longlongman / CasRel-pytorch-reimplement

Pytorch reimplement of the paper "A Novel Cascade Binary Tagging Framework for Relational Triple Extraction" ACL2020. The original code is written in keras.
200 stars 53 forks source link

The performance of pytorch-reimplement #14

Open Onion12138 opened 3 years ago

Onion12138 commented 3 years ago

It seems that the performance isn’t as good as expected when I apply the model to other Chinese Datasets. I wonder whether it is due to the model itself or the missing details in the pytorch reimplement. Looking forward to your reply, many Thanks.

longlongman commented 3 years ago

Sorry to hear about that, as you can see, I use the code on a Chinese Dataset(CMED) and achieve better results compared with the Keras version. I did not test the model on other Chinese Datasets, but I have done my best to keep the torch version as same as the Kears version.

jxrjlxc02 commented 3 years ago

大佬,请问 f1 = 47.59,在真实场景能使用么,会不会偏低了些?

longlongman commented 3 years ago

确实不能用在实际场景,肯定是需要根据实际场景做调整的

1qingqing1 commented 2 years ago

请问一般如何根据实际场景做调整呢,可否传授一些经验