2.13.1 with old pinnings

[xhochy]

Let's at least get this release out. We'll probably have to wait 1-2 more releases until we can loosen the protobuf pin.

Fixes #345
Fixes #329
(doesn't fix 335)

[conda-forge-webservices]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[xhochy]

This builds locally for me on Linux in the CPU-only and the CUDA 11.8 version. If someone can tell me what the correct configuration for CUDA 11.8 is, I can build and upload the full matrix.

[xhochy]

Packages are ready. Linux packages are on uwe.korn-tf-gpu and macOS Packages on uwe.korn-tf-experimental. Logs:

• linux_64_c_compiler_version11cuda_compilernvcccuda_compiler_version11.8cxx_compiler_version11numpy1.22python3.8.____cpython-log.txt
• linux_64_c_compiler_version11cuda_compilernvcccuda_compiler_version11.8cxx_compiler_version11numpy1.22python3.9.____cpython-log.txt
• linux_64_c_compiler_version11cuda_compilernvcccuda_compiler_version11.8cxx_compiler_version11numpy1.22python3.10.____cpython-log.txt
• linux_64_c_compiler_version12cuda_compilerNonecuda_compiler_versionNonecxx_compiler_version12numpy1.23python3.11.____cpython-log.txt
• linux_64_c_compiler_version11cuda_compilernvcccuda_compiler_version11.8cxx_compiler_version11numpy1.23python3.11.____cpython-log.txt
• linux_64_c_compiler_version12cuda_compilerNonecuda_compiler_versionNonecxx_compiler_version12numpy1.22python3.10.____cpython-log.txt
• linux_64_c_compiler_version12cuda_compilerNonecuda_compiler_versionNonecxx_compiler_version12numpy1.22python3.9.____cpython-log.txt
• linux_64_c_compiler_version12cuda_compilerNonecuda_compiler_versionNonecxx_compiler_version12numpy1.22python3.8.____cpython-log.txt
• linux_64_c_compiler_version10cuda_compilernvcccuda_compiler_version11.2cxx_compiler_version10numpy1.22python3.9.____cpython-log.txt
• linux_64_c_compiler_version10cuda_compilernvcccuda_compiler_version11.2cxx_compiler_version10numpy1.22python3.10.____cpython-log.txt
• linux_64_c_compiler_version10cuda_compilernvcccuda_compiler_version11.2cxx_compiler_version10numpy1.22python3.8.____cpython-log.txt
• linux_64_c_compiler_version10cuda_compilernvcccuda_compiler_version11.2cxx_compiler_version10numpy1.23python3.11.____cpython-log.txt
• osx_arm64_numpy1.23python3.11.____cpython.log
• osx_arm64_numpy1.22python3.10.____cpython.log
• osx_arm64_numpy1.22python3.9.____cpython.log
• osx_arm64_numpy1.22python3.8.____cpython.log
• osx_64_numpy1.23python3.11.____cpython.log
• osx_64_numpy1.22python3.8.____cpython.log
• osx_64_numpy1.22python3.9.____cpython.log
• osx_64_numpy1.22python3.10.____cpython.log

[xhochy]

@h-vetinari @hmaarrfk Can you copy and merge?

[h-vetinari]

@h-vetinari @hmaarrfk Can you copy and merge?

Never copied artefacts yet, so I'll defer to @hmaarrfk.

[xhochy]

The copy command is listed here: https://github.com/conda-forge/cfep/blob/main/cfep-03.md#specification I would also do this but I don't know whether that is OK according to the CFEP.

[hmaarrfk]

you can also upload directly.

my instructions are in a comment #291

many thanks

[hmaarrfk]

@xhochy I feel like a while back we discussed that given the amount of core devs that were watching this package, it would be OK for them to directly upload.

Its pretty hard to store this many packages (even locally) so sometimes it has been prohibitive for me to keep creating random anaconda accounts.

[jonas-eschle]

Amazing, thanks a lot for the work!