macanv / BERT-BiLSTM-CRF-NER

Tensorflow solution of NER task Using BiLSTM-CRF model with Google BERT Fine-tuning And private Server services
https://github.com/macanv/BERT-BiLSMT-CRF-NER
4.68k stars 1.25k forks source link

结果f1为0 #315

Open Chandler-Bing opened 4 years ago

Chandler-Bing commented 4 years ago

作者您好,用自己的数据训练得到output文件夹下面的predict_scroe.txt内容里P、R、F1都是0,但是label_test.txt里面还是有对应完全正确的实例。请问这个怎么回事呢? image

Chandler-Bing commented 4 years ago

已解决,谢谢

deskl0 commented 4 years ago

已解决,谢谢

你好,我遇到了相同的问题. 请教一下如何解决的

qzd147 commented 4 years ago

你好请问是怎么解决的呢?我遇到了这个问题:processed 0 tokens with 0 phrases; found: 0 phrases; correct: 0. 在label_test.txt里什么数据也没有

Chandler-Bing commented 4 years ago

已解决,谢谢

你好,我遇到了相同的问题. 请教一下如何解决的 我的情况是用错了label,原作者用的是B-GOODS,我没看清楚,用的是B_GOODS,下划线和横杠的区别

Chandler-Bing commented 4 years ago

你好请问是怎么解决的呢?我遇到了这个问题:processed 0 tokens with 0 phrases; found: 0 phrases; correct: 0.在label_test.txt里什么数据也没有

我的情况是用错了label,原作者用的是B-GOODS,我没看清楚,用的是B_GOODS,下划线和横杠的区别