make release-tag: Merge branch 'master' into stable

MLBazaar · Oct 26, 2018 · c59e8f6 · c59e8f6
2 parents c3537a7 + 62c96fe
commit c59e8f6
Show file tree

Hide file tree

Showing 7 changed files with 39 additions and 19 deletions.
diff --git a/HISTORY.md b/HISTORY.md
@@ -1,5 +1,11 @@
 # History
 
+## 0.1.1
+
+* Removal of unnecessary dependencies
+* Separate `tabular` into `single_table` and `multi_table`
+* Improve datasets stats collection.
+
 ## 0.1.0
 
 * D3M Dataset parser

diff --git a/mit_d3m/__init__.py b/mit_d3m/__init__.py
@@ -4,7 +4,7 @@
 
 __author__ = """MIT Data To AI Lab"""
 __email__ = '[email protected]'
-__version__ = '0.1.0'
+__version__ = '0.1.1-dev'
 
 import os
 import shutil

diff --git a/mit_d3m/dataset.py b/mit_d3m/dataset.py
@@ -123,12 +123,12 @@ def get_data_modality(self):
             return first_type
 
         elif len(resource_types) == 1:
-            return 'tabular'
+            return 'single_table'
 
         else:
             second_type = resource_types[1]
             if second_type == 'table':
-                return 'tabular'
+                return 'multi_table'
 
             return second_type
 

diff --git a/mit_d3m/loaders.py b/mit_d3m/loaders.py
@@ -535,7 +535,8 @@ def load(self, d3mds):
 
 
 _LOADERS = {
-    'tabular': TabularLoader,
+    'single_table': TabularLoader,
+    'multi_table': TabularLoader,
     'timeseries': TabularLoader,
     'image': ImageLoader,
     'text': TextLoader,

diff --git a/mit_d3m/stats.py b/mit_d3m/stats.py
@@ -68,6 +68,24 @@ def get_problems(dataset, path):
     return problems
 
 
+STATS_COLUMNS = [
+    'dataset',
+    'data_modality',
+    'task_type',
+    'task_subtype',
+    'metric',
+    'dataset_id',
+    'problem_id',
+    'target',
+    'size',
+    'size_human',
+    'test_features',
+    'test_samples',
+    'train_features',
+    'train_samples',
+]
+
+
 def get_stats(datasets, path):
     data = []
     for dataset in datasets:
@@ -83,7 +101,7 @@ def get_stats(datasets, path):
 
             data.append(stats)
 
-    return pd.DataFrame(data)
+    return pd.DataFrame(data)[STATS_COLUMNS]
 
 
 if __name__ == '__main__':

diff --git a/setup.cfg b/setup.cfg
@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.1.0
+current_version = 0.1.1-dev
 commit = True
 tag = True
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+))?

diff --git a/setup.py b/setup.py
@@ -10,18 +10,13 @@
     history = history_file.read()
 
 install_requires = [
-    'baytune==0.2.1',
-    'boto==2.48.0',
-    'featuretools==0.2.0',
-    'mlblocks==0.2.3',
-    'mlprimitives==0.1.2',
+    'boto3>=1.9.27',
     'networkx==2.1',
-    'numpy==1.15.2',
-    'pandas==0.23.4',
-    'pymongo==3.7.2',
-    'requests==2.18.4',
-    'scikit-learn==0.20.0',
-    'scipy==1.1.0',
+    'numpy>=1.15.2',
+    'pandas>=0.23.4',
+    'pymongo>=3.7.2',
+    'scikit-learn>=0.20.0',
+    'scipy>=1.1.0',
 ]
 
 setup_requires = [
@@ -51,7 +46,7 @@
 
     # fix style issues
     'autoflake>=1.1',
-    'autopep8>=1.3.5',
+    'autopep8==1.3.4',
 
     # distribute on PyPI
     'twine>=1.10.0',
@@ -91,6 +86,6 @@
     test_suite='tests',
     tests_require=tests_require,
     url='https://github.com/HDI-Project/mit-d3m',
-    version='0.1.0',
+    version='0.1.1-dev',
     zip_safe=False,
 )