fix(tokenizers): add missing tokenize_raw to TokenizerTransformers#2
Open
tatsuru-okada-business wants to merge 1 commit into
Open
Conversation
The index pipeline (index/tokenize.py: tokenize_encode_offsets) calls
tokenizer.tokenize_raw(), which every tokenizer implements (mecab, icu,
llama, moses) except TokenizerTransformers. As a result, building an
index with --backend transformers always crashes:
softmatcha-index --index /tmp/idx --backend transformers \
--model tohoku-nlp/bert-base-japanese-v3 corpus.txt
...
AttributeError: 'TokenizerTransformers' object has no attribute 'tokenize_raw'
This mirrors the llama tokenizer's implementation (tokenize_raw ==
tokenize on the stripped line). Verified: index builds successfully and
search returns expected results with tohoku-nlp/bert-base-japanese-v3.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
Building an index with
--backend transformersalways crashes:Cause
The index pipeline (
src/softmatcha/index/tokenize.py,tokenize_encode_offsets) callstokenizer.tokenize_raw(). Every tokenizer implements it (mecab, icu, llama, moses) exceptTokenizerTransformers, so the transformers backend is effectively unusable for indexing.Fix
Add
tokenize_rawmirroring the llama tokenizer's implementation (identical totokenizeon the stripped line). 3 lines, no behavior change for other backends.Verification
With this patch,
softmatcha-index --backend transformers --model tohoku-nlp/bert-base-japanese-v3builds a complete index (metadata/offset/rough/sa/index.bin) andsoftmatcha-searchreturns expected soft matches on a ~2MB Japanese corpus.🤖 Generated with Claude Code