fix(train): improve quality of training #274
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
During several tests with a fresh model it introduced (or perhaps kept) some metallic noise. Letting it run without the `torch.no_grad()` call improves this significantly.
Codecov Report
📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more @@ Coverage Diff @@
## main #274 +/- ##
=======================================
Coverage 20.51% 20.51%
=======================================
Files 38 38
Lines 3247 3247
Branches 418 417 -1
=======================================
Hits 666 666
Misses 2564 2564
Partials 17 17
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
|
Two examples of the TF2 Spy after only 200 iterations with the fix: |
|
😍🥰😎 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR includes several things discussed in #270
_temp_epochvariable is initialized in the LightningModule's__init__method (thanks @ne0escape)__init__method of the LightningModule to the top (bit of cleanup)with torch.no_grad():call for the generator loss calculation which ends up vastly improving the quality of the model (less metallic noise during several tests with a model from scratch)log_audio_dictfunction so it use the correcttotal_batch_idxvaluevalidation_stepis being run for every model apparently (we check ifbatch_idxis 0 and only then save)Please tell me if you'd like to see any changes made before potentially merging this 🙏