中文长下文词语预测(Chinese WPLC)数据集
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
pengf fd8a9e25b6 上传文件至 'nezha_evaluate' 1 month ago
..
README.md 上传文件至 'nezha_evaluate' 1 month ago
nezha_evaluate.py 上传文件至 'nezha_evaluate' 1 month ago
tokenization.py 上传文件至 'nezha_evaluate' 1 month ago

README.md

NEZHA-EVALUATE

The scripts and model are based on https://github.com/huawei-noah/Pretrained-Language-Model/tree/master/NEZHA-Gen-TensorFlow

Release Notes

First version: 2021/04/22

Environment

The scripts are tested sucessfully with Tensorflow 1.13 and Python 3.6.

The python package fire is required. You may need to install the fire package with the command:

pip3 install fire

Usage

Step 1: Download the folder named models_gpt from the link below and move the folder to the same directory with the scripts.

Step 2: Move the file named dev.json to the same directory with the scripts.

Step 3: Run the script nezha_evaluate.py with the command:

python3 nezha_evaluate.py

Model download