ITADN
neurlang/classifier

版本发布 8

v0.1.8
? · 2025-04-04

- Query Key Value version of cross attention layer - add multi (homograph, homonym) trainer and dataset - fix bug: we were hashing the placeholder symbol _ as the solution word - Changes to multi transformer and to it's dataset - implement masked attention but not use it (it performed worse) ## What's Changed * rename dirty to lexicon by @thewh1teagle in https://github.com/neurlang/classifier/pull/2 ## New Contributors * @thewh1teagle made their first contribution in https://github.com/neurlang/classifier/pull/2 **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.7...v0.1.8

v0.1.7
? · 2025-03-12

- add phonemizers (transformer) and their dataset - add deps - doc fix - add sum and cross attention layer **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.6...v0.1.7

v0.1.6
? · 2025-02-28

- fix loading bug of certain models **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.5...v0.1.6

v0.1.5
? · 2025-02-26

* global premodulo support * quaternary training support * fix slight inneficiency **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.4...v0.1.5

v0.1.4
? · 2025-02-06

* fix cuda * fix tiny bug in the new cpu algo * make learning.Reducing() public **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.3...v0.1.4

v0.1.3
? · 2025-02-01

- cuda learning is broken in this release - fix build on android/arm64 **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.2...v0.1.3

v0.1.2
? · 2025-02-01

- new experimental learning - cuda learning is broken in this release, might be fixed later - move the pre-xorring+subtracting burden to json loader/writer - only build vectorized hash on supported archs (fixes goruut build) **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.1...v0.1.2

v0.1.1
? · 2025-01-30

- multiclassifier updates - finalized phonemizer training script. This phonemizer trainer (v0.1.1) will be used for the next goruut release. - fix loading of zlib - used transition memoization optimalization, use improvement possible - remember local minima and avoid them - parallel: add LoopUntil() **Full Changelog**: https://github.com/neurlang/classifier/compare/v0.1.0...v0.1.1