Add PyKEEN to ecosystem-ci

[mberr]

Before submitting

• Was this discussed/approved via a GitHub issue? (no need for typos and docs improvements)
• Did you create/update your configuration file?
• Did you set runtimes in config for GitHub action integration?
• Did you add your config to CI in Azure pipeline (only projects with 100+ GitHub stars)?
• Are all integration tests passing?

What does this PR do? [optional]

Project: https://github.com/pykeen/pykeen

PyL integration via

• 🔥⚡ PyTorch Lightning pykeen/pykeen#905
• 🔥⚡ PyTorch Lightning - Part 2 pykeen/pykeen#917
• 🚅⚡ Test Training with PyTorch Lightning pykeen/pykeen#930

Did you have fun?

Make sure you had fun coding 🙃

[mberr]

In PyKEEN we are testing with Python 3.8 and 3.10, which are the minimum and maximum Python version we officially support. We test for both, Ubuntu and Windows, but do not have specific tests for macOS.

If you have any suggestions here, feel free to share them. The main reason we are interested in integrating with PyTorchLightning/ecosystem-ci is the possibility to run tests with GPUs, which we at PyKEEN do not have available for continous testing.

[Borda]

that is fine we can test only Linux and Win or just Ubuntu 🦦

[mberr]

This is my Slack account at https://pytorch-lightning.slack.com

[mberr]

https://dev.azure.com/PytorchLightning/compatibility/_build/results?buildId=72651&view=logs&j=fb683405-d979-52da-6de9-2541dff429a6&t=bdee9137-b6d6-59ea-6392-0d699b7aef3e&l=12676

the errors seem to originate from a tqdm in Lightning-only code 🤔

       torch.multiprocessing.spawn.ProcessRaisedException: 
E       
E       -- Process 1 terminated with the following error:
Traceback (most recent call last):
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/torch/multiprocessing/spawn.py", line 69, in _wrap
    fn(i, *args)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/strategies/launchers/spawn.py", line 101, in _wrapping_function
    results = function(*args, **kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 809, in _fit_impl
    results = self._run(model, ckpt_path=self.ckpt_path)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1234, in _run
    results = self._run_stage()
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1321, in _run_stage
    return self._run_train()
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1343, in _run_train
    self._run_sanity_check()
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1411, in _run_sanity_check
    val_loop.run()
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
    self.advance(*args, **kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py", line 154, in advance
    dl_outputs = self.epoch_loop.run(self._data_fetcher, dl_max_batches, kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/loops/base.py", line 204, in run
    self.advance(*args, **kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 122, in advance
    self._on_evaluation_batch_start(**kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py", line 249, in _on_evaluation_batch_start
    self.trainer._call_callback_hooks(hook_name, *kwargs.values())
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 1634, in _call_callback_hooks
    fn(self, self.lightning_module, *args, **kwargs)
  File "/home/AzDevOps_azpcontainer/.local/lib/python3.8/site-packages/pytorch_lightning/callbacks/progress/tqdm_progress.py", line 291, in on_validation_batch_start
    self.val_progress_bar.reset(convert_inf(self.total_val_batches_current_dataloader))
  File "/opt/conda/lib/python3.8/site-packages/tqdm/std.py", line 1408, in reset
    self.last_print_t = self.start_t = self._time()
AttributeError: 'Tqdm' object has no attribute '_time'

@aniketmaurya any idea how to fix / investigate this issue?

[aniketmaurya]

Seems like TQDM version compatibility issue. Cc: @Borda

[codecov]

Codecov Report

Merging #50 (1adafc5) into main (21ddc52) will not change coverage.
The diff coverage is n/a.

Additional details and impacted files

@@        Coverage Diff         @@
##           main   #50   +/-   ##
==================================
  Coverage    85%   85%           
==================================
  Files         2     2           
  Lines       230   230           
==================================
  Hits        196   196           
  Misses       34    34           

[mberr]

Seems like TQDM version compatibility issue. Cc: @Borda

any updates on this?

[Borda]

@mberr I am very sorry for the delay, but I ll take look at it this week as we are rolling some more updates 🦦

[mberr]

@mberr I am very sorry for the delay, but I ll take look at it this week as we are rolling some more updates 🦦

@Borda no problem - great that this is now regaining momentum;

I accepted your proposed changes in 597f7c2...4cd6fae and merged with the current main branch.

[mberr]

This error message looks strange to me:
https://github.com/Lightning-AI/ecosystem-ci/actions/runs/3878585128/jobs/6627870005#step:9:11

The respective version is available through pypi: https://pypi.org/project/torch-max-mem/0.0.4/

This one seems to come from main branch: https://github.com/Lightning-AI/ecosystem-ci/actions/runs/3878585128/jobs/6627869826#step:12:1