potok/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-07-30 15:53:03 +04:00
Code Packages Releases Wiki Activity

update convolution opencl kernels in dnn module (#11762)

Browse Source 
* optimize ocl kernel enqueue in fc layer

Signed-off-by: Li Peng <peng.li@intel.com>

* use CV_LOG_INFO in convolution auto tuning

Signed-off-by: Li Peng <peng.li@intel.com>

* update convolution IDLF kernel

extend parameter tuning range, also cleanup
ocl kernel implementation

Signed-off-by: Li Peng <peng.li@intel.com>

* update in-memory convolution cache config

fp16 and fp32 cache config are stored separately

Signed-off-by: Li Peng <peng.li@intel.com>
This commit is contained in:
Li, Peng
2018-06-25 22:06:18 +08:00
committed by Vadim Pisarevsky
parent a2bc075924
commit ab8022f74e
4 changed files with 989 additions and 590 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/dnn/src/layers/fully_connected_layer.cpp
+2 -2
View File
@@ -310,7 +310,6 @@ public:
innerProductOp = Ptr<OCL4DNNInnerProduct<float> >(new OCL4DNNInnerProduct<float>(config));
}
UMat biasOnesMat = UMat::ones(outerSize, 1, umat_blobs[0].type());
for (size_t i = 0; i < inputs.size(); i++)
{
MatShape inshape, outshape;
@@ -320,7 +319,6 @@ public:
UMat srcMat, dstMat;
srcMat = inputs[i].reshape(1, inshape.size(), &inshape[0]);
dstMat = outputs[i].reshape(1, outshape.size(), &outshape[0]);
dstMat.setTo(0.0f);
if (!innerProductOp->Forward(srcMat, (use_half) ? half_blobs[0] : umat_blobs[0],
(bias) ? (use_half ? half_blobs[1] : umat_blobs[1]) : UMat(),
@@ -332,6 +330,7 @@ public:
if (!use_half && bias && (outerSize > 1))
{
UMat biasOnesMat = UMat::ones(outerSize, 1, umat_blobs[0].type());
UMat& biases = umat_blobs[1];
cv::gemm(biasOnesMat, biases, 1, dstMat, 1, dstMat, 0);
}
@@ -354,6 +353,7 @@ public:
if (bias)
{
UMat biasOnesMat = UMat::ones(outerSize, 1, umat_blobs[0].type());
UMat& biases = umat_blobs[1];
cv::gemm(biasOnesMat, biases, 1, dstMat, 1, dstMat, 0);
}