Skip to content

Code and models for the paper "Aspect Sentiment Classification with Aspect-Specific Opinion Spans", EMNLP 2020.

Notifications You must be signed in to change notification settings

xuuuluuu/Aspect-Sentiment-Classification

Repository files navigation

ABSA

[EMNLP 2020] Aspect Sentiment Classification with Aspect-Specific Opinion Spans

Requirement

Python 3.7.3
Pytorch 1.1.0
Numpy 1.17.2
CUDA: 10.0

Suggested Environment Installation Procedures:
conda create -n yourenvname python=3.7 anaconda
conda install -n yourenvname pytorch==1.1.0 torchvision cudatoolkit=10.0 -c pytorch

Running

python train_crf_glove.py  

By default, the model runs on laptop dataset with provided hyper-parameters.
To run on other datasets:

  • In line 131-144, uncomment the main funtion for different dataset.
  • Edit the embed_num on line 148 and path directory in line 171-176, .

To run with the saved best model:

  • Set training to False on line 147.

About

Code and models for the paper "Aspect Sentiment Classification with Aspect-Specific Opinion Spans", EMNLP 2020.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages