Results of the experiment cannot be reproduced on PKU, trainning stopped after 1 iteration

TylunasLi commented 6 years ago

我们的实验环境 : Python 3.5 Tensorflow 1.4.0 4x Nvidia Geforce 1080Ti 对代码进行了小幅调整以便在我们的环境上运行第一个实验测试的是Sighan2005-PKU 但是，无论BI-LSTM-CRF还是ID-CNN-CRF 每次训练完第1个迭代后，验证集、测试集都只有0.58-0.62 然后提示 'Score too low, break to save time' 退出。同时，用同样脚本分割的语料训练 foolNLTK（https://github.com/rockyzhengwu/FoolNLTK），几乎一样的BI-LSTM-CRF第一轮训练后验证集有0.88的F1值。初步怀疑评测脚本有问题，验证集、测试集分数一致

hankcs commented 6 years ago

https://github.com/hankcs/ID-CNN-CWS/issues/5

SeekPoint commented 5 years ago

我页遇到同样问题 --但是，无论BI-LSTM-CRF还是ID-CNN-CRF 每次训练完第1个迭代后，验证集、测试集都只有0.58-0.62 然后提示 'Score too low, break to save time' 退出。

我已经确认我是python3.6+tf1.4

hankcs / ID-CNN-CWS

Results of the experiment cannot be reproduced on PKU, trainning stopped after 1 iteration #6