search

zijundeng / BDRAR

Code for the ECCV 2018 paper "Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection"
123 stars 29 forks source link

About performance of trained model. #7

Open zhangchbin opened 5 years ago

zhangchbin commented 5 years ago

Hi @zijundeng , I have loaded your pre-trained model and inference, but I got Ber = 5.20. I have ensured that my evaluation code is right, while I used your network, data loader and other. I don't know whether is there some problems with your pre-trained model or other. Thanks.

aaronjie commented 4 years ago

@zhangchbin I got a similar result. Do you solve this problem now? thanks :)

zhangchbin commented 4 years ago

@zhangchbin I got a similar result. Do you solve this problem now? thanks :)

I have solved this problem, because the difference between pytorch1.0 and pytorch0.4.0, you can change the version.

aaronjie commented 4 years ago

@zhangchbin I got a similar result. Do you solve this problem now? thanks :)

I have solved this problem, because the difference between pytorch1.0 and pytorch0.4.0, you can change the version.

Thank you very much for your quick reply. I tried this morning with pytorch 0.4.0, however, it seems useless. I don't know whether it is related to my testing code. If possible, would you please kindly share me with your testing code? Thanks again. My email address is：914181182@qq.com

zhangchbin commented 4 years ago

@zhangchbin I got a similar result. Do you solve this problem now? thanks :)

I have solved this problem, because the difference between pytorch1.0 and pytorch0.4.0, you can change the version.

Thank you very much for your quick reply. I tried this morning with pytorch 0.4.0, however, it seems useless. I don't know whether it is related to my testing code. If possible, would you please kindly share me with your testing code? Thanks again. My email address is：914181182@qq.com

You can refer to my repo 'evaluation-for-shadow-detection'. If it is useful to you, please star it, thanks.

aaronjie commented 4 years ago

@zhangchbin I got a similar result. Do you solve this problem now? thanks :)

I have solved this problem, because the difference between pytorch1.0 and pytorch0.4.0, you can change the version.

Thank you very much for your quick reply. I tried this morning with pytorch 0.4.0, however, it seems useless. I don't know whether it is related to my testing code. If possible, would you please kindly share me with your testing code? Thanks again. My email address is：914181182@qq.com

You can refer to my repo 'evaluation-for-shadow-detection'. If it is useful to you, please star it, thanks.

Thank you so much :)