A fork of the C++ deep learning framework Caffe that adds a compress command which roughly implements the NIPS 2015 paper Learning both Weights and Connections for Efficient Neural Networks by Han et al. Handles fully connected layers only.
A fork of the C++ deep learning framework Caffe that adds a compress command which roughly implements the NIPS 2015 paper Learning both Weights and Connections for Efficient Neural Networks by Han et al. Handles fully connected layers only.