paper:https://arxiv.org/abs/1408.5882
This is NOT an original implementation. There may be some minor differences from the original structure.
- Python 3.5
- tensorflow-gpu==1.3.0
- numpy==1.13.1
- jupyter==1.0.0
- Use pre-trained vectores from word2vec
- Currently, word embeddings are randomly initialized
- Download "sentence polarity dataset v1.0" from http://www.cs.cornell.edu/people/pabo/movie-review-data/
- run "data_prepare.ipynb" with jupyter notebook
- modify "config.py"
- run
python train.py
- modify sampling.py
- run
python test.py
MIT
Ryo Kamoi