This project is targeted for sentence classification by RNN structure.
This project require following environments
Environment | Version |
---|---|
Python | 3.6.x or later |
TensorFlow | 1.4.x - 1.12.0 |
numpy | 1.13.x |
This project use following dataset for train and test.
- Sentence polarity dataset v1.0 Movie Review data from Rotten Tomatoes
Learning Step:
Use sentece_classification.py
. Once executed, learning is executed with the default hyperparameter and model data and vocabulary are generated under the "run" directory.
Classification App:
Use Classifier.py
. Once execute, classification is executed with the default hyperparameter and the classification result of the test sentence written in the program is output.
If you want to print all hyper parameter and runtime parameter, please execute the following command.
python sentence_classification.py --help
or
python Classifier.py --help
This software is released under the Apache License 2.0, see LICENSE.