modules/dnn/src/layers/fully_connected_layer.cpp · 5a3bca8d5c54dad3b7ec98814616772675a9ac83 · submodule / opencv

Merge pull request #12565 from dkurt:dnn_non_intel_gpu · 24ab7515

Dmitry Kurtaev authored Sep 26, 2018

* Remove isIntel check from deep learning layers

* Remove fp16->fp32 fallbacks where it's not necessary

* Fix Kernel::run to prevent localsize > globalsize

24ab7515

fully_connected_layer.cpp 17.5 KB

Replace fully_connected_layer.cpp