Dual LSTM Encoder for Dialog Response Generation
Jupyter Notebook Python
Switch branches/tags
Nothing to show
Clone or download
Pull request Compare This branch is 1 commit ahead of dennybritz:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
models
notebooks
scripts
.gitattributes
.gitignore
LICENSE
README.md
requirements.txt
udc_hparams.py
udc_inputs.py
udc_metrics.py
udc_model.py
udc_predict.py
udc_test.py
udc_train.py

README.md

Retrieval-Based Conversational Model in Tensorflow (Ubuntu Dialog Corpus)

Please read the blog post for this code

Overview

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.

Additions and performed work

The original code from this blog post was created with TesnorFlow < 1.0

I have made the necessary changes for the code to function properly with the latest version of TensorFlow. Mostly, this meant changing the order of the parameters for various functions, and taking out methods that are deprecated.

Setup

This code uses Python 3 and Tensorflow >= 0.9. Clone the repository and install all required packages:

pip install -U pip
pip install numpy scikit-learn pandas jupyter

Get the Data

Download the train/dev/test data here and extract the acrhive into ./data.

Training

python udc_train.py

Evaluation

python udc_test.py --model_dir=...

Evaluation

python udc_predict.py --model_dir=...