Skip to content

Add support for Split pretokenizer w/ behavior=removed & invert=false #1033

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 16, 2024
Merged

Add support for Split pretokenizer w/ behavior=removed & invert=false #1033

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/tokenizers.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1518,6 +1518,8 @@ class SplitPreTokenizer extends PreTokenizer {

if (this.config.invert) {
return text.match(this.pattern) || [];
} else if (this.config.behavior?.toLowerCase() === 'removed') {
return text.split(this.pattern).filter(x => x);
} else {
return regexSplit(text, this.pattern);
}
Expand Down
10 changes: 10 additions & 0 deletions tests/models/roberta/tokenization.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -691,4 +691,14 @@ export const TEST_CONFIG = {
decoded: "<s> \tH\u00e4LLo!how \n Are yoU? </s>",
},
},

// Split tokenizer with behavior="Removed" and invert=false
"onnx-community/camembertv2-base": {
SIMPLE: {
text: BASE_TEST_STRINGS.SIMPLE,
tokens: ['How', 'are', 'you', 'doi', '##ng', '?'],
ids: [1, 14473, 9556, 10577, 6471, 9274, 38, 2],
decoded: "[CLS] How are you doing? [SEP]",
}
},
};
Loading