oADmOMkW2kMHNugIxxXVfTmsL0BYRpHMGk/j4t2RrTjJVRndmb3xF4I60NKD6fT//xH/+hb33rW/rd736nBx988JBJ1sYY7dixQy+//PJRbyMRpC3L0l//+lcNHz48nUsGAOBjLHOwHQRA+i3dGtO2A666t7U0oGNaCimR4yb18+t/Z4U0c1VEP72gxOvloEBl5NGpbdu2uu2227R27VrNmjVL3//+9zV69GhZliXXdY/5JsVLu999992kpmcDyUgciUUVJ4Aj8VlGoSgPEECmtDwSK1HNCLR0dl+fjJHeWhvl8RmeyfhoxLPOOktnnXWWpPiE7vfff18rV67UypUrtXv3boVCIbmuqy5duqhPnz668MIL1a1bt0wvEwWmqOmeEKWME8AR+G1p+oqovnam1ysBCsNUjsTCcbQrszSii61Fm2Oauz6qCX35XUHmeXrOQFlZmc4880ydeSZXJ/BWc8+0p6tIjXuvLNONT9R7vQwgr1iGmQpApoSirmatie9MnzeAgISjm9TPp0WbY5q5ijANb9CEAkiiggzAsRiTHy+2Abng7bVRBcPSyG62OpRzqYqjm5g4b3oVZxfCGzxCAZISx1c6bD0BKXXvlWVeLyElHPdgBQuA9Go+EotdaRzH2X18sow0Z21UjRGu4ZB5hGlABweQOTwOAzgKKliAzJi2nPOl0TptSy2N7GYrFJXeWcd508g8wjQgydd0T+DoGwBH4rIzDWTE3qCjeRtjKvJJ43t7OtoHOWJSv/iLLjMp9YYHCNOA4pN6JcI0gCOLOtKk/lzYA+k2Y2VUjiud1cenkgAvYeH4JvaLPzbPWMXONDKPMA1Isi0jy3DONIAjizquAj4u7IF0az5fmn5ptNKEvn7ZVrzMuyHMhRwyizANNLEsyaFpGsARRB0pYHu9CiD/TWsaPjaZfmm0Uptio1HdbIVj0tv0TSPDCNNAE4ujbwAcgeu6cpyD7SAA0mPd7phW7XLUrtTo1K7c4dB6E/txRBa8QZgGWsiXjel8OY4IuS1ffg8TsxT8NmXeQDoldqXPG+CXbXF/Q+slZlrMWMnONDKLMA008dtG0ZjXqwCQbaJNYbqI+WNAWk1bEQ9CkwdwZ0Nyzurjl8+S5m6Iqj6UJzsjyAmEaaDJ5AE+hQnTAA4TaXpcKGYAGZA2juNqeqJfmuFjSFJ5kdGYHj5FYtIc+qaRQYRpoEmx3yjCOG8Ahwk1PS6UcUwPkDaLNse0q95V72pLfdrTL43kNR+RtZK+aWQOYRpoUuI3ijlSLF8apwEP5Uu/tCSFmq7Lyou8XQeQz15rOhLrQqZ44wRN6t80hIwwjQwiTANNSpuevyn1BtBSQyT+AltlCU+ZQLokwvT5hGmcoDN7+eS3pfc2xlRH3zQyhCsDoEl5UbyEMxTNjwfgfNoZBLwUbArTbUso8wbSoT7k6q01UVlGOrc/w8dwYsqKjMb28CnmSLPXsDuNzCBMA00OhmmPFwLkuHx7IScYkWxLKgt4vRIgP81aHVE4Jp3e06e2pVya4sRxRBYyjUcsoEliuFC+7EwDSI2GsKsSv5Ex7EwD6ZAo8b5gILvSODmT+jX1Ta9iZxqZQZgGmrAzDeBwrusqGHF16VD6OIF0mbo8/sRLvzRO1hm9fCrySfM3xrQv6Hi9HBQAwjTQpKIpTDfm0c50vpXbIvvl2+9cY0RyHKm6lF1pIB0273P0wbaY2hQbje3JzjROTknAaHxvnxxXmrmK3RGkH2EaaFIWkIyRGqgMAtCkLhx/ca26jKdLIB2mrYg/6Z7b3ye/zYtWOHnnDYhXOEzniCxkAFcHQBPLMirxGzWE82dnGsDJSRyv0r6Mi3wgHTgSC6l2XtN509NXEKaRfoRpoIWLBvubz5QFkJx8K/GWpP2N8ceDThU8XQKp5jiupjYPHyNMIzVGdbfVptjoo+2OtuynbxrpxdUB0EJlsVEkJkVj+ROo8zHgAJlyoNFVsd+orIidafz/9u47PIpq4R/4d2ZrdlMIBCGQ0KRD6EWagiBFQUQsoauIXO97LVeBe6+8V/BaUCnWV/GHSlFs6NWLXimCYEGR3hQQISGVUNJ3k+zOzPn9sckaJIFssptt38/z7MOQmZ2czczOzHfOmXPI2w5lqThXLNC6kYyr43hJSt6h10kY2tb1/D1rp8nXeOQiqiS2vJMhO4+9RB4JxZs2mhAoKhMYn8QaMyJfcDfxbm/g0HPkVWzqTfWFYZqokoblYbriOUkiCl/FZa6evOOjeaok8gX3+NIdecOKvGt4+/Ka6RMKhOA1HfkOrxCIKqnosdceYp2QhWKtIQWOUN2/8ktcz9o1Y5gm8jq7Q+D7UwpkydWTN5E3dW6qQ9NoCRn5Gk6c43PT5Du8QiCqJNLo+jeUxpomotrJtbuOA4mxPFUSedt3J50oU4C+LXSItfA7Rt4lSRKbelO94NGLqJIoswRIQKni75IQBYdQrZUGXGHaapQQyc7HiLxu83HXiZa9eJOvVIw3vYVhmnyIYZqoEl35WNO2EHxmOpRDD5G3ORQBW5nAHT2N/i4KUUj6is9Lk49VPDe97TcFqhZ613UUGBimif5gXBcDbA7BDiuIriCUb9BcKG/i3bIhT5NE3pZdoOFwtoooE9C/JZ+XJt9oEatD2zgZeXaBA5mqv4tDIYpXCUR/0MgqQdWA0hBsFRTK4YfqV6jvSxdsrg5rWjFME3ldRa30sHYGGHR8jIJ8p6KpN5+bJl/hVUIVioqKsHDhQiQlJSEyMhIxMTHo27cvli5dCofDUef1FxYW4rnnnsPAgQPRuHFjmEwmJCQkYNiwYVi4cCHy8/Pr/iGo1uLKe/QuDrEevYmo5s7bBMwGyT1cHhF5z1flwYbPS5Ov8blp8jW2rfmD06dPY+jQoUhNTQUAWCwWlJWVYc+ePdizZw/Wrl2LrVu3IjY2tlbr37ZtGyZNmoScnBwAgF6vR2RkJDIzM5GZmYnt27fjlltuQY8ePbz0ichTjay/jzXdONLPhSEKQKFeK13iECgqFbjnGhMkiWGayJs0TfB5aao3w9q6os73pxSUKQImPY/p5F2sma5EVVWMGzcOqampiI+Px1dffQWbzQa73Y4PPvgAUVFR2L9/P6ZMmVKr9e/YsQM33XQTcnJyMGLECHz//fcoKytDXl4e7HY79uzZg/nz5yMmJsbLn4w8EVcRpkO0ZjrUgxD5VjjsP2eKXE28O1yl83NJiELPoSwVOUUCLWNltI3jZSj5VlykjB7NdShxAj+mcKgW8j7WTFeyatUqHD58GADwySefYMCAAQAAWZZx5513QtM0TJ48GRs2bMDWrVsxfPjwGq/bbrdj+vTpKCkpwcSJE/HRRx9Bln8/iURERKB3797o3bu3dz8UeSzaLMGgAwpKQjNME9Hl5RQLyBJwNS/0ibxuw1FXrfToTga2/KB6MaK9AQcyVWw+7sTQdmwNQd7FK4VKVq9eDQAYNmyYO0hXlpycjNatWwMA1qxZ49G633nnHZw6dQoRERFYvnz5RUGaAoskSbi1uxEFJQJaiA6lEA61i+R94bDfKKrAhWINE7sb2RyQyAcqwvSYTgw1VD9GlT9OsOkYn5sm72OiK2e327Fjxw4AwJgxY6pcRpIkjB49GgCwefNmj9ZfEb7Hjx+PuLi4OpSU6kPTKBmaAIrr3t8cUUgIhyANuDoe0wTQnk28ibwu367hh1QFBh1wPWsIqZ4MbqOHxQjsy1CRU/4YD5G3MEyXO3r0KDTN9QXr2rVrtctVzDtz5gxyc3NrtO6KDswA4LrrrsOpU6cwc+ZMJCQkwGQyoWnTphg/fjw2bNhQx09B3tI02lUjVVgamjXTQPiEI6q7cNpXKp6Xbt+Yp0cib9vyqwJVA4a00SPKzJYfVD/MBglD27pu3lR0fkfkLXxmulxWVpZ7unnz5tUuV3leVlYWGjZseMV1p6amuofUysjIQLdu3WCz2WA0GmGxWJCTk4P169dj/fr1+NOf/oTXX3+9Vp8hOzv7isskJCTUat3hJj7adSFdUKohgfecKIyFU5BWNYHsQg0xERIaWvm9v5yMjIzLzq8YsYKosg1HXddCbOJN9W1UBwO+/MWJTcecmNrH5O/iUAhhmC5XVFTknrZYLNUuV3le5fdcTl5ennt60aJFiI6Oxvvvv4+JEyfCYDAgPT0d8+bNwwcffIDly5ejY8eOeOihhzz+DP369bviMkKEbk2rNzW0SNDrgPwQ74TsjTusmP2Rzd/FIAoI54oFFBX482BeaF1JYmKiv4tAQUYIgY3H+Lw0+cfoTgbgU2DzMSc0TUCW2TKCvIO33utBRfPxiunly5cjOTkZBoPrZJKYmIi1a9eiZ8+eAICnnnoKisLu+/1JkiTcklTeCRlvQFCYCqdaaQDIKHAdq7s05fPSRN52KEtFVoFAYgMZnfkdo3rWrrGMVg1lnC0WOJCp+rs4FEJYM10uKirKPW2326tdrvK8yu+p6boTExNx5513XrKMLMt49NFHMXXqVJw/fx579+5F//79a7T+Crt27UJ8fLxH76HqJTSQoWpAUalATETo3sFk7TRVJdyCtEMRyCnUcFsPI2IieJ/5StLT0y87/8CBAxg3blw9lYaCQeVevDkkFtU3SZIwqqMBb/xQhk3HnOiVyAhE3sE9qVyzZs3c05mZmejWrVuVy2VmZlb5nsup/Jx1x44dq12uU6dO7unTp097HKbj4+P5TLQXJcS4LqjzSgRiIvxcGKJ6FG5BGgCyCjVoAujejDVmNXGlc01N+vCg8MIhscjfRlcK0/+4gRd25B28/V6uU6dO7rGfjxw5Uu1yFfOaNm1ao87HAKBhw4buQH25u7GVn2fmXVv/a97AtT+E+nPTQHiGJ6pauO4LaXkadDLY/JTIBwpKNOxIcQ2JNbw9wzT5x/XtDdDLwI4UJaRHa6H6xTBdzmKxYNCgQQCAjRs3VrmMEAKbNm0CAIwcOdKj9Vcsf/To0Wo7ATt69Kh7unXr1h6tn7wv0iTBapJwvliw4zYKC+EapAtKNBSUCNxzjQlGPW9kEnlbxZBYgzkkFvlRtFnCgFZ6KBqw7QSHyCLvYJiuZMaMGQCAbdu24aeffrpk/rp163Dq1CkAwPTp0z1a99133w3A9ZzZhx9+eMl8TdOwbNkyAK5m4b169fJo/eQb0/oYUeIUsDn8XRLfC9cgRS7hvP3T8lwdj/VK4JNPRL7w5S8cEosCw+jyfXDTMYZp8g6G6UpmzJiBpKQkCCEwceJEbN26FYAr6K5btw6zZs0CAIwZMwbDhw+/6L2rVq2CJEmQJAnbt2+/ZN1DhgzBbbfdBgC4//778eGHH8LpdH2R09PTMWXKFOzfvx8A8PTTT7ubnJN/tY1zNfk8W6xdYUmi4BXOQdqpCqTnu8aWbh7DGjMib+OQWBRIRnV07YMbjznZ6pC8grfhK9Hr9Vi/fj2GDRuG1NRUjBgxAhaLBZqmobS0FADQs2dPrF27tlbrX7VqFc6ePYtvv/0WycnJMJlMsFgsF41D/fjjj7tryMn/WjeSIUvA2WKBNo38XRrfY8/e4SecgzTgqpVWNeDRYWb2VUHkAxVDYiU0kDnsHPldz+Y6NI6UkHJBw2/nNbRrzH2S6obVn3/QqlUrHDp0CI8//ji6du0KSZJgMBjQu3dvLFmyBDt37kRsbGyt1m21WrFt2zasWLEC1157LaxWK4qLi9G8eXMkJydjx44deOKJJ7z8iagujHoJE7oZkWvToGrhcQcz3MNVOAn3ba0JgZRcDUY9kBTPCyoiX6joxftGDolFAUCWJYzswKbe5D2sma5CVFQUnnjiCY+C7V133YW77rrrisvJsox7770X9957bx1KSPWpXWPXeNPnbQJNonghQKEh3IM0AGQXCpQ4BOZcb4Zex+82kS98+QubeFNgGdXRgLV7Hdh0zIm/DDH7uzgU5FgzTXQFnZq4aqyyC8PnuWkGrdDG7et6jvO3cyp0MtC3Be8rE/lCvl3DD6kcEosCy8jy56a/PuFEmRIerQ7Jdximia6goVVGtFlCTpEGLYw6q2DgCk3cri45RQKFpQL3DzLBYmStNJEvVAyJNYRDYlEAaRIlo0dzHewOYMcpxd/FoSDHME1UA/cNNMGhABds4ROmAQavUMPt6SKEwK/nVMgyMLA1a8uIfGXDUQ6JRYFpdEc+N03ewTBNVAPh2NSbQscbd1gZpCs5UyRQUCLw50Em1pYR+YgQwt35GMM0BZrKQ2QR1QXDNFENNI6UEG2WkFUQPr16V2AIC27cfhfTNIGjOa5npQe14QU+ka8czFSRXSiQ2EBGZw6JRQFmYGs9Ik2uoduyC1hRQrXHME1UA5Ik4cFrzXCqQGYYHnQZyIITt9ul0vI12MoEHr7OjEgTa6WJfOXzn8uHxOrMIbEo8Bj1Eq5v57qhuvk4a6ep9himiWqoR4IOeh1w8rwGEUYdkVVgMAsebNZdNacqcPysCpNewsDW7MGbyJc+Pex6XvqWJLYAocA0is9NkxcwTBPVkEkv4f5BZhSXCZwPs47IKjCgBT5uo+qdOKfBoQD/uMEMo541ZUS+cjpXxf4MFdHm32v/iAJNRZjefNwZdo/wkfcwTBN5oE+i67mv1Nzwa+pdgWEtcHHbVK+4TODUBRUxERJ6Nufzm0S+9Nnh8ibenQy8cUUB6+o4HdrGybhgE9iXofq7OBSkGKaJPNDQKiO5lxFnijTYHeF7F5OhLfBwm1RPCIFDWQqEABaNjYAs8+KeyJc+K2/iPaEba6UpsLGpN9UVwzSRh65ppQcEkJrLu5jkf3w++soy8jVcsAncN9CExFjWShP50gWbhm9PKjDqgNGdjP4uDtFluYfIOsowTbXDME3koTaNZESaJKTlaVBU1k6T/3AbXFmpU+DnM65Ox0a0Zy0Zka99fsQJTQAjOhgQzXHcKcANa2eAQQfsPK2goCR8H+Gj2mOYJvKQJEl4ZJhrmKyMMBwmqzKGOf9gbXTNCCFwKFuFUwWevDECEUZe2BP52mfsxZuCSKRJwuA2eqgasPVXxd/FoSDEME1UC92b6WDQASkXNGhhOExWZQx19Yt/75rLKNCQU6hhRj8TOjVl824iX7OVCWw67oQkATd3ZRNvCg6jOvC5aao9hmmiWjDqJfzPENcwWVkF4R2mAQa8+sK/c82VOASOZKswGyTc2Jk1ZET1YfNxJ0qdwMBWejSJ4iUmBQf3c9PHnBBhXkFCnuORjqiWBrbSw6ADjp9VoXF8QjY99iH+bT2jCYF9mQoU1dV7t4XNu4nqxaeH2MSbgk/35jo0jXb1hXP8bHg/vkeeY5gmqqUIo4S/DjXD7hBIy+fBtwJDn3fx7+m5385pyLUJ/HmwCW0bs3k3UX1wqgJf/OJqJntLEpt4U/CQJAkj2dSbaolhmqgO+rfUw6SXcOKcBpW1024MgHXH2ujaOVes4fg5FdFmCTd0YO0YUX357qSCPLtA13gdb2JR0OEQWVRbDNNEdWDUS5hzvRmlToHUXNZOV8YgWDsM0bVndwjsy1Cgl4GXJ1qg17F5N1F9+bS8F+8JbOJNQeiGDgZIEvDNSSdKnawcoZpjmCaqoz6JOkQYJZw4p6JM4QG4MoZCz/DvVXuKKrA7TYFDARaNtaCRlac3ovoihMBnh9nEm4JX40gZvRN0KHEC3/zG2mmqOV5tENWRXidhwSjXuNPHclR/FyfgsKb1yvg3qhtXh2MqCksF5g43o2MTNjElqk9701Vk5GtoESujZwK/fxScKkZ+qHj2n6gmGKaJvKBzUx1u72FEWp6GCzY2964Kw+KlGKLrTgiBn8+oyCnUcHd/E4a00fu7SERh57PDv/fiLUl8vIKC07gurlYVnx/hEFlUcwzTRF4gSRJu6mKALAOHszlUVnUYHl34d/CeUxc0pF7QEBcpYTwv5In84tPyJt4T2MSbglivBB3ioyWcztNwJJstDalmGKaJvKSRVcbD15lRVCqQws7ILitcgyRDtHdlFWj4JUdFpEnC67dboZMZpInq269nVfxyRkUjq4TBbBlCQUyWJYytqJ3+mU29qWYYpom8aHAbPaxGCcfOqiguY+305YRTsAynz1pfcoo07M9QYNZLeP12C8wGBmkif/jPEVcT73FdDOxBn4LeuC6u56Y/L9+via6EYZrIiww6Cc+Pj4AmgAOZCjQ+c3NFoRw0Q/mz+VOuXcPedAV6HfDa7RY0sPBURuQvnx5iL94UOoa3N8BiBHaeVpFVwFaGdGW8AiHyshaxOvxlsAl5doGT53kgrqlQCp6h9FkCTVGpwK7TCgDg5VutaBzJ0xiRv2QXaNh5WkGEwTVOL1GwsxgljOnk2pc/PcTaaboyXoUQ+cDw9gZEmiQcP6siv4SB2hPBGkQryh2MZQ8WRaUCP6YqUDRgyXgLmjfgKYzIn9YfcUAIYHQnAyxGNvGm0HBbd1cri48PMkzTlbGnCCIf0OskvDDBgtkf2rAvQ8WQNhIMfJbMI5VD6eyPbH4sSfUYnOtPRZB2qALPjbOgbWOOZUvkb5+V9+J9S1c28abQcVNnI0x6G749qeBskYaronjjlqrHvYPIR5pGy/jHDRGwlQkcylI5ZmEdBFKNL2ug619x2e9