Thanks for your code, but I am confused about some parameters during training wider fa

<a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/us

Some parameters are confused for training wider face about mxnet-ssh HOT 3 OPEN

deepinsight commented on September 21, 2024

Some parameters are confused for training wider face

from mxnet-ssh.

hdjsjyl commented on September 21, 2024

I see the parameter 'color_jitter' for data augmentation, it is not used. correct?

from mxnet-ssh.

nttstar commented on September 21, 2024

It will end at epoch-80.
Softmax already did per-instance normalization.
Fixed parameters are correct.
Color_jitter is required.

from mxnet-ssh.

hdjsjyl commented on September 21, 2024

@nttstar Thanks for your reply. This information is important to me. Thanks. Other questions:

I found that the program runs fast. And when I run the code with not good cpu and gpu, the gpu utility is always near to 100%. It is very good, can you explain it?
Did you use focal loss to train the wider face? How is the result?
This is an excellent work, thank you very much.

from mxnet-ssh.