Catchher
Catchher
Aha，I meet the same issue, particularly in Optical-Flow branch with only 83% acc.
Thanks for ur reply, I'm looking forward to ur good news. @vra
THANKS. I got it.
Yes, But it seems not easy for me. I modified the code in next_step_selector.py, and I guess i make a mistake, because I don't know what's the correct steps to...
But I'll go ahead and implement it, any suggestion would be helpful. THANKS.
@AmedeoSapio Hi, I modified the relevant code according to your suggestions, and then successfully ran the test program, and the data verification of the test program passed. 2 folding is...
I have implemented 2-folded pipeline successfully. Thanks. @AmedeoSapio
> halo, any tricks? No tricks, only increased the batchsize and learning rate to 128 and 1e-1 respectively.
pre-trained model, this model trained by multi-gpu, when you test it with a single gpu, please remove 'module.' in any variable name. https://pan.baidu.com/s/1jZQ2dp_4YR5TvA4ARkKZ-g code: gh6e
> i am currently new in face recognition project, i want to ask what is actual difference between your model and this rep provided? does the model also improve accuracy...