hustercn / chinesetextclassification Goto Github PK

使用卷积神经网络和循环神经网络进行中文文本的分类

CNN做句子分类的论文可以参看: Convolutional Neural Networks for Sentence Classification

还可以去读dennybritz大牛的博客：Implementing a CNN for Text Classification in TensorFlow 以及字符级CNN的论文：Character-level Convolutional Networks for Text Classification

本文是基于TensorFlow在中文数据集上的简化实现，使用了字符级CNN和RNN对中文文本进行分类，达到了较好的效果。

文中所使用的Conv1D与论文中有些不同，详细参考官方文档：tf.nn.conv1d

Tensorfow >= 1.3 Python 2/3 numpy scikit-learn scipy

python run.py cnn/rnn train/test

please choose one model from cnn and rnn to train first, and then test on test text. Finally you can get a predict text set to predict.

size of training text is too large to upload, you can download it from HERE, and then put it in path data/cnews