implement an efficient bpe function using stack #339
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Outdated suggestions cannot be applied.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.
The original bpe function looks for the bigram with the smallest
bpe_ranks
value in a word then merge this bigram, and repeating this process in a while loop.We use a stack to implement the process in a linear O(n) complexity, and the code is also much simpler than before.
See the performance comparison below:
text
is from https://www.reedbeta.com/blog/programmers-intro-to-unicode/my_enc.encode
represents the encode function using the original bpe fucntion, andmy_enc.my_encode
represents the encode function using the modified bpe function