The first public PyTorch implementation of Skip-Thought Vectors
An implementation of Skip-Thought Vectors in PyTorch.
d = DataLoader("./data/dummy_corpus.txt")
Trainnotebook runs at the rate of 1 epoch / 2 days on an Nvidia 1080 Ti.
./saved_modelswhen the average training loss in the last 20 iterations dips below the previous best.
Only implemented on classification tasks * Download the movie review dataset and put
./tasks/mr_datadirectory. * You may also test on other classification tasks by downloading the datasets and providing their path and tasks type in