Skip to content
A Pytorch Implementation of EsCVAE: Learning to Converse Emotionally Like Humans.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
corpus
modules
.gitignore
README.md
constants.py
escvae.py
main.py
seq2seq.py
utils.py

README.md

The EsCVAE model: A Pytorch implementation.

A Pytorch Implementation of EsCVAE: "Learning to Converse Emotionally Like Humans: A Conditional Variational Approach" (Zhang R, Wang Z., CCF International Conference on Natural Language Processing and Chinese Computing 2018). A Tensorflow implementation can be found in this repository.

Usage

Modify the utils.py to set the data_path and run the main.py to run the model.

More Details of the data format will be updated soon.

Some Notes

  1. Some details in this code is different from the paper. For example, we did not implement the beam search decoder here.
  2. The hyper parameters are not fine-tuned in this implementation. Please modify the hyper parameters by yourself.

Acknowledge

How do I cite EsCVAE?

@inproceedings{zhang2018learning,
    title={Learning to Converse Emotionally Like Humans: A Conditional Variational Approach},
    author={Zhang, Rui and Wang, Zhenyu},
    booktitle={CCF International Conference on Natural Language Processing and Chinese Computing},
    pages={98--109},
    year={2018},
    organization={Springer} 
}
You can’t perform that action at this time.