royny / SelF-Eval

Implementation of COLING 2022 Oral presentation paper "SelF-Eval: Self-supervised Fine-grained Dialogue Evaluation".

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Implementation of COLING 2022 Oral presentation paper "SelF-Eval: Self-supervised Fine-grained Dialogue Evaluation".

Prerequisites

First create an environment:

conda create -n self python=3.6

Then install the required packages:

pip install -r requirements.txt

Install Texar locally:

cd texar-pytorch
pip install .

Training

When training, make sure the CHECKPOINT_DIR_PATH key in pretrain.sh is a directory path. This will be the path where we store our checkpoints and experiments.

sh pretrain.sh

Testing

First change the MODE key to test in the file pretrain.sh. Then specify the path to the checkpoint at CHECKPOINT_DIR_PATH. Note, different from training phase, the key now needs to be the path to the checkpoint file. Finally run:

sh pretrain.sh

About

Implementation of COLING 2022 Oral presentation paper "SelF-Eval: Self-supervised Fine-grained Dialogue Evaluation".


Languages

Language:Python 98.4%Language:Shell 0.7%Language:Makefile 0.4%Language:Batchfile 0.3%Language:Perl 0.2%