Evaluate the effectiveness of the Whisper model.
Project description
whisper-evaluate
Evaluate the effectiveness of the Whisper model.
Install
从pypi安装:
pip install whisper-evaluate
从本地安装:
git clone https://git.xindong.com/fengyanglu/whisper-evaluate.git
cd whisper-evaluate
pip install dist/whisper_evaluate-1.2.0-py3-none-any.whl
How to use
参考test.py
from whisper_evaluate import Report
import pprint
if __name__ == '__main__':
# 设置参数: 任务名称
rep = Report("test")
# 生成评估报告
result = rep.report("test_data/test-data-standard-latest.csv",
"test_data/test-data-standard-latest-whisper1.1-res.txt",
"test_data/test_antispam_0727.txt")
pprint.pprint(result)
# 保存统计备查数据 和 计算结果
rep.save()
# 单独计算whisper结果
result = rep.whisper_report("test_data/test-data-standard-latest.csv",
"test_data/test-data-standard-latest-whisper1.1-res.txt")
pprint.pprint(result)
# 单独计算netease结果
result = rep.netease_report("test_data/test-data-standard-latest.csv",
"test_data/test_antispam_0727.txt")
pprint.pprint(result)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
whisper-evaluate-1.2.0.tar.gz
(5.4 kB
view hashes)
Built Distribution
Close
Hashes for whisper_evaluate-1.2.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5ed2f665cddb34c0934e54f0cbc6b520b31d43871f8e286535c924e570cfca5b |
|
MD5 | c7f7e8891dc244a5ed02a9041345c9aa |
|
BLAKE2b-256 | a829bad7cd56614a8a2173dca5471a9cf6738bd30cb836ee1aa69ab61fa41b3c |