Fix: ignore tokens should be set to -100, not -1#19
Open
maxlchen wants to merge 2 commits into
Open
Conversation
Loss cannot be computed if labels[~masked_indices] = -1. From BertForMaskedLM documentation: "Indices should be in [-100, 0, ..., config.vocab_size] ... Tokens with indices set to -100 are ignored (masked), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size]"
Fix: ignore tokens in target tensor should be set to -100, not -1
This file contains hidden or 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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Issue #, if available:
Description of changes:
Correcting the value of unmasked indices in the target tensor in mask_tokens().
Currently, line 348 is: labels[~masked_indices] = -1
But, the value of -1 breaks the loss function. From BertForMaskedLM documentation: "Indices should be in [-100, 0, ..., config.vocab_size] ... Tokens with indices set to -100 are ignored (masked), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size]"
This value constraint appears to hold in every version of the HuggingFace API which has documentation for BertForMaskedLM.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.