v0.3.1
MinishLab/model2vecv0.3.1Oct 29, 2024by Pringled
AI Summary
This maintenance release improves token handling, adds optional fast encoding, and fixes a normalization bug.
Key Highlights
- Updated tokens to be more agnostic
- Made encode_batch_fast optional
- Fixed normalization leading to NaN for empty docs
New Features
- Make encode_batch_fast optional
Full Release Notes
## What's Changed * fix: update added tokens to be more agnostic by @stephantul in https://github.com/MinishLab/model2vec/pull/107 * fix: don't rely on reported vocab size, log warning if inconsistent by @stephantul in https://github.com/MinishLab/model2vec/pull/109 * docs: Fixed broken links by @Pringled in https://github.com/MinishLab/model2vec/pull/112 * feat: make encode_batch_fast optional by @stephantul in https://github.com/MinishLab/model2vec/pull/113 * fix: normalize would lead to NaN for empty docs by @stephantul in https://github.com/MinishLab/model2vec/pull/114 * docs: Add tokenlearn results by @Pringled in https://github.com/MinishLab/model2vec/pull/116 * docs: Updated plot by @Pringled in https://github.com/MinishLab/model2vec/pull/117 * Bump version by @Pringled in https://github.com/MinishLab/model2vec/pull/118 **Full Changelog**: https://github.com/MinishLab/model2vec/compare/v0.3.0...v0.3.1