Repository files navigation

APyTorch based implementation of Tree-LSTM from Kai Sheng Tai's paperImproved Semantic Representations From Tree-Structured Long Short-Term MemoryNetworks.

• PyTorch Deep learning library
• tqdm: display progress bar
• meowlogtool: a logger that write everything on console to file
• Java >= 8 (for Stanford CoreNLP utilities)
• Python >= 3

First run the script./fetch_and_preprocess.sh

hints: you should set java bin in $PATH to runjavac

This downloads the following data:

• Stanford Sentiment Treebank (sentiment classification task)
• Glove word vectors (Common Crawl 840B) --Warning: this is a 2GB download!

and the following libraries:

• Stanford Parser
• Stanford POS Tagger

python sentiment.py --name <name_of_log_file> --model_name <constituency|dependency> --epochs 10

We have not fully test on fine grain classification yet. Binary classification accuracy on both model are the same in original paper.

Kai Sheng Tai for theoriginal LuaTorch implementation
Pytorch team for Python library
Riddhiman Dasgupta for his implement on sentiment relatednesshttps://github.com/dasguptar/treelstm.pytorch which I based on as starter code.

MIT