PyTorch BigGraph: a distributed system for learning large graph embeddings - up to billions of entities and trillions of edges - Sharding and Negative Sampling - WikiData embeddings (78 mil entities, 4131 relations) - Blog: ai.facebook.com/blog/open-sour… - Code: github.com/facebookresear…

Timothee Lacroix implemented Python API for our library fastBPE: github.com/glample/fastBPE It is faster than sentencepiece, and follows the same algorithm as the original Python implementation of Rico Sennrich fastBPE can easily work on 200GB of text. Check it out!