search

mila-iqia / myia

Myia prototyping
MIT License
455 stars 46 forks source link

Any roadmap for CPU acceleration such as by MKL-DNN #77

Open pengzhao-intel opened 6 years ago

pengzhao-intel commented 6 years ago

MXNET used NNVM with MKL-DNN backend for the CPU acceleration and got a very nice speedup as below.

https://github.com/apache/incubator-mxnet/releases/tag/1.2.0

On average, after the MKL-DNN change, the inference speed of MXNet + MKLDNN outperforms MXNet + OpenBLAS by a factor of 32, outperforms MXNet + MKLML by 82% and outperforms MXNet + MKLML with the experimental flag by 8%. The experiments were run for the image classifcation example, for different networks and different batch sizes.

Do you have any roadmap for CPU operators?

abergeron commented 6 years ago

For the runtime implementations of operators we will delegate to things like NNVM, TensorComprehensions or others, so we should benefit from anything they do.