Skip to content

CNN and RNN with Attention for Chinese Text Classification in Tensorflow

License

Notifications You must be signed in to change notification settings

YCG09/tf-text-classification

Repository files navigation

Environment setup

  • Requirements: python3, tensorflow1.10, sklearn, numpy, pandas and jieba
pip install -r requirements.txt

Training

  • CNN
python train_cnn.py
  • RNN with attention
python train_rnn.py

the data format is as follows: label '\t' text

Testing

  • evaluation mode:
python test.py --checkpoint_dir=./runs/textcnn/trained_result_1548144557/checkpoints --model_type=CNN
  • prediction mode:
python test.py --checkpoint_dir=./runs/textcnn/trained_result_1548144557/checkpoints --model_type=CNN --test_mode=prediction

Inference

python inference.py --checkpoint_dir=./runs/textrnn/trained_result_1548145204/checkpoints --model_type=RNN

About

CNN and RNN with Attention for Chinese Text Classification in Tensorflow

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages