RandolphVI / Multi-Label-Text-Classification

About Muti-Label Text Classification Based on Neural Network.
Apache License 2.0
553 stars 143 forks source link

How to use f1_score from sklearn? #5

Closed saja1994 closed 4 years ago

saja1994 commented 5 years ago

Thank you for sharing this helpful code. please, I want to ask you if you can help me to implement f1_score from sklearn into your implementation to measure the f score because I got low f1 values using your code and want to try different f1 average ways like macro and micro, and compare them.

Thanks a lot.

RandolphVI commented 5 years ago

Hi @saja1994 ,

What you mentioned, implement f1_score() from sklearn, it’s important since users want to calculate the micro-f1 or macro (actually what I implement is called sampled-f1, but is not used basically).

And I have implemented the code already, and I will update the code in one or two days (a little bit busy recently, ddl).

Hope this helps!

saja1994 commented 5 years ago

Thank you so much, I really appreciate your effort.

RandolphVI commented 5 years ago

Hi @saja1994 ,

I have updated the code already, now you can predict the evaluation metrics using sklearn.

saja1994 commented 5 years ago

Thank you so much.