Language modeling flex heads #210
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
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Outdated suggestions cannot be applied.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.
Closes #53.
Waiting for #208.
This PR adds three different prediction heads for
XModelWithHeadsclasses, depending on the model architecture:add_causal_lm_head()adds a causal LM head for classes that support this type of head in transformers, e.g. GPT-2, BERT, ...add_masked_lm_head()adds a masked LM head for models with MLM, e.g. BERT, RoBERTa, ...add_seq2seq_lm_head()adds a sequence-to-sequence LM head for encoder-decoder models, e.g. BARTAll heads can be automatically converted from their respective static-head counterparts (e.g.
seq2seqlmfromBartForConditionalGeneration).To ensure that all conversions work as expected, a new test module was added in
test_adapter_conversion.py.