Improving DL Performance Using Binary Convolution Support in OpenVINO Toolkit – Intel AI

Improving DL Performance Using Binary Convolution Support in OpenVINO Toolkit – Intel AI

Improving DL Performance Using Binary Convolution Support in OpenVINO Toolkit – Intel AI

One way to accelerate DL networks is to replace standard convolutions with binary convolutions. To provide more DL training capabilities to the OpenVINO community, we are releasing support of binary models in the Neural Network Compression Framework

Source: www.intel.ai/binary-convolution-openvino/

Subscribe to our Digest