Aligned Word Embeddings. Make AI word models speak the same language across different datasets.
github.com/vinid/aligned_word_embeddings
Vaya's read on this project
Problem, audience, market, and the verdict — sign in to see it.
Updates
March 2020
- Added encoding
February 2020
- bugfix
December 2019
- bugfixes
- bugfixes
- added dataframe functions and data augmentation routines
- Added t-test on arrow for SWEAT.plot
November 2019
- added corresp_transpose to comparative.utils, bugfixes
- colorcoded word arrows for SWEAT.plot
- changed display of median and mean in SWEAT.plot boxplots
- changed return type of SWEAT.test from tuple to dict
- Added zipf measure
September 2019
- bugfixes
August 2019
- adding dump and load of an awec object
- addedd comments, first draft lexicon_stability
- Added Hotelling test
- Added Lexicon class, fixed typo in utils
- Added comments and parse_lexicon, model_deltas functions
- Added comparative.utils module
- fixed bug in nltk.corpus.stopwords signature
- - fixes