The code here implements the Dual LSTM Encoder model from The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems.
- The original code (https://github.com/dennybritz/chatbot-retrieval) was created with TensorFlow < 1.0
- Some of the functions were changed to work properly on later versions (TensorFlow < 1.4.0) by https://github.com/transposed/chatbot-retrieval
- After changing some function parameter orders, I have managed to run the codes on TensorFlow 1.4.0
This code uses Python 3 and Tensorflow >= 1.4.0. Clone the repository and install all required packages:
pip install -U pip
pip install numpy scikit-learn pandas jupyter
Download the train/dev/test data here and extract the acrhive into ./data
.
python udc_train.py
python udc_test.py --model_dir=...
python udc_predict.py --model_dir=...