Dear TF community people,
I am looking for a 3D depthwise separable convolution. Apparently, the current version of TF does not include this implementation (If I misunderstand, please let me know). I am wondering if there is any way to use 3D depthwise separable convolution using Tensorflow and Keras. All I was able to find is the following one from GitHub: GitHub - alexandrosstergiou/keras-DepthwiseConv3D: Keras w/ Tensorflow backend implementation for 3D channel-wise convolutions
Regards,
Evan