B+vmbLejUlEGayN8KSwW2nnBCJwNju7CJN4WOKLOEUR0N0MTvY6gTVYc100Q+1L+lDlP7GPHuHgcaWiS0bsQQUBd/DLD1UWPN0OxfhaUCOyvVSHdswu8QUSD4/IgDThW4vp0eDa2sm6HQMrGbEeuPOPHJQQfuG2j2d3EogDFME/mQJEkYn2TEp4ec+PmMihizxIsOL6ou6NYmZDM0B56CElfnRooKPDeONdJEgeSjA64auzt6sIk3hZ5xXQ0w6ICvTyjItWm8dqNqMUwT+ZjZIOHliRbc96ENe9JVDLlaQgTHxPUpBuPgl2fX8NNpBaoAltxiQTs+I00UMApKNGw86oQsAbd2Z5im0BNrkTG8nQEbjzmx/ogTd/U3+btIFKB4m4WoHlwVJePpmyJQpgjsTlOganx+mqg654s1/JiqQBPACwzSRAHn85+dcJQ38ebwdBSqbuvBXr3pyngEJKonXeL1+OtQMwpKBA6yQzKiKmUVuGqkJQl4eaIFbeIYpIkCzUf7XeHidjbxphA2vqsBOhn46rgTBRzmlKrBME1Uj4a102NybyMy8zWcOMcDM1Flp3NV7M1QYNBJWH6HFS1iGaSJAk2+XcOmY65evCd0Y5im0BUXKeO6q/VwqMAXPzv9XRwKUAzTVSgqKsLChQuRlJSEyMhIxMTEoG/fvli6dCkcDu829Xj22WchSZL7RaFNkiTc2t2ImAgJx8+qSM1V/V0kIr8TQuDEORWHslRYDRJWJFvQhON6EgWk9UfYxJvCx23lfQJ8cohNvalqPAr+wenTp9GtWzc88cQTOHLkCIQQKCsrw549ezBnzhxcc801yMvL88rvOn78OJ544gmvrIuCh0nvqnWzGiUczlaRXcgaagpfQgj8fEbFsRwV0WYJb06yItbCUxNRoFp3kL14U/iY0M0ISQI2HHWiuIyP59GleMVSiaqqGDduHFJTUxEfH4+vvvoKNpsNdrsdH3zwAaKiorB//35MmTKlzr9L0zTMnDkTpaWlGDBggBdKT8Ek0iThtdstMOqAfRmuYReIwo2qCezLUJFyQUMjq4S3JlkRaWILHaJAVbmJ9y1JDNMU+ppGyxjcWo9SpytQE/0Rw3Qlq1atwuHDhwEAn3zyCUaMGAEAkGUZd955J9544w0AwIYNG7B169Y6/a5XXnkFO3bswJQpUzBy5Mi6FZyCUkOrjFcnWiEB2J2u8I4nhZUyReDHVAVZBRqm9zPhzWQrzBwyjiig/eeIE04VGN5Ojzg28aYwMbG8qffHB9jUmy7FI2Elq1evBgAMGzasytri5ORktG7dGgCwZs2aWv+elJQUzJ8/H40aNcILL7xQ6/VQ8IuPkbF4vAUOFdh1WkGpk4GaQl9xmcCOUwry7AIPXmfGbd0N0OsYpIkC3bryMHFHT465S+GjIkx//rMDhaW8TqOLMUyXs9vt2LFjBwBgzJgxVS4jSRJGjx4NANi8eXOtf9esWbNgs9mwbNkyNG7cuNbrodDQrrEOC0ZFwOYQ2JWmwKnyQE2hK9euYUeKEzanwILREbihg4GdLxIFgTy7hs3HndDLwC1JBn8Xh6jeJDSQcX07PUqcrJ2mS+n9XYBAcfToUWia67nVrl27VrtcxbwzZ84gNzcXDRs29Oj3rFixAlu3bsWIESMwffr02he4CtnZ2VdcJiEhwau/k7yjTws9Hh1mxtJtpdiVpqB/Cz1r6ijk5BRp2JuuAACWjreg/VUc+ipYZWRkXHZ+Tk5OPZWE6st/DruaeI/qaEAjK+tiKLzc1c+Er08oWL27DPdcw5YZ9DuG6XJZWVnu6ebNm1e7XOV5WVlZHoXpzMxMzJ07FxEREe7nr72pX79+V1xGCNZ6Bqrr2upRppjw6ndl2J2uoF8LPXQyAzWFhvR8FQczVehl4OWJViQ04MV4MEtMTPR3EaiefXSAvXhT+JrQzQjLOhu+PakgPU9FYixvBpMLr2bKFRUVuactFku1y1WeV/k9NTF79mwUFBRg4cKFaNOmjeeFpJAmSRJGdjTgT4NMOF8ssDtdharx5gcFNyEEfj2r4kCGCrNewht3MkgTBZs8u4av2MSbwlikScL4rq4bSR/sZ1Nv+h1rpuvJu+++i//+97/o0aMHHnnkEZ/8jl27diE+Pt4n66b6IUkSbuxsgCqAFT+UYV8G0DtBB5k11BSENE3gULaK9DwN0WbX+OpRZu7LoSA9Pf2y8w8cOIBx48bVU2nI1z477ISiAWM6GdCQTbwpTE3pbcT7+xx4b68Dc6+P8HdxKEAwTJeLiopyT9vt9mqXqzyv8nsu5+zZs3j44Yeh0+mwYsUK6PW++bPHx8fzmegQIEkSxnY2QFGBlT+VYX8m0DNBB5mdNFEQcSgCu9MV5NoEknsZcVsPI0x67sOh4krnmpr04UHB46Pymrjb2cSbwtjIjgY0sko4kKnilzMqOjdlU29iM2+3Zs2auaczMzOrXa7yvMrvuZy//e1vuHDhAu677z507NgRxcXFF70cjt+bi1T1Mwo/sizhliQDpvU1IqtAw/4MFRqbfFOQKCwV+O6UK0j/zxATJvVikCYKVrk2DVt+dcKgYxNvCm8GnYTby4fJem9vmZ9LQ4GCYbpcp06dIMuuP8eRI0eqXa5iXtOmTWvc+VhKSgoA4PXXX0dUVNQlr0WLFrmXrfjZvHnzavtRKETIsoSJ3Y3uQL2XgZqCQE6Ra+irEqfAwjERGN3JyMcUiIJYRRPvGzoYEGvhZSOFt8m9y8P0Pgc79SUADNNuFosFgwYNAgBs3LixymWEENi0aRMAYOTIkfVWNgpfOlnCbd2NuKu/CWcKNezJYKdkFJiEEDh5XsWuNAUSgJdvtaB3Ip8kIgp27MWb6HeDWuvRIlZGygUNP6Yq/i4OBQCG6UpmzJgBANi2bRt++umnS+avW7cOp06dAgCPxojevn07hBDVvhYsWOBetuJnL774Yt0+DIUMWZYwIcmAe64xIadQYy/fFHBUTeBglusZskijhDeTrWjViM+SEQW7C5WaeI/vyibeRLIsYXIv142ltXv5SCYxTF9kxowZSEpKghACEydOxNatWwEAmqZh3bp1mDVrFgBgzJgxGD58+EXvXbVqFSRJgiRJ2L59e30XnUJcxTPUswaacK5Iw640hYGaAkKJU+CHFAXpeRqSexmxcrKVvf0ShYgP9jmgasCojgY0YBNvIgDAlD6uMP3RfgecKq/Fwh3b4FWi1+uxfv16DBs2DKmpqRgxYgQsFgs0TUNpaSkAoGfPnli7dq2fS0rhSJIkjOtigCwBb+wow94MFX0S2cs3+U9BiYZdaSpKnQIPDzVjWFs9n48mCiErd7k6Wbqrn8nPJSEKHF3j9ejWTIdDWSo2H3Pipi58BCKc8TbjH7Rq1QqHDh3C448/jq5du0KSJBgMBvTu3RtLlizBzp07ERsb6+9iUpiSJAk3df69yfehLJUdYJBfnCnUsCNFgUMVeGZsBIa3NzBIE4WQQ1kK9qaraGR13cglot9NKe+I7N09bOod7iTBK/GglpGRgcTERABAeno6x5kOE6om8OF+B97f60BirIxuzVhDTfVDCIFTFzT8kqPCqANenGBBYiyfj6aL7d69G/369QPAc1Ow+uunNrz4TRkevNaEl261+rs4RAElPU9Fy38VwKwHcp6MRZSZ12CBzleZiTXTREFIJ0u4o4cRk3obkZ6n4UCmCo33xcjHNE3gcPbFHY0xSBOFHoci3DVud7OJN9ElEmN1uO5qPUqcwKeHWDsdzhimiYKUXifhzp5GTOljRGa+hr3pHIeafMepCuxKU3A6V8MdPV0djXHMWaLQ9N9fnDhvE+jRXIceCexeh6gqU/u4bjS9s6fMzyUhf+KVEFEQ08kSbu9hxIx+5eNQc9gs8oHiMoHvTyk4Vyxw30ATJvc2wmxgkzaiULXyJ1c4YK00UfVu626A2QBsPaEgq0Dzd3HITximiYKcTpZwa7fyTsmKNOxOU6BwqAbyknPFGr4/5USxQ2D+yAiM7WKAjh2NEYWsM4UavjzqGlt6cm/2UkxUnZgIGTd3MUII4L29rJ0OVwzTRCGgYhzq+waacK5YYOdpBQ6FgZpqTwiBk+dV7DytAABeuMWCa1rpIbGjO6KQ9u6eMqgacHNXA+IieZlIdDnT+rpuOL3DXr3DFo+SRCFCkiSM7WLAg9eZkWcX+CFVQYmTgZo8p2oCBzJ/72hsRbIVbRuzozGiUCeEwMpd7HiMqKZGdTSgcaSEQ1kqDmYq/i4O+QHDNFEIkSQJN3Qw4O8jzCgqFdhxSkFRKQM11VypU+DHVAUZ+Rom9TZi1RQrGll5qiAKB7vTXDfR4qMljOrIsaWJrsSgkzC5l6t2+q2dbOodjniFRBSCBrUx4KmbIlCqCOxIceJcMTvGoCvLs2v47pSCPLvAg9eZkdzTCJOezbqJwkVFx2PT+pig1/G7T1QT9w6o6NXbAbuDFRjhhmGaKER1b67HixMsEAB+Oq0gLU/1d5EoQAkhkJqr4ocUBU5V4JmxEbihgwEyOxojChslDoH395U38e7PJt5ENdU1Xo+BrfXILxH4+ACfnQ43DNNEIaxNnA7/704rzAYJBzNV/HJGgSZ415R+p6gC+zNVHM5SYTZIWH6HFUnNOK4sUbj57LADBaUC17TUoWMT9pFA5InZ5bXT/+9HNvUONwzTRCGucaSMlZOsiLVIOHlew16ORU3l8uwavj2lIDNfw+TeRqyabEXTaJ4WiMLRyl3lY0uzVprIY7f3MKJBhIQdKQp+zmZHZOGEV01EYcBqkvD2JCvu7m/CmUINP6Zy6KxwpmkCR3NUfJ+ioMQh8NgNZiT3MiLCyGbdROEoLU/Fll8VRBiAO3tybGkiT0UYJUzr4/rusHY6vDBME4UJvU7ChG4GPHCta+is71MUFJcxUIcbm0NgR4qC386piDG7hr0a0NrA8aOJwtjqXQ4IAdzazYiYCF4aEtXGfQNdrTrW7HaghB2RhQ0eMYnCiCRJGNnRgMdHR8DmEPj+FHv6DieZ+Rq+PelEfqnAQ9eZsXqKFVdF8TRAFM40TWAVm3gT1dlFHZEdZEdk4YJXUURhqG8LV0/fALDztIJTF1QIdkwWshyKwN50BfsyFOgkCS9OsGBEBwN07K2bKOxt/03BqQsaWsbKGNaWnQ8S1cV97Igs7DBME4Wpq+N0WJFsRaRRws/ZKvZnsmOyUJRTpOGbkwqyCjRM6WPE6ilWXB3HnnqJyOWlb0sBALMGmDgcHlEdVXRE9v0pBQcz2RFZOGCYJgpjjawyVk62YnJvIzLzNXx/is9RhwqnKnAgU8Gu0woUTeBfN0bgzp5GWE28WCYil9/Oqfj8ZyfMBmD2QDbxJqori1HCzGtc36WXvin1c2moPjBME4U5s0FCci8j5g43o7BM4NuTTqTmstl3MDtTqGH7bwrS8zTc0dM15FXPBD07GSOii7zyXSmEAKb1MSEukpeERN7wl8EmyBLw3j4HzhaxX5pQxyMnEUGSJFx7tQGvTrTAqJNwOEvFrjQVZRw+K6iUOAR2pynYnaZA1QSeGBOBqX3YOy8RXaqgRMPbP7me63zoWtZKE3lLq0Y63JJkQJkCvPEDn50OdbzCIiK3lg11WDPVipnXmHC2SMM3vym8qxoEVE3g13Mqtv3mxJlCDdP7mbB6SiR6JbI2moiq9vZPDhSXATd00KNLPDseI/Kmh64zAwBe21EKBysmQhrDNBFdxGyQcEs3I54ZGwFVCPx0WsHhbIWdkwUgIQSyClxNuo/nqDAbJLwwwYLbexgRZWaIJqKqqZrAy+Udjz1cftFPRN4zpI0ePZrrcKZQ4KMDHCYrlDFME1GVkprpsXKSFQ2tElIvuHqEzrWxljpQXLBp2JGiYG+6Aocq8PcRZqydZkXbxuypm4gub/0RJ1JzNbRvLGN0R4O/i0MUciRJct+oeumbUvZDE8IYpomoWg0sMlZOsuKxG8wocQrsSFVwKEuBU+VJwV/ySzT8dFrBDykK8ksE7h9swrtTIzGoDceNJqKaebG8l+GHrjNzOCwiH0nuZcRVkRL2pKv4MZXDZIUqhmkiuixZljCgtQFvJVvRyCLhdK6Gr084kZ7HHr/rU0GJhl1pCr47qeBssYYZ/UxYNdmKGztzuCsiqrl96Qq+PamgQYSE6X3Z8RiRr5j0Ev40yPUde/EbdkQWqtjjBBHVSFyka0zqg5kqntxcggOZKtLyNHSJ16EBe4v2mVy7ht/Oacgp7whuah8jhrYzoEkU/+ZE5LmKWulZA0yI5I04Ip+6f5AZi7aU4t+HHEjLU9Eilo9ihRpejRFRjUmShB4JeqydFon7B5uQWyLw3SkFBzIV2B2spfYWIQTOFWv4IcWJHacU5BRrmNbXiDcnWXFnLxODNBHVSnaBhg/2O6CTgb8MYa00ka81jZaR3NMIVQNe/pa106GINdNE5DGzQcKNnY3ok6jHpmNOfLTfgcx8DS0bymgbp4PZwNqO2tCEQHahwKnzKvJLBCQJmHmNCYPb6BEXyQBNRHXz+o5SOFXg9h5G1pAR1ZNHhprxzh4Hlu8oxT9GmNHIyvN5KGGYJqJauypKxrS+JgxqrcfjG0qQckFDWp6G1o10aNNIhknPUF0TTlUgLU9DSq6GEoeALAN/GuT6uzaw8KRLRHVndwgs/8FVM/bwdayVJqovPRL0GNvFgC9+duKlb0rxrxst/i4SeRHDNBHVWZs4Hd6ZasXxsxqe2FiC386pSLmgomVDGW0a6hBhZKiuSmGpQEquisx8DaoGGPXAo8PM6NdSDwv/ZkTkRW/+WIZzxQLXtNRhQCte/hHVp/+9IQJf/OzEy9+W4dFhZsSwr5mQwaMpEXmFJEno2ESHtdOsOJqj4pmvSnHqvIaUCxqaxcho1VBGbIQESQrvkKioAtmFGk7nacizu54zj4mQMGeYGV3jddDrwvvvQ0TeV6YIPP91CQBg/siIsD8OE9W3/q30uKGDHl8dV/DKd2X435ER/i4SeQnDNBF5lSxL6BKvx7vTrDh5XsOOFAUfH3A9Ux1tltAiVkZCAxmGMAqNmhC4YBPIyNeQXeiqhZZl1/PQfVro0TyGNxmIyHfW7C5DZoFA92Y63NTZ4O/iEIWl/x0Zga+OF2HZ9lI8MMTE2ukQwTBNRD4hSRLaNtahbWMdxnQyYHeagjd3luFItopfclQ0iZKRECOjcaQEnRx6QVLTBM7bBLKLNJwp1OBQXD9vECHhwevM6NqUzd+JyPcUVWDRFtdwWP/LWmkiv7n2agOGt9dj668KXv62DP8cxdrpUMAwTUQ+d1WUjJu6GHFDBwN+PqPiYKaKTw46kF2gwaBzDR0RHy0jzhrcwbrU6RrS6myxwNliDYrq+nmkScIDQ0xIaqZjL55EVK/e3+dAygUNnZrIuLUba6WJ/OmJ0RHY+msRlm5z1U6zk9HgxzBNRPXGqJfQM0GPngl6TOjmCtb/910Z0vM0pOdp0MlAXKSMqyIlxFllWI0I6FoUhyKQaxc4b9NwwSZQWPr7WNuxFgmzBpjQqYmOw1oRkV9omsAzX7melX7shgjIQXyzkigUDGpjwMgOemw+ruCFb0rxxBj27B3sGKaJyC9iImQMbC1jYGsDLtg0HM1R8etZDZ8ddiCnEABUmPQSYi2uVwOzhJgIyW/PWquaQFGZQH6JQEGJQF6J6/8oz88mvYS7+pvQNk7G1XE6RJp40UpE/vXvQ04cO6uhTSMZyT2N/i4OEQF4YowFm48X4oXtpXjoWjMassVaUGOYvoKioiIsXboUn3zyCVJSUqDT6dC+fXskJyfjgQcegNHo2cnpwoULWL9+PbZu3Yp9+/bh9OnTUBQFjRs3Rp8+fTBjxgxMmDDBR5+GKDA1ssoY3EbG4DbA5N5GpOZqSLmg4t09Dpwp0nCm8PdlIwwSos0SIk2A1SjBYpQQYZAQYUCdm4irmkCJEyhxCpQ4BWwOoLhMuF6O34NzRTlm9DWhZUNXT+VxVnYiRkSBQ9MEntzsqpX++wgzRwogChDXtNLjxs4GfPmLE8u2l+Kpm1g7HcwkIYS48mLh6fTp0xg6dChSU1MBABaLBaqqoqysDADQs2dPbN26FbGxsTVep8FggKIo7v+bzWbodDrYbDb3z8aMGYOPP/4YFsuVv1wZGRlITEwEAKSnpyMhIaHGZSEKBg5FIKtAQ1ahhuxCgX8fdKC4TECr4sil1wEmnQSDzjWtlyXIEiBLQEXO1QQgBKBqgKIJKBrgVF1Dx6haFQWQXKF9QpIBTaJcz3bHx8iINvPClKg6u3fvRr9+/QDw3OQvH+0vw52rbWgZK+PX+TEw6nnMIgoUe9IU9F1WiEgTkPLPBnwcrB74KjOxZroaqqpi3LhxSE1NRXx8PNasWYMRI0ZA0zSsW7cOs2bNwv79+zFlyhR8+eWXNV6voijo168f7rrrLowaNQpt2rQBAKSmpuKpp57CW2+9hQ0bNmD27Nl45513fPXxiIKGUS+hVSMdWjXSAQAmdjdC01zNrS/YXf/ml7hqj4tKBexOga2/KlAdAk7t4trkymQZ0EmAQSdhWDs9rEYJVqOr1jvK9Hvz8tgIiTU6RBRUVE1gwQZXrfTjoyIYpIkCTJ8Wetzc1YD1R5xYsq0Uz45j7XSwYpiuxqpVq3D48GEAwCeffIIBAwYAAGRZxp133glN0zB58mRs2LABW7duxfDhw2u03q+//hrDhg275OetWrXCm2++Cb1ejzfeeAPvvvsunnnmGfcdFCL6nSxLaGiV0NBa9fw/Dfp9WiuvfQZcuVqWXCGaHfEQUah6b68Dx85qaBsnY3pfPitNFIgWjo7A+iNOvPJdKR4ZasZVUaydDkbcatVYvXo1AGDYsGHuIF1ZcnIyWrduDQBYs2ZNjddbVZCubObMme7pPXv21Hi9RFQ1WZZg1LteJr2rAzMGaSIKVU5VYOFGV630wtERbFlDFKAqRjaxO+DudZ+CD8N0Fex2O3bs2AHA9fxyVSRJwujRowEAmzdv9trvNpvN7mlVVb22XiIiIgp9r39fhlMXNHRuqkNyL9ZKEwWyf42JgE4GXv2+DD9nK1d+AwUcNvOuwtGjR6FprnahXbt2rXa5inlnzpxBbm4uGjZsWOffvX37dvd0UlKSR+/Nzs6+4jLsBIaIiOoqIyPjsvNzcnLqqSRU2fliDQvKa6UX3xxR5xEOiMi3usbr8T+DTXj52zI8+G87tvw5iiODBBmG6SpkZWW5p5s3b17tcpXnZWVl1TlM5+fnY9GiRQCAIUOGoEOHDh69v6Ln1Mth5+1ERFRX7M8jMD2+oQT5JQI3djbgxs6slSYKBk+MjsD7+xz4+oSCTw46cVsPfneDCZt5V6GoqMg9fbnhqSrPq/ye2tA0DdOmTUN2djZMJhNeeeWVOq2PiIiIwsfBTAVv/FAGvQwsG8+egYmCRQOLjEVjXd/ZRz6zw+5gxVcwYc10gHjooYfwxRdfAABee+01dO/e3eN17Nq1C/Hx8d4uGhER0UXS09MvO//AgQMYN25cPZWGhBB4+FM7NAE8fJ0ZHZro/F0kIvLA3f2MeOOHUuxOU7FoSwmevJE3xIIFw3QVoqKi3NN2u73a5SrPq/weT82ZMwevvvoqAOCFF17APffcU6v1xMfH85loIiLyuSuda2rShwd5z78PObH9NwWNIyU8Psp85TcQUUCRZQmvTrSi/wuFWPx1Ke7qZ8LVcbwpFgzYzLsKzZo1c09nZmZWu1zleZXf44l58+Zh6dKlAIDFixfj4YcfrtV6iIiIKPyUOAQe/cx1c/+ZmyIQE8FLO6Jg1K+lHvf0N6JMAf76afWVeRRYeMStQqdOnSDLrj/NkSNHql2uYl7Tpk1r1fnY3LlzsXjxYgDA888/jzlz5tSitERERBSulmwrxek8DT0TdLi7v8nfxSGiOlg01oIYs4TPf3Ziwy8OfxeHaoBhugoWiwWDBg0CAGzcuLHKZYQQ2LRpEwBg5MiRHv+OOXPmYMmSJQBcQXru3Lm1LC0RERGFo1PnVTyzxTUU1su3WjgUFlGQuypKxr/GRAAAHvy3HSXsjCzgMUxXY8aMGQCAbdu24aeffrpk/rp163Dq1CkAwPTp0z1a95w5c9xNu5csWcIgTURERB4RQuB/Praj1AnM6GvE4DYGfxeJiLzgz4NN6N5Mh9/Oa3h8Q4m/i0NXwDBdjRkzZiApKQlCCEycOBFbt24F4BrCat26dZg1axYAYMyYMRg+fPhF7121ahUkSYIkSdi+fftF8/72t7+5g/SyZcvw6KOP+v7DEBERUUj5YJ8DG4850dAiYTGHwiIKGXqdhLcnWaGTgWXbS7EzVfF3kegyGKarodfrsX79erRq1QqZmZkYMWIErFYrrFYr7rjjDhQWFqJnz55Yu3ZtjdeZlpaG559/HgAgyzKee+45NG3atNpXRTNwIiIiogppeSruX+fqoGjpeAsaR/JyjiiU9ErU4x8jzNAEcPd7xSh1srl3oOLQWJfRqlUrHDp0CEuWLMG///1vpKSkwGAwoEuXLpg0aRIeeOABGI3GGq9P07SLpnNyci67fHFxca3LTkRERKFH1QSmv2tDQanArd0MmNGv5tchRBQ8/ndkBD477MSRbBULN5bg2XFsgRKIJCEEb3UEsYyMDCQmJgIA0tPTOc40ERH53e7du9GvXz8APDd523NbSvD3L0rQLEbCoXkxaGRlrTRRqNqTpuCaFwshBLDzr9Ho24L1oLXlq8zEIzARERFRENibruB/v3R1SLRqciSDNFGI69NCj3nXu5p73/WeDWUK60ADDY/CRERERAHO7hCY8k4xFA14ZKgZN3Rg791E4eDxURHo1ETGL2dUPLmJvXsHGoZpIiIiogD36Gd2HD+roVszHZ4ZG+Hv4hBRPTEbJKycFAlZAp7dWoqf2Lt3QGGYJiIiIgpg6484sPyHMpj0wHvTrDDpJX8XiYjqUf9Wesy93gxVA+5YXYwLNu3Kb6J6wTBNREREFKDOFGqY+b4NALD4Zgu6xLMDIqJw9OSNERjUWo+0PA1T37FB0/j8dCBgmCYiIiIKQEII3P2eDedtAqM7GvCXISZ/F4mI/MSgk/DhjEhcFSlh4zEnntpc6u8iERimiYiIiALSq9+VYeMxJ+KsElZOtkKS2LybKJw1byDj/emu56cXbirB5mNOfxcp7DFMExEREQWYb0868eh/7ACAtyZZ0TSal2xEBFzf3oAnb4yAEMDkd4px8rzq7yKFNR6ZiYiIiAJIygUVE98uhlMF5l5vxs1djf4uEhEFkL8PN+PmrgZcsAmM/X9FyLOzQzJ/YZgmIiIiChBFpQI3v1mM8zaBsV0MWMRhsIjoD2RZwtppkejRXIdjZzXctrIYTpUdkvkDwzQRERFRAChTBO5YVYwj2Sq6NNVh7bRI6GQ+J01El4o0Sfh8VhTioyV8fULB/evsEIKBur4xTBMRERH5maIKTFpTjI3HnLgqUsL6eyMRbWaQJqLqJTSQ8fmsKFiMwFs7y/DsFvbwXd8YpomIiIj8SNUEZrxnw6eHnGhokbDlz1FoE6fzd7GIKAj0TtTjvWmRkCTgsf+W4MFPbKyhrkcM00RERER+omkCsz+y4729DkSbJWy+PwpJzfT+LhYRBZHxSUasmWKF2QC88l0Z5q0vYaCuJwzTRERERH6gagL3fWTHWzvLYDUCG2ZHoncigzQReW5qHxM+vScSBh2wZFsp5v+Xgbo+MEwTERER1TOnKjD1HZs7SH8xKwoDWxv8XSwiCmKjOxmx7q5I6GVg0ZZSPPYFA7WvMUwTERER1aNSp8BtK4vxwX4HYsqbdg9txyBNRHU3PsmIj+921VA/u7UUc9nk26cYpomIiIjqia1MYNyKIqw/4kScVcK2v7BGmoi8q3KgXrqtFHe/Z+M41D7CME1ERERUDwpKNIxaXoQtvyqIj5bwzQPR6JnAZ6SJyPtu7mrEhtlRiDQBq3c7MOGtYtgdDNTexjBNRERE5GNnCjUMf60IO1IUtIyV8d2D0ejclMNfEZHvDG9vwPa/RKNxpIT//uLEiNeKkGvT/F2skMIwTURERORD+9IV9F1WiL3pKto3lvHdg1G4muNIE1E96J2ox46HotGqoYwfUxUMfKkQJ86p/i5WyGCYJiIiIvKRD/eVYfDLhcjI13Dd1a6L2sRYBmkiqj/tGuvww0PR6NFch+NnNfRbVoivjjv9XayQwDBNRERE5GWKKvDYF3Ykr7GhxAnMHmjC5vujEBfJSy8iqn/xMTK+fzAaE7sbkF8iMOaNIizdVgJN43PUdcEjOhEREZEXnTyvYsjLhVi0pRQ6Gfi/2yxYfocVRr3k76IRURizmiR8NCMSC0dHQNWAOf8pwdgVxThbxOeoa4thmoiIiMgLhBBYs6sMPRYXYOdpFS1iZWz/SxT+PNjs76IREQEAZFnCgtER+GJWJOKsEjYcdaLb8wXYfIzNvmuDYZqIiIiojnJtGiatsWHGezYUlwGTehlxcG40BrfhGNJEFHhu6mLEoXkxGNFej5wigVHLi/DoZ3aUcPgsjzBMExEREdVSRW10h2cK8OF+B6JMwDtTrVg7zYoGFl5mEVHgio+RselPUXhuXAT0MrBseym6Ly7AdydZS11TPMoTERER1cKxHBXX/18RZrxnw3mbwPD2ehyYG4OpfUyQJD4fTUSBT5YlzBsegV2PRKN7Mx1OnNNw7StFmPZuMdLyOITWlTBMExEREXmgxCHwzy/t6PZ8Abb/puCqSAlrp1nx1f1RaMPxo4koCPVM0GPXI9H415gImPTAu3scaP90Af7+uR0FJeygrDoM00REREQ1oGkC7+8tQ6dFBXhqcykUDfjzYBOOPxaDyb1ZG01Ewc2ol/DPURH4dX4MpvUxokwBnttairZPFeDV70rhVPk89R8xTBMRERFdhhACX//qxDUvFmLyOzacztPQM0GHHx+Kxv/dxmejiSi0tIjVYc3USOx9NBrD2ulx3ibwwCd2dHm2AO/uKWOoroRHfyIiIqIqCCHw358dGPRSEYa/VoTdaSqax0hYPdmKPY9Eo38rvb+LSETkM70S9dj65yh8MSsSnZu6nqee9q4NVz9ZgBe3l6K4jKGaZwEiIiKiSsoUgX8fdOD5r0txINPVAU/jSAmPDjXjgWvNsBjZnJuIwoMkSbipixGjOhrw/j7XcfFItoq/fmbHvzaX4M+DTHjgWjOaRIVnHS3DNBERERGA386p+H8/lmHlT2U4b3PVuDSPkTDv+gjcO8DEEE1EYUuvkzCtrwlT+xix8agTz39diu2/KXj6q1Is2VaKu/qZcP8gE7o3D694GZ63EIjKZWRkQJIkSJKEjIwMfxeHAgD3Cfoj7hOhzakKfHzAgRGvFaLd0wVY/HUpztsEeiXosOJOC07+swEevM5VG819IXhwWwUXbq/gkZmZiRu7mLD9gRisn1KK27ob4FCBN34oQ4/Fhej2XAEWf12CzPzw6AE8vG4dEBERUdgTQmB/hop1BxxYuasMOUWuWmirEZjUy4TZA03o04KXSEREl9MzXsO4vlE4cU7F/31Xivf2OXA4W8W89SX42+clGN5Oj+l9TZjQzYhIU2i27OGZgoiIiEKeUxX45jcFnx12YP0RJ9Ir1Zp0b6bD7IEmTOljQrQ5NC/4iIh8pV1jHV681YrF4y3YfMyJd/Y48NlhB7b8qmDLrwrMH9lwfTsDburserVsqPN3kb2GYZqIiIhCjhACR3M0bP3ViS2/OrH9NwWFpb/3PJvYQMb4JAOm9Dahf0sdx4gmIqojg87VWdlNXYwoKNHw8UEn3tldhm9PKfjyFye+/MWJ/wHQNV6HkR0MGNpWjyFt9EE9vCDDNBEREQW9MkXgQIaKH1MV7Dyt4NuTTmQXXjxsS7dmOtySZMD4rkb0TGCAJiLylZgIGTOvMWHmNSacK9aw4Rcn/vuLE5uOOXEkW8WRbBXLtgOSBPRorsO1bfS4ppUe/Vro0bqRHDTHZ4ZpIiIiCiq2MoEj2SoOZSs4mKlib7qCfRkqHOrFyzWPkTC8vQHD2xkwvL0BzRsEb+0HEVGwahwpY3o/E6b3M8GpCvyYqmD7CQXbTzrxY6qC/Rkq9meoeOnbMgBAI6uEbs10SIp3vbrG69GlqQ5RAfgYDsM0ERERBRxFFUjL03Dygobfzqnuf38+45oWF1c6Q5ZcNc/XtNRjQCvXq/1VwVO7QUQUDgw6CddebcC1VxvwOCJQpgjsOq3g+1MKdqcr2HVaQWaBwLYTCradUC56b3y0hLZxOlwdJ7v/bdNIh4QGMq6KlKDX1f/xnmGaiIiIfE4IgaIyIM+uIb9EuF95doEzRRqyCjRkFWrIKhDILtSQka9BqWZkFZMe6NJUh27NdOjWTI8ezXXok6gPyFoLIiKqnkkvYcjVBgy52uD+WXaBhsPZKg5nKzicpeJwtopfclRkFwpkFyr47tSl65EloEmUhGYxMppFy2gSJaOhRUJDi4RYiwRR5PBJ+Rmmg5yi/H7HJjs7248lCU6V/2b8+xHAfYIuxX3Cc1lZWe7p/k8eh82Qj8JScUlt8mVJQPNoGS0bymjVUIeWDSW0aqhD2zgd2jSSL6mBKDgPFHip/NXhvhA8uK2CC7dX8KivbdU5EujcDriznev/miaQUySQmqsiNU9D6gUNp/M0pOepyCnWcLZIILsQyM4E9la1QtsZ92Tl/FRXkhAendoowOzevRv9+vXzdzGIiIiIiIgC3q5du9C3b1+vrIs9cRARERERERF5iDXTQa60tBSHDx8GADRu3Bh6PVvuExGRfxUXF2Pnzp0AgAEDBsBqtfq5REREFM4URcG5c+cAAElJSTCbzV5ZL8M0ERERERERkYfYzJuIiIiIiIjIQwzTRERERERERB5imCYiIiIiIiLyEMM0ERERERERkYcYpomIiIiIiIg8xDBNRERERERE5CGGaSIiIiIiIiIPMUwTEREREREReYhhmoJGUVERFi5ciKSkJERGRiImJgZ9+/bF0qVL4XA46rTuHTt2IDk5GYmJiTCZTGjcuDGuv/56rF69GkKIat+3atUqSJJ0xdeWLVvqVD66mN1ux4YNG/DUU0/h1ltvRcuWLd1/64ULF3rld9R1f/Pl/kqXCuR9gseJwOXt7+mFCxewcuVKTJ06FZ07d4bVaoXJZEJCQgJuueUWfPrppz74FKHP18fTwsJCPPfccxg4cCAaN27s3mbDhg3DwoULkZ+fX/cPEQYC+ThMF6uPbfVHzz777EXnvJAhiIJAamqqaNWqlQAgAAiLxSJMJpP7/z179hS5ubm1Wvf8+fPd6wEgGjRoIIxGo/v/o0ePFqWlpVW+d+XKlQKAkGVZNGnSpNrXt99+W5ePT3+wbdu2i7ZZ5deCBQvqvP667m++3F+paoG8T/A4EZh88T3V6/UX7Xtms1lYrdaLfjZmzBhhs9l89KlCj6+Pp19//bVo0qSJe316vV40aNDgom22f/9+732gEBbIx2G6mK+31R8dO3ZMmM3mi35PqGDNNAU8VVUxbtw4pKamIj4+Hl999RVsNhvsdjs++OADREVFYf/+/ZgyZYrH637zzTfx9NNPAwCSk5ORnp6OvLw8FBUV4d1330VUVBQ2btyIBx544LLrSUxMxJkzZ6p9DRkypFafnaoXGxuL4cOHY+7cuXj//ffRtGlTr6y3rvubL/dXurxA3Scq8DgROHz1PVUUBf369cNrr72GkydPoqSkBMXFxUhJScHMmTMBABs2bMDs2bN98bFCjq+Ppzt27MBNN92EnJwcjBgxAt9//z3KysqQl5cHu92OPXv2YP78+YiJifHyJwtdgX4cpt/5alv9kaZpmDlzJkpLSzFgwACf/A6/8neaJ7qSN998030X64cffrhk/nvvveeev2XLlhqvV1EU0bRpUwFA9OrVS2iadskylWuUDh06VO38li1bevSZqG4URbnkZy1btvTKHdW67m++2l/p8gJ5n+BxIvD46nv69ddfX3b+7Nmz3etNS0vzuNzhxpfHU5vNJtq0aSMAiIkTJwpVVb1V7LAVyMdhupgvt9UfvfjiiwKAmDJliliwYAFrponq2+rVqwEAw4YNq/KOVnJyMlq3bg0AWLNmTY3Xu2fPHpw5cwYA8Oijj1b5/MaMGTPQpEkTaJrmLgf5n06n89m667q/+Wp/pcsL5H2CAo+vtumwYcMuO7+idhpwnYPo8nz53XvnnXdw6tQpREREYPny5ZBlXhLXFY/DwcOX26qylJQUzJ8/H40aNcILL7xQL7+zvvHIQQHNbrdjx44dAIAxY8ZUuYwkSRg9ejQAYPPmzTVe9+nTp93TnTt3rnbdHTp0AABs3Lixxuum4FTX/c2X+yv5B7dp6PHnNjWbze5pVVW9tt5Q5OvtVBG4xo8fj7i4uDqUlHyNx+HgNWvWLNhsNixbtgyNGzf2d3F8gmGaAtrRo0ehaRoAoGvXrtUuVzHvzJkzyM3N9fj3XO6ipmLe8ePHq+0p8ty5c+jduzciIyMRERGBNm3aYOrUqdi+fbvHZSH/qev+Vl/7K9Ufb25THicCgz+/p5W3dVJSklfWGap8uZ3KysrcLQOuu+46nDp1CjNnzkRCQgJMJhOaNm2K8ePHY8OGDXX8FOQNPLcGpxUrVmDr1q0YMWIEpk+f7u/i+AzDNAW0rKws93Tz5s2rXa7yvMrvuZxWrVq5p48cOVLlMoqi4Pjx4+7pc+fOVbmc3W7Hvn37YDQaoWkaUlJSsHbtWgwbNgz33HMPFEWpUZnIv+q6v/lyfyX/8OY25XEiMPjre5qfn49FixYBAIYMGeJu9URV8+V2Sk1Ndd8cz8jIQLdu3fD222/j3LlzsFgsyMnJwfr163HjjTfi/vvvr+UnIG/huTX4ZGZmYu7cuYiIiMAbb7zh7+L4FMM0BbSioiL3tMViqXa5yvMqv+dyevfu7e658LnnnqvyQvaNN97A+fPn3f8vLCy8aH6zZs2wYMECHDx4EKWlpcjNzXU3RxoxYgQAYOXKlfjrX/9aozKRf9V1f/Pl/kr+4Y1tyuNEYPHH91TTNEybNg3Z2dkwmUx45ZVX6rS+cODL7ZSXl+eeXrRoEQwGA95//30UFxcjLy8PaWlpSE5OBgAsX74cL730kqfFJy/iuTX4zJ49GwUFBVi4cCHatGnj7+L4FMM0hS2dTucemP7o0aO46aabsHfvXjgcDuTk5OCFF17Ao48+CoPB4H7PHzsoGTlyJBYuXIhu3brBZDK51ztw4EBs2rQJ48ePBwC89tprOHHiRP18MCIKKDxO0EMPPYQvvvgCgGs7d+/e3c8lCm8VTYYrppcvX47k5GT3+T4xMRFr165Fz549AQBPPfUUW44Q1dC7776L//73v+jRowceeeQRfxfH5ximKaBFRUW5p+12e7XLVZ5X+T1XMnv2bPz9738H4Oqwok+fPu7npR555BE0aNAAjz32mHv52NjYGq9blmUsWbIEgOtk/fnnn9f4veQfdd3ffL2/Uv3z9TblcaL+1ff3dM6cOXj11VcBAC+88ALuueeeWq8rnPhyO1VeLjExEXfeeecly8iyjEcffRQAcP78eezdu7dG6ybv47k1eJw9exYPP/wwdDodVqxYAb1e7+8i+RzDNAW0Zs2auaczMzOrXa7yvMrvqYlFixZh586dmDlzJpKSkpCYmIjevXvjsccew5EjR9w1SbGxsR73RNi2bVt3L6GnTp3y6L1U/+q6v9XH/kr1qz62KY8T9as+v6fz5s3D0qVLAQCLFy/Gww8/XKv1hCNfbqfKz9Z27Nix2uU6derknq48AgjVL55bg8ff/vY3XLhwAffddx86duyI4uLii16VO/Kt6mfBiGGaAlqnTp3cTaur6ySs8rymTZuiYcOGHv+e/v37480338ShQ4eQlpaGPXv24Omnn0ZcXBy+/fZbAMCAAQOqHIuaQkdd97f62l+p/nCbhp762qZz587F4sWLAQDPP/885syZU4vShi9fbqeGDRu6A/XlzutCCPc0z//+w+Nw8EhJSQEAvP7664iKirrkVdEJIwD3z+bNm+ev4noFwzQFNIvFgkGDBgGofpxnIQQ2bdoEwPVsojelpaXhq6++AgDMmDHD4/efPHnS3YFZ69atvVo28r667m/+3l/J++pjm/I4Ub/qY5vOmTPH3Xz/+eefx9y5c2tZ2vDl6+1UsfzRo0cvCs2VHT161D3N76b/8NxKgYxhmgJeRYjdtm0bfvrpp0vmr1u3zt000pvj2DmdTtx3331QVRVdu3bFhAkTLppf3cm38vyKCyhZljF27FivlY18p677m7/2V/KdumxTHicCky+/p3PmzHE37V6yZAmDdB34cjvdfffdAID09HR8+OGHl8zXNA3Lli0D4GoW3qtXL4/WT97Fc2tw2L59O4QQ1b4WLFjgXrbiZy+++KL/CuwNgijAOZ1OkZSUJACI5s2biy1btgghhFBVVXz00UciOjpaABBjxoy55L0rV64UAAQAsW3btkvmnzx5UsyfP1/s3btXlJSUCCGEUBRFbN++XQwePFgAEJGRkWLfvn2XvDclJUX07dtXLF++XJw8eVJomuYu148//ihGjRrl/t3333+/F/8iJIQQubm54ty5c+5XYmKiACDmzp170c+Lioouet+V9om67G/eeD/VXiDuEzxOBCZfnVfmzZvnnrds2bL6+CghzZfnfyGEuO222wQA0aBBA/HBBx8Ih8MhhBAiLS1NJCcnu9+/atUqn33GUBOIx2Gqmq+21eUsWLDA/d5QETqfhEJaSkqKaNWqlfsLaLFYhNlsdv+/Z8+eIjc395L3XekLv3//fvd8SZJEbGys0Ov17p81a9ZM/PDDD9WWqWI5AMJkMom4uDhhMpku+vndd98tnE6nt/8kYa9ly5YX/Z2re82YMeOi99XkJFDb/c1b76faCcR9gseJwOXt88rp06fdP5dlWTRp0uSyr8WLF9fjpw1evjr/CyFEcXGxuPbaay/6fsbGxl703Xz88cd9/AlDSyAeh6lqvtxW1QnFMM1m3hQUWrVqhUOHDuHxxx9H165dIUkSDAYDevfujSVLlmDnzp0eDVtVeb2PP/44rr32WsTHx8NmsyEmJgYDBgzAc889h+PHj2PAgAFVvrdJkyZ45ZVXMHnyZHTu3BnR0dHIz8+HwWBAx44dcc899+D777/H22+/HRZDA4SSuu5vvtpfyX9qu015nAhc3v6e/nHs4pycnMu+iouLffGxQo4vj6dWqxXbtm3DihUrcO2118JqtaK4uBjNmzdHcnIyduzYgSeeeMLLn4hqi+dWCkSSEFd4oIuIiIiIiIiILsKaaSIiIiIiIiIPMUwTEREREREReYhhmoiIiIiIiMhDDNNEREREREREHmKYJiIiIiIiIvIQwzQRERERERGRhximiYiIiIiIiDzEME1ERERERETkIYZpIiIiIiIiIg8xTBMRERERERF5iGGaiIiIiIiIyEMM00REREREREQeYpgmIiIiIiIi8hDDNBEREREREZGHGKaJiIiIiIiIPMQwTUREREREROQhhmkiIiIiIiIiDzFMExEREREREXmIYZqIiIiIiIjIQwzTRERERERERB5imCYiIiIiIiLyEMM0ERERERERkYcYpomIiIiIApAQAl9++SUmTZqEdu3awWKxID4+HkOHDsWaNWugKIq/i0gU1iQhhPB3IYiIiIiI6HcnT57EzJkz8c0331S7zOjRo/Hxxx/DarXWY8mIqALDNBERERFRANm/fz+GDx+OvLw8AECjRo0wdOhQREVFYc+ePThy5Ih72XvvvRcrVqzwV1GJwhrDNBERERFRgDh+/DgGDBiAvLw86HQ6/POf/8Tf//53mEwm9zJLlizB3LlzAQCSJGH//v3o3r27v4pMFLb4zDQRERERUQBwOp2YPHmyu0b6jTfewIIFCy4K0gAwZ84cjBgxAoDruer33nuv3stKRAzTREREREQB4dlnn8W+ffsAAHfddRdmzpxZ7bL33HOPe3rjxo0+LxsRXYrNvImIiIiI/CwvLw+tW7dGQUEBzGYzUlNT0aRJk2qX37t3L/r06QMAiIyMRFFRUX0VlYjKsWaaiMLeO++8A0mSIEkSpk6dCsDV1O69997D2LFj0bJlSxgMBkiShA8//NDPpSUiolD00ksvoaCgAICrVvpyQRoAoqKi3NPFxcUoLCz0afmI6FJ6fxeAiMjfDhw44J7u3r079u7di7vvvhuHDx++ZFl28EJERN4mhMCqVavc/7/33nuv+J4/jjEty6wjI6pvDNNEFPYqh+nCwkJcd911sNlssFqtGDp0KOLj43Hu3Dns3r0b7du3919BiYgoJH3//fc4ffo0AKB9+/bo3bv3Fd/zx2bdZrPZJ2UjouoxTBNR2Dt48KB7+plnnoEQAvPmzcM///lPREZGuuepqso7/0RE5HX/+c9/3NM333xzjd5z4cIF93RMTAz0el7WE9U3dkBGRGEtPT0dLVq0uOhnq1atwowZM/xUIiIiCjd9+/bFnj17av3+Pn36YPfu3V4sERHVBKtYiCisVW7iDQAPPfQQgzQREdWb4uJi7N+/v07raNeunZdKQ0SeYHsQIgprlcN0o0aN8K9//ct/hSEiorBz8OBBqKoKALjqqqtw++231+h9q1evRnFxMQCgV69ePisfEVWPYZqIwlrlMD1jxgxER0f7rzBERBR2Tpw44Z7u3bs3Xn311Su+Jzc3F6+99pr7/0OGDPFJ2Yjo8tjMm4jCWuUwPWHCBP8VhIiIwlLlMJ2QkFCj92zfvh0V3R5ZrdYa9f5NRN7HME1EYauoqAgpKSkAAJPJhH79+vm5REREFG4q98rdqFGjGr1n/fr17umxY8eyJ28iP2GYJqKwdeDAAfed/a5du8JoNPq5REREFG7Kysrc0xaL5YrLOxyOi4bSmjp1qk/KRURXxjBNRGGrchPvnj17+q8gREQUtiqPUqsoyhWX/+ijj5Cfnw8AaNasGUaNGuWrohHRFTBME1HYYpgmIiJ/a9asmXs6Nzf3sstqmoalS5e6/z9//nwYDAaflY2ILo9hmojCFsM0ERH5W5s2bdzTe/fuveyyb731lvvc1bp1a9x7772+LBoRXYEkKrctISIKE4qiIDIyEmVlZZBlGYWFhbBarf4uFhERhZn09HS0aNECACBJEg4dOoSuXbtesty+ffswZMgQ2O126HQ6bN++HYMHD67v4hJRJayZJqKwdPToUXenL+3bt2eQJiIiv0hMTHSPEy2EwJQpU3D27NmLlvn0008xfPhw2O12AMCTTz7JIE0UANiPPhGFJTbxJiKiQLFs2TL0798fmqbh0KFDaNu2LUaOHAmr1Yq9e/fi559/di87f/58/OMf//BjaYmoAsM0EYUlhmkiIgoUffr0wdtvv417770XiqKgqKgIn3zyyUXLxMbG4qWXXsK0adP8VEoi+iOGaSIKSwcPHnRPM0wTEZG/zZgxAz169MCLL76I7777DpmZmYiJicHVV1+NiRMnYvr06YiLi/N3MYmoEnZARkREREREROQhdkBGRERERERE5CGGaSIiIiIiIiIPMUwTEREREREReYhhmoiIiIiIiMhDDNNEREREREREHmKYJiIiIiIiIvIQwzQRERERERGRhximiYiIiIiIiDzEME1ERERERETkIYZpIiIiIiIiIg8xTBMRERERERF5iGGaiIiIiIiIyEMM00REREREREQeYpgmIiIiIiIi8hDDNBEREREREZGHGKaJiIiIiIiIPMQwTUREREREROQhhmkiIiIiIiIiDzFMExEREREREXmIYZqIiIiIiIjIQwzTRERERERERB5imCYiIiIiIiLyEMM0ERERERERkYcYpomIiIiIiIg8xDBNRERERERE5CGGaSIiIiIiIiIP/X+SlpTHqamVyQAAAABJRU5ErkJggg==", + "image/png": "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", "text/plain": [ "
