Implementation of a Recursive Neural Network in Pytorch
Files are detailed as follows:
- model.py
- Pytorch implementation of a Recursive Neural Network
- main.py
- Main training file. Usage is:
python main.py
, runpython main.py --help
for a list of arguments.
- Main training file. Usage is:
- tree.py
- Objects and functions for tree processing
- utils.py
- Utility classes/functions to process the Stanford Sentiment Treebank (SSTB)
- transform_parse_trees.py
- Script to transform the SSTB into accepted format
- trees/
- Directory containing preprocessed trees in accepted format