MKL: update MKL-DNN version to not compute bias for BWD_F

baidu-research / DeepBench

Benchmarking Deep Learning operations on different hardware

Apache License 2.0

1.07k stars 239 forks source link

MKL: update MKL-DNN version to not compute bias for BWD_F #33

Closed rsdubtso closed 7 years ago

rsdubtso commented 7 years ago

The original driver for MKL-DNN computed both gradients for weights and bias for BWD_F. This patch disables bias gradient computations by default. The aim of this change is to make it easier to compare MKL and MKL-DNN performance.

dmudiger commented 7 years ago

Looks OK to me, merged.