extractor was not included in the optimizer parameter.
So, I think the extractor was only used for inference, but wonder why used train().
In this case, batch_norm different results than eval().
because running_mean, running_var changed when train().
Hi, Thanks for your code!!
extractor was not included in the optimizer parameter. So, I think the extractor was only used for inference, but wonder why used train(). In this case, batch_norm different results than eval(). because running_mean, running_var changed when train().
https://github.com/YoungGod/DFR/blob/530779514bba281acb4f9be2c1c96a1b41909cbb/DFR-source/anoseg_dfr.py#L148
https://github.com/YoungGod/DFR/blob/530779514bba281acb4f9be2c1c96a1b41909cbb/DFR-source/anoseg_dfr.py#L207
https://github.com/YoungGod/DFR/blob/530779514bba281acb4f9be2c1c96a1b41909cbb/DFR-source/anoseg_dfr.py#L78