" ] diff --git a/docs/example.rst b/docs/example.rst index c27613b14..c644363c5 100644 --- a/docs/example.rst +++ b/docs/example.rst @@ -90,7 +90,7 @@ From a Python interpreter You can use **cobaya** interactively within a Python interpreter or a Jupyter notebook. This will allow you to create input and process products *programatically*, making it easier to streamline a complicated analyses. -The actual input information of **cobaya** are Python *dictionaries* (a ``yaml`` file is just a representation of a dictionary). We can easily define the same information above as a dictionary: +The actual input information of **cobaya** is provided as Python *dictionaries* (a ``yaml`` file is just a representation of a dictionary). We can easily define the same information above as a dictionary: .. literalinclude:: ./src_examples/quickstart/create_info.py :language: python diff --git a/docs/installation.rst b/docs/installation.rst index 69657be56..74c917ef8 100644 --- a/docs/installation.rst +++ b/docs/installation.rst @@ -141,7 +141,7 @@ Problems with file locks ^^^^^^^^^^^^^^^^^^^^^^^^ By default Cobaya uses `Portalocker `_ to lock output chain files to check that MPI is being used correctly, that only one process is accessing each file, and to clean up files from aborted runs. -If Portalocker is uninstalled it will still work, but files may need to be cleaned up manually. You can also set an environment variable to turn off file locking if it causes problems (e.g. on NERSC home). +If Portalocker is not installed it will still work, but files may need to be cleaned up manually. You can also set an environment variable to turn off file locking if it causes problems (e.g. on NERSC home). .. code:: bash diff --git a/tests/test_cosmo_run.py b/tests/test_cosmo_run.py index bca412f87..23c17422c 100644 --- a/tests/test_cosmo_run.py +++ b/tests/test_cosmo_run.py @@ -96,6 +96,9 @@ def test_cosmo_run_resume_post(tmpdir, skip_not_installed, packages_path=None): # note that continuing from files leads to text-file precision at read in, so a mix of # precision in the output SampleCollection returned from run run(info, resume=True, override={'sampler': {'mcmc': {'Rminus1_stop': 0.2}}}) + run(info, resume=True, override={'sampler': {'mcmc': {'Rminus1_stop': 0.15}}, + 'theory': {'camb': {'stop_at_error': False}}}, + allow_changes=True) updated_info, sampler_init = run( info['output'] + '.updated' + Extension.dill, resume=True, override={'sampler': {'mcmc': {'Rminus1_stop': 0.05}}} @@ -111,7 +114,8 @@ def test_cosmo_run_resume_post(tmpdir, skip_not_installed, packages_path=None): 'skip': 0.2, 'thin': 4 } - output_info, post_results = run(updated_info, override={'post': info_post}, force=True) + output_info, post_results = run(updated_info, override={'post': info_post}, + force=True) samp2 = post_results.samples(to_getdist=True) samp_test = samp.copy() samp_test.weighted_thin(4) @@ -170,7 +174,7 @@ def test_cosmo_run_resume_post(tmpdir, skip_not_installed, packages_path=None): samp_thin = sampler_init.samples(skip_samples=0.2, to_getdist=True) samp_thin.weighted_thin(4) assert samp_thin.numrows == samp_revert.numrows + \ - post_products_revert["stats"]["points_removed"] + post_products_revert["stats"]["points_removed"] if not post_products_revert["stats"]["points_removed"]: assert np.isclose(samp_revert.mean("sigma8"), samp_thin.mean("sigma8")) else: diff --git a/tests/test_input.py b/tests/test_input.py index e1c0ba7e0..705f6a762 100644 --- a/tests/test_input.py +++ b/tests/test_input.py @@ -81,3 +81,4 @@ def test_run_file(tmpdir): default_info = get_default_info(likname, "likelihood") updated_info = yaml_load_file(root + '.updated.yaml') assert updated_info["prior"] == default_info["prior"] + run_script([input_file, '--resume', '--allow-changes', '--debug'])