Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 379 Bytes

README.md

File metadata and controls

10 lines (8 loc) · 379 Bytes

word2vec from scratch

start here

  • Updated to avoid deprecated libraries like Torchtext
  • Merged into a single file app since it's intended for learning
  • Added hugging face datasets for training data
  • Improved performance on some CPU tasks
  • Removed a lot of redundancy between SkipGrams and CBOW

Based on https://github.com/OlgaChernytska/word2vec-pytorch