cnn-text-classification-tf

The base project is dennybritz/cnn-text-classification-tf. It can classify the positive and negative movie reviews. I modified it to have 5 labels(0,1,2,3,4).

It is slightly simplified implementation of Kim's Convolutional Neural Networks for Sentence Classification paper in Tensorflow.

Requirements

• Python 3
• Tensorflow > 0.12
• Numpy

Install tensorflow

Please see the install guide in tensorflow website.

Training

Print parameters:

./train.py --help

optional arguments:
-h, --help            show this help message and exit
--data_file DATA_FILE
File for training or evaluation (default: /data/train.txt for training, /data/dev.txt for evaluation)
--embedding_dim EMBEDDING_DIM
Dimensionality of character embedding (default: 128)
--filter_sizes FILTER_SIZES
Comma-separated filter sizes (default: '3,4,5')
--num_filters NUM_FILTERS
Number of filters per filter size (default: 128)
--l2_reg_lambda L2_REG_LAMBDA
L2 regularizaion lambda (default: 0.0)
--dropout_keep_prob DROPOUT_KEEP_PROB
Dropout keep probability (default: 0.5)
--batch_size BATCH_SIZE
Batch Size (default: 64)
--num_epochs NUM_EPOCHS
Number of training epochs (default: 100)
--evaluate_every EVALUATE_EVERY
Evaluate model on dev set after this many steps
(default: 100)
--checkpoint_every CHECKPOINT_EVERY
Save model after this many steps (default: 100)
--allow_soft_placement ALLOW_SOFT_PLACEMENT
Allow device soft device placement
--noallow_soft_placement
--log_device_placement LOG_DEVICE_PLACEMENT
Log placement of ops on devices
--nolog_device_placement

Train:

./train.py

Train, including the Kaggle dataset:

./train.py --data_file ./data/train-all.txt

Evaluating

./eval.py --checkpoint_dir="./runs/1459637919/checkpoints/"

Replace the checkpoint dir with the output from the training. To use your own data, change the eval.py script to load your data.

References

• Convolutional Neural Networks for Sentence Classification
• A Sensitivity Analysis of (and Practitioners' Guide to) Convolutional Neural Networks for Sentence Classification