Skip to content
/ CoSIm Public

Code and dataset for NAACL 2022 paper "CoSIm: Commonsense Reasoning for Counterfactual Scene Imagination" Hyounghun Kim, Abhay Zala, Mohit Bansal.

License

Notifications You must be signed in to change notification settings

hyounghk/CoSIm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

01a923d · Nov 26, 2022

History

4 Commits
Nov 23, 2022
Jul 6, 2022
Jul 6, 2022
Jul 6, 2022
Nov 26, 2022
Jul 6, 2022

Repository files navigation

CoSIm

Code and dataset for NAACL 2022 paper "CoSIm: Commonsense Reasoning for Counterfactual Scene Imagination" Hyounghun Kim, Abhay Zala, Mohit Bansal.

Prerequisites

  • Python 3.8
  • PyTorch 1.4 or Up
  • For others packages, please run this command.
pip install -r requirements.txt

Dataset

Please download image features etc. from here and unzip in data/cosim_feats folder.
Raw image files can be downloaded from here.

Usage

To train the models, please run the script run/lxmert_pretrain.bash like below:

bash run/run_train.bash 0  

About

Code and dataset for NAACL 2022 paper "CoSIm: Commonsense Reasoning for Counterfactual Scene Imagination" Hyounghun Kim, Abhay Zala, Mohit Bansal.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published