Skip to content

Latest commit

 

History

History
24 lines (19 loc) · 1004 Bytes

README.md

File metadata and controls

24 lines (19 loc) · 1004 Bytes
Raw

Affective Manifolds

Using affective manifolds, we can model a machine's mind to feel like a human.

This is the code for the following paper:

  • Benyamin Ghojogh, "Affective Manifolds: Modeling Machine's Mind to Like, Dislike, Enjoy, Suffer, Worry, Fear, and Feel Like A Human", arXiv:2208.13386, 2022.
  • Link to the paper: https://arxiv.org/abs/2208.13386

How to run the code

For running the code of affective manifolds:

  • Install the required packages:
pip install -r requirements.txt
  • Put the dataset files train.csv and test.csv in the path ./dataset/.
  • Change the config file in the path ./configs/config.json. You can use the example config files in the path ./configs/example_configs/.
  • Run the code main_affective.py:
python main_affective.py

The part of code for Siamese network is inspired by the code main_siamese.py, which is taken from the following Kaggle page: