Skip to content

Latest commit

 

History

History
28 lines (19 loc) · 786 Bytes

RELEASE.md

File metadata and controls

28 lines (19 loc) · 786 Bytes

Version 1.14.0

Major Features and Improvements

  • Adds a reserved_tokens parameter to vocabulary APIs, a list of tokens that must appear in the vocabulary and maintain their order at the beginning of the vocabulary.

Bug Fixes and Other Changes

  • approximate_vocabulary now returns tokens with the same frequency in reverse lexicographical order (similarly to tft.vocabulary).
  • Transformed data batches are now sliced into smaller chunks if their size exceeds 200MB.
  • Depends on pyarrow>=10,<11.
  • Depends on apache-beam>=2.47,<3.
  • Depends on numpy>=1.22.0.
  • Depends on tensorflow>=2.13.0,<3.

Breaking Changes

  • Vocabulary related APIs now require passing non-positional parameters by key.

Deprecations

  • N/A