Practical tutorial how to implement Deep Compression for a network in Caffe

[kamadforge]

Is there any practical (step by step) tutorial how to speed up the network in Caffe in a way similar to the paper on the Deep Compresion or the alike. There is an issue #4294 that touches on pruning by means of using a sparse layer but I am not sure about the details (e.g. prototxt sample) how to use it. Also, here is a link on quantization with TensorFlow. Is it possible to do it in Caffe, as well?

[shelhamer]

There is not currently a bundled tutorial on how to do model compression, though there are several options. Please ask modeling and usage questions on the mailing list.

From https://github.com/BVLC/caffe/blob/master/CONTRIBUTING.md:

Please do not post usage, installation, or modeling questions, or other requests for help to Issues.
Use the caffe-users list instead. This helps developers maintain a clear, uncluttered, and efficient view of the state of Caffe.

[yiakwy]

This is a great direction where we can contribute to !!!