This repository has been archived by the owner on Dec 16, 2022. It is now read-only.
Add per node beam size option to beam search #1835
Merged
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Outdated suggestions cannot be applied.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.
I added an option to the beam search classes called
per_node_beam_size
, which sets the maximum number of candidates to consider, at each step, from the same action history. In vanilla beam search,per_node_beam_size = beam_size
, which is left as the default behavior. As discovered in Beam Search Strategies for Neural Machine Translation, settingper_node_beam_size
to a value smaller thanbeam_size
can result in better predictions, as it introduces more diversity into the search.