Question for eval_BreastPathQ_SSL.py line 123.

srinidhiPY / SSL_CR_Histo

Official code for "Self-Supervised driven Consistency Training for Annotation Efficient Histopathology Image Analysis" Published in Medical Image Analysis (MedIA) Journal, Oct, 2021.

https://doi.org/10.1016/j.media.2021.102256

MIT License

62 stars 21 forks source link

Question for eval_BreastPathQ_SSL.py line 123. #5

Closed hust-linyi closed 2 years ago

hust-linyi commented 2 years ago

Hi, thanks for sharing the code! Amazing work!

In eval_BreastPathQ_SSL.py line 123, https://github.com/srinidhiPY/SSL_CR_Histo/blob/adc301ec74b69defc51cfe2ab3effccc7c1f60c5/eval_BreastPathQ_SSL.py#L123

target = target.long() It could translate the float tensor into int64, thus the val_loss could be wrong. Is this a bug here?

Thanks

srinidhiPY commented 2 years ago

Hi, Really sorry for the late reply.

I looked at the issue in line 123, and it's an unintentional bug over here. It should have been the target.float() (as in Line 55) instead of target.long() data type. Since the pretrained model is made available you could use these scripts to fine-tune the model again. Let me know if you face any issues. Thanks