Skip to content

xerrors/LabelPrompt

Repository files navigation

LabelPrompt

code for LabelPrompt: Effective Prompt-based Learning for Relation Classification

Setup

To set up the environment, follow these steps:

conda create -n labelprompt python=3.8
conda activate labelprompt

pip install -r requirements.txt

Dataset

The dataset can be found at KnowPrompt.

for few-shot

To generate few-shot datasets, run the following command:

python genetate_k_shot.py --dataset retacred --data_file train.txt

Run

To run the model, use the following command:

python main.py --gpu 0 --dataset retacred --data_path ./dataset/retacred/k-shot/8-1

About

code for LabelPrompt: Effective Prompt-based Learning for Relation Classification

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors