https://github.com/abehandler/cnn-sentence-classification
https://github.com/abehandler/cnn-sentence-classification
Last synced: over 1 year ago
JSON representation
- Host: GitHub
- URL: https://github.com/abehandler/cnn-sentence-classification
- Owner: AbeHandler
- Created: 2017-03-20T23:47:38.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2016-08-08T10:32:28.000Z (almost 10 years ago)
- Last Synced: 2025-01-20T10:44:55.259Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 488 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Sentence Classification With Convolutional Neural Networks
This is an extension of yoonkim's implementation of CNN_sentence to classify test sentences. In addition to the original Random and Word2Vec embeddings, GloVe support has been added.
The original implementation can be found here: https://github.com/yoonkim/CNN_sentence
The file conv_net_classes.py is the same file from original implementation.
The files conv_net_sentence.ipynb and process.ipynb are modified files.
The original rt-polarity.neg and rt-polarity.pos files consisted of 5331 examples each. I took 50 examples from each of them to create
the test.data file. Here rt-polarity.neg, rt-polarity.pos, test.data consists of 5281, 5281 and 100 examples.
First 50 examples of test.data are of negative sentiment, rest are of positive sentiment.