yang0817manman

which knowledge-distillation loss is best in "kd_losses "?

1

thanks your shariing. Can you tell me which lossfunction is best in " "kd_losses " for classification task?

hello,thanks for sharing your code. Can you tell me there is wheather goog strategy for trainning large scale dataset. I have trained sphereface-20 with ms-celeb-1m.Then I finetune spherefacenet20_ms-celeb-1m.caffemodel with VGGFace2...

how to debug your code？

4

Hi , thanks for sharing your work. I want to train my traindataset but the error appeared all the time ,then I debug the code and the following text always...

seek forhelp

2

hello,thank you for sharing your code. I am interesting in TP-GAN and want to run a test, can you provide train dataset or a pretrained modelfor me? Thank you

how to get "data/Out/BFM.mat?"

3

Thanks for your work. I send qq e-mail , but i did not accept reply all the time.Can you share your resource and help me? thx.

Hi，thanks for sharing your work . I downloaded your sharing the model "MobileResNet50_merge_v2" and tested my test dataset, the result is surprising, so I am interested in this network and...

inital loss is nan?

3

非常感谢你的分享。为什么我用你分享的loss进行试验，初始loss一直都是nan，无论怎么修改学习率都是nan？请问这是什么原因？

about"the same shape, except at concat_axis"?

1

helloe, thanks for sharing your work. I want to train my dataset with providing network ,but the error appeared. I merely modified batchsize. the error is following: "I1130 16:59:28.788107 13376...