Fix incorrect dimensions when loading the shift predictor #28
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
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Outdated suggestions cannot be applied.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.
When loading the shift predictor, its dimensions were not being loaded correctly.
It does not match the directions of the latent deformator. For example, with a deformator of 64 dimensions, by default the shift predictor was initialized with 128 dimensions. This would not fail during training because of pytorch's use of both one-hots and integers in the cross-entropy loss of the predictor.
This small change in setting the shift predictor's dim explicitly to
directions_countfixes this.Correct me if my interpretation is wrong