James Fung

I will take a look

Multichannel is not supported, because some of the effect don't make sense in multichannel setup, for example, cross-feed and quad / cross channel convolution are one of those

@tuan3w The only fatal error of your implementation is concatenation. [https://github.com/deezer/spleeter/blob/39af9502ab1156c013f17f8d8cd1c53d46459857/spleeter/model/functions/unet.py#L127](https://github.com/deezer/spleeter/blob/39af9502ab1156c013f17f8d8cd1c53d46459857/spleeter/model/functions/unet.py#L127) Each U-Net encoder **convolutional layer** output is being concated with decoder output. We are not concatenating the encoder batch...

@tuan3w What's the MSE/MAE of output mask between your output and official Spleeter(Tensorflow)? If the mask function is identical or similar (1e-3), then you are implement absolutely correct. You don't...

Recently is busy on my projects, may be help you to find remaining bugs in the future.

@generalwave 据我经验Tensorflow和Matlab的Padding几乎无别。 至于Pytorch和Tensorflow的区别，除了Padding='same'外的特例我不知道。 我试过将SRGAN Pytorch的CNN系数转到Matlab里，两者预测的结果是一致。 就算Tensorflow和Pytorch的Padding不一样，理论上完全能预补零解决。 Spleeter官方没公开训练集，重头训练并匹配原论文的结果是没可能。

可能SRGAN那方的Padding刚好导致输入输出大小一样，所以和Matlab的'same'无别，所以结果吻合。 个人在C实现TF或Pytorch的CNN都没问题，设好stride, padding, dilation和**offset**，然后送去im2col(),gemm()就ok

What's the compilation status and runtime status of the AIDL thingy?