• python3.7
• pytorch
• torchnlp

The main idea is from paper DialogueRNN An Attentive RNN for Emotion Detection in Conversations.

And there are two main parts of this model

1. Feature extraction

   Used BiLSTM to get one sentence's vector.

2. Emotion Detection

   Mainly composed of three GRUCells.

DailyDialogue

In this project pretrained word vectors GloVe was used. And in utils.py you can adjust the store path TORCHNLP_CACHEDIR of vectors file.

• train

  $ python3 main.py --istrain

• test

  $ python3 main.py

• other options

  $ python3 main.py --help

If you think this project is helpful to you, plz star it and let more people see it. :)