josephdviviano/lstm-variational-dropout

theoretically grounded application of dropout to LSTMs

Variational Dropout for LSTMs

A simple pytorch implementation of

A Theoretically Grounded Application of Dropout in Recurrent Neural Networks Yarin Gal & Zoubin Ghahramani, 2016. arxiv

This is a simple version of previous implementations I found (e.g., WeightDrop from awd-lstm-lm and torchNLP).

For tests and usage, please see:

python test_model.py