New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

LightNER 在 cluener数据集下F1 Score 74% #413

Closed

zr-icu opened this issue Feb 28, 2024 · 1 comment

Labels

question

zr-icu commented Feb 28, 2024 •

edited

Loading

如题，在few-shot代码下使用 bart-large-chinese将 freeze_plm 设置成 True 训练45 epoch。F1 Score 才达到 48.5%

然后将 freeze_plm 设置成 False 训练 45 epoch。F1 Score 达到76%

使用的是全量的 cluener 训练集数量为 1w+。感觉训练出来的指标好低呀，请问可以帮忙看下么？

zr-icu added the question label

Contributor

flow3rdown commented Feb 28, 2024

LightNER主要适用场景为少样本场景，另外LightNER在中文场景下的性能存疑，主要是因为生成式的方式在中文场景下寻找实体边界的能力较弱。您可以尝试调一下参数，例如learning rate, batch size等，参考issue #127

zr-icu closed this as completed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment