make generate images function work with parallel in CIFAR files

atong01 / conditional-flow-matching

TorchCFM: a Conditional Flow Matching library

https://arxiv.org/abs/2302.00482

MIT License

1.25k stars 101 forks source link

make generate images function work with parallel in CIFAR files #83

Closed kilianFatras closed 11 months ago

kilianFatras commented 11 months ago

What does this PR do?

Remove unused FLAG variables in cifar experiments.
Make generate_images function work with parallel

Fixes #82 and partially 73

Before submitting

[x] Did you make sure title is self-explanatory and the description concisely explains the PR?
[x] Did you make sure your PR does only one thing, instead of bundling different changes together?
[x] Did you list all the breaking changes introduced by this pull request?
[x] Did you test your PR locally with pytest command?
[x] Did you run pre-commit hooks with pre-commit run -a command?

Did you have fun?

Make sure you had fun coding 🙃

codecov[bot] commented 11 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (3c0cfdd) 35.90% compared to head (d964396) 35.90%.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #83 +/- ## ======================================= Coverage 35.90% 35.90% ======================================= Files 67 67 Lines 7419 7419 ======================================= Hits 2664 2664 Misses 4755 4755 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

kilianFatras commented 11 months ago

-- DO NOT MERGE --

After two tests, the above fix does not seem to perform as well as the main branch. I got an FID of 4.9 for the two tests while it should be 3.6. For now, I consider the parallel training not fixed.