chen wei
chen wei
The output loss value: ``` step 3600, training loss 1.00463 step 3700, training loss 0.970242 step 3800, training loss 0.906492 step 3900, training loss 0.0988686 step 4000, training loss 0.00080747...
Thanks! I know nothing about mxnet. I want to test the mxnet code. When I run the program. I got the following output AttributeError: module 'mxnet.symbol' has no attribute 'LSoftmax'...
I found it is really hard to convergence. I compared my output cos_m_t and other parameters. We have same output, but my model can't convergence. I changed the m value...
Does the definition of `J` equals the equation `(4)` in original paper? I can't understand the derivation from `1` to `2` in your [latex version](https://luoyetx.github.io/2017/05/derivatives-for-lsoftmax/#derivatives-for-lsoftmax). I have wrote some definition...
I have uploaded two new model, you can test with the new `ckpt` file.
It's using `resnet` in `L_Resnet_E_IR.py`.
The face image is not aligned. The image is already processed in the `insightface` project. The default image align method is `mtcnn`.
I don't know why, so I added `Max Batch Size Test`.
Could you test the model size after adding the `prelu` and adding the same numbers of `Batch Norm` layers as the paper.
@HsuTzuJen Thanks for you compare the different models.