История

saidbleik db9f076d68 minor edits		2020-05-20 14:26:05 +00:00
..
bert	update move_model_to_device args	2020-02-13 21:21:17 +00:00
gensen	Added tests to Gensen	2019-08-02 18:25:06 -04:00
glove	update the embedding notebook and glove folder	2019-08-01 22:55:37 +00:00
pretrained_embeddings	minor	2019-08-19 02:05:16 -04:00
pytorch_modules	minor edits	2019-08-16 15:08:31 +00:00
transformers	minor edits	2020-05-20 14:26:05 +00:00
xlnet	update move_to_device refs	2020-01-18 17:56:03 +00:00
README.md	filling in README for XLNet modules	2019-08-29 14:41:00 -04:00

README.md

Models

The models submodule contains implementations of various algorithms that can be used in addition to external packages to evaluate and develop new natural language processing systems. A description of which algorithms are used in each scenario can be found on this table

Summary

The following table summarizes each submodule.

Submodule	Description
bert	This submodule includes the BERT-based models for sequence classification, token classification, and sequence encoding.
gensen	This submodule includes a distributed Pytorch implementation based on Horovod of learning general purpose distributed sentence representations via large scale multi-task learning by refactoring https://github.com/Maluuba/gensen
pretrained embeddings	This submodule provides utilities to download and extract pretrained word embeddings trained with Word2Vec, GloVe, fastText methods.
pytorch_modules	This submodule provides Pytorch modules like Gated Recurrent Unit with peepholes.
xlnet	This submodule includes the XLNet-based model for sequence classification.