Skip to content

An implementation of the CNN model described in "A Convolutional Neural Network for Modelling Sentences". Kalchbrenner et al. 2014

Notifications You must be signed in to change notification settings

sandeepwww/Dynamic-CNN-for-Modelling-Sentences

 
 

Repository files navigation

Dynamic-CNN-for-Modelling-Sentences

An implementation of the CNN model described in "A Convolutional Neural Network for Modelling Sentences". Kalchbrenner et al. 2014. Here is the original code in MATLAB.

I referred to https://github.com/FredericGodin/DynamicCNN for implementation hints. The model I implemented however, is very specific to the experiments described in the paper.

Binary Sentiment Analysis

Due to unavailability of a high-end GPU, one epoch of training took ~4 hours and this achieved an accuracy of ~78% on the test set of Stanford Sentiment Treebank. The authors reported an accuracy of 86.8 % by a well-trained model in the paper.

About

An implementation of the CNN model described in "A Convolutional Neural Network for Modelling Sentences". Kalchbrenner et al. 2014

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%