Train specific layers #28
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
This should speed up training and still get good results. Also turns caching to disk off by default.
fofr
reviewed
Sep 6, 2024
Comment on lines
+152
to
+155
| cache_latents_to_disk: bool = Input( | ||
| description="Use this if you have lots of input images and you hit out of memory errors", | ||
| default=False, | ||
| ), |
There was a problem hiding this comment.
What sort of speed up do we get?
![ellipsis-dev[bot]](https://avatars.githubusercontent.com/in/64358?s=60&v=4){kind=small}
| sudo curl -o /usr/local/bin/cog -L "https://github.com/replicate/cog/releases/latest/download/cog_$(uname -s)_$(uname -m)" | ||
| #sudo curl -o /usr/local/bin/cog -L "https://github.com/replicate/cog/releases/latest/download/cog_$(uname -s)_$(uname -m)" | ||
| # this is the latest main which has new base images for torch 2.4.0 | ||
| sudo curl -o /usr/local/bin/cog -L "https://storage.googleapis.com/andreas-scratch-public/cog-26208388-linux-x86_64" |
There was a problem hiding this comment.
Consider using the official GitHub release URL for installing Cog to ensure reliability and security. The current URL points to a Google Cloud Storage location, which might not be as reliable.
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Outdated suggestions cannot be applied.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 should speed up training and still get good results.
Also turns caching to disk off by default.
Depends on #27
Summary:
This PR optimizes training layers using regex, updates package versions, changes caching behavior, and modifies GitHub workflow.
Key points:
layer_match.pyformatch_layers_to_optimizefunction.train.pyto acceptlayers_to_optimize_regexandcache_latents_to_diskas inputs.train.pyto usematch_layers_to_optimizefor layer selection.train.pyto not cache latents to disk.cog.yamlto PyTorch 2.4.0 and torchvision 0.19.0.cog.yaml..github/workflows/push.yamlto install a specific Cog version.Generated with ❤️ by ellipsis.dev