文件列表:

.travis.yml (1788, 2017-02-15)
CMakeLists.txt (13982, 2017-02-15)
CONTRIBUTORS.md (6082, 2017-02-15)
LICENSE (569, 2017-02-15)
Makefile (9954, 2017-02-15)
NEWS.md (2912, 2017-02-15)
R-package (0, 2017-02-15)
R-package\.Rbuildignore (61, 2017-02-15)
R-package\DESCRIPTION (819, 2017-02-15)
R-package\LICENSE (1478, 2017-02-15)
R-package\R (0, 2017-02-15)
R-package\R\callback.R (5213, 2017-02-15)
R-package\R\context.R (1058, 2017-02-15)
R-package\R\executor.R (2102, 2017-02-15)
R-package\R\gru.R (15917, 2017-02-15)
R-package\R\initializer.R (3637, 2017-02-15)
R-package\R\io.R (1502, 2017-02-15)
R-package\R\kvstore.R (220, 2017-02-15)
R-package\R\lr_scheduler.R (2953, 2017-02-15)
R-package\R\lstm.R (15497, 2017-02-15)
R-package\R\metric.R (2067, 2017-02-15)
R-package\R\mlp.R (2718, 2017-02-15)
R-package\R\model.R (20521, 2017-02-15)
R-package\R\ndarray.R (4782, 2017-02-15)
R-package\R\optimizer.R (14517, 2017-02-15)
R-package\R\profiler.R (940, 2017-02-15)
R-package\R\random.R (2484, 2017-02-15)
R-package\R\rnn.R (14725, 2017-02-15)
R-package\R\rnn_model.R (9105, 2017-02-15)
R-package\R\symbol.R (4188, 2017-02-15)
R-package\R\util.R (793, 2017-02-15)
R-package\R\viz.graph.R (6215, 2017-02-15)
R-package\R\zzz.R (995, 2017-02-15)
... ...

Implementing CNN for Text Classification in MXNet ============ It is slightly simplified implementation of Kim's [Convolutional Neural Networks for Sentence Classification](http://arxiv.org/abs/1408.5882) paper in MXNet. Recently, I have been learning mxnet for Natural Language Processing (NLP). I followed this nice blog ["Implementing a CNN for Text Classification in Tensorflow" blog post.](http://www.wildml.com/2015/12/implementing-a-cnn-for-text-classification-in-tensorflow/) to reimplement it by mxnet framwork. Data preprocessing code and courpus are directly borrowed from original author [cnn-text-classification-tf](https://github.com/dennybritz/cnn-text-classification-tf). ## Performance compared to original paper I use the same pretrained word2vec [GoogleNews-vectors-negative300.bin](https://drive.google.com/file/d/0B7XkCwpI5KDYNlNUTTlSS21pQmM/edit?usp=sharing) in Kim's paper. However, I don't implement L2-normalization of weight on penultimate layer, but provide a L2-normalization of gradients. Finally, I got a best dev accuracy 80.1%, close to 81% that reported in the original paper. ## Data Please download the corpus from this repository [cnn-text-classification-tf](https://github.com/dennybritz/cnn-text-classification-tf), :) 'data/rt.vec', this file was trained on the corpus by word2vec tool. I recommend to use GoogleNews word2vec, which could get better performance, since this corpus is small (contains about 10K sentences). When using GoogleNews word2vec, this code loads it with gensim tools [gensim](https://github.com/piskvorky/gensim/tree/develop/gensim/models). ## Remark If I were wrong in CNN implementation via mxnet, please correct me. ## References - ["Implementing a CNN for Text Classification in Tensorflow" blog post.](http://www.wildml.com/2015/12/implementing-a-cnn-for-text-classification-in-tensorflow/) - [Convolutional Neural Networks for Sentence Classification](http://arxiv.org/abs/1408.5882)

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：