Jongsoo Park

Got it. This is a good point. I'll take a look.

Oops! I forgot to add those files. Thanks for pointing out. Now the necessary files to reproduce resnet-50 pruning results should be there. Please let me know if you encounter...

Hi, good question. Even if pruned model has lots of zeros we didn't take advantage of that when storing to files. This is because our main goal was reducing inference...

Regarding speed up in GPU, I didn't optimize for GPU inference speed using the sparsity. In Xeon and Xeon Phi, you should be able to see speedups.

Technically, yes. ARM also should be able to take advantage of sparsity to speed up inference but you'd need to implement an optimized sparse convolution similar to the one for...

Thanks for reporting the issue. Please try with a new MKL version (>=17.0) . Sorry about the trouble. I'll change the code so that you can compile without MKL DNN...

You're right that SkimCaffe is already using libxsmm but only for sparse-matrix-times-dense-matrix (SpMDM) multiplication for fully-connected layers. Unfortunately, sparse convolution routine (in include/caffe/utils/sconv.hpp) has not been integrated into libxsmm. It...

Thanks for reporting this issue. Unfortunately, SkimCaffe doesn't support doing inference in GPU and I'm happy to help if anyone wants to contribute that part. The first step would be...

Sorry about late reply. The error is generated from gflags and probably the command line arguments are not parsed correctly. What's the version of gflags you're using? Can you check...

It would be also helpful to know if you're getting the same error if you use the Caffe main branch.