.circleci: Add workflows for Python 3.8

[seemethere]

Done by just editing .circleci/cimodel/data/dimensions.py to include 3.8 and then regenerated using .circleci/regenerate.sh

cc @kostmo, @mingbowan, @ezyang, @soumith

Signed-off-by: Eli Uriegas eliuriegas@fb.com

[soumith]

if things pass, shipit!

[seemethere]

After a second glance through at these changes I don't think any python 3.8 specific tests will be run on PRs and that these changes only apply to the nightly/postnightly branches.

@kostmo where would the other places to change be so that we can have full Python 3.8 coverage?

[kostmo]

Yes, Python version lists are scattered in a few different places. I think this will find the ones that matter:

kostmo@ubuntu:~/github/pytorch-repos/pytorch$ git grep --name-only '3.6' -- .circleci/cimodel/**/*.py
.circleci/cimodel/data/binary_build_data.py
.circleci/cimodel/data/caffe2_build_data.py
.circleci/cimodel/data/caffe2_build_definitions.py
.circleci/cimodel/data/dimensions.py
.circleci/cimodel/data/pytorch_build_data.py
.circleci/cimodel/data/pytorch_build_definitions.py

I wish it was as simple as a single list!

[kostmo]

💊 CircleCI build failures summary and remediations

As of commit 01abf37:

None of the build failures appear to be your fault.

• 1/1 recognized as flaky ❄️
  • Re-run these jobs?

Detailed failure analysis

One may explore the probable reasons each build failed interactively on the Dr. CI website.

❄️ 1 failure recognized as flaky

The following build failures have been detected as flaky and may not be your fault:

pytorch_linux_xenial_cuda10_1_cudnn7_py3_NO_AVX_NO_AVX2_test (1/1)

Step: "Test" (full log | pattern match details) ❄️

Jan 28 06:35:31 RuntimeError: quantization scale should be > 0

Jan 28 06:35:31   File "test_quantization.py", line 722, in test_quantized_rnn 
Jan 28 06:35:31     y, (h, c) = cell_dq(x, (h, c)) 
Jan 28 06:35:31   File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 540, in __call__ 
Jan 28 06:35:31     result = self.forward(*input, **kwargs) 
Jan 28 06:35:31   File "/opt/conda/lib/python3.6/site-packages/torch/nn/quantized/dynamic/modules/rnn.py", line 373, in forward 
Jan 28 06:35:31     return self.forward_tensor(input, hx) 
Jan 28 06:35:31   File "/opt/conda/lib/python3.6/site-packages/torch/nn/quantized/dynamic/modules/rnn.py", line 334, in forward_tensor 
Jan 28 06:35:31     input, hx, batch_sizes, max_batch_size, sorted_indices) 
Jan 28 06:35:31   File "/opt/conda/lib/python3.6/site-packages/torch/nn/quantized/dynamic/modules/rnn.py", line 315, in forward_impl 
Jan 28 06:35:31     self.batch_first, dtype=self.dtype, use_dynamic=True) 
Jan 28 06:35:31 RuntimeError: quantization scale should be > 0 
Jan 28 06:35:31  
Jan 28 06:35:31 ---------------------------------------------------------------------- 
Jan 28 06:35:31 Ran 36 tests in 55.359s 
Jan 28 06:35:31  
Jan 28 06:35:31 FAILED (errors=1, skipped=1) 
Jan 28 06:35:31  
Jan 28 06:35:31 Generating XML reports... 
Jan 28 06:35:31 Traceback (most recent call last): 
Jan 28 06:35:31   File "test/run_test.py", line 456, in <module> 
Jan 28 06:35:31     main() 

---

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

This comment has been revised 15 times.

[seemethere]

Whoo! macOS builds are ready to go:

• https://circleci.com/gh/pytorch/pytorch/4354608?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link

[facebook-github-bot]

@seemethere is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

[facebook-github-bot]

@seemethere merged this pull request in da39091.