GH-387: modify FlairEmbeddings to handle large texts #444
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Outdated suggestions cannot be applied.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 #387
This PR adds a fix for retrieving embeddings from a character LM for long sequences. The idea is to chop long sequences into chunks and push each chunk through the LM, while always remembering the last hidden state as new initial hidden state. This lowers memory requirements (shorter sequences at once) but increases runtime (more calls to RNN).
In detail:
LanugageModel.get_representation()andFlairEmbeddingsnow have thechars_per_chunkparameter that defaults to 512. Lowering this parameter reduces memory but increases runtime.LanguageModelTrainercan now shuffle sentences in each splitDeprecated
DocumentMeanEmbeddingsremoved, as well as most mentions of deprecatedCharLMEmbeddingsRemoved slow unit tests