Optimizing CNN Model Inference on CPUs — mentioned by ajtulloch

Tue, 26 Mar 2019 03:00:55 GMT

FWIW, I believe that the current state of the art for batch-size 1, fp32 inference for ResNet-50 on Intel CPUs is AWS's work in https://arxiv.org/abs/1809.02697. After the low-hanging fruit outside of model execution are picked, this kind of work is probably quite relevant.

Paper and discussion history. A mention is not an endorsement.

Optimizing CNN Model Inference on CPUs · HackerPapers discussions

Optimizing CNN Model Inference on CPUs — mentioned by ajtulloch