How can can I speed up new.forward() process in OpenCV in Python having CUDA?
when I write
net.setPreferableTarget(cv2.dnn.DNN_TARGET_OPENCL)
I'm getting speed twice as slow And show up next message as output:
OpenCV(ocl4dnn): consider to specify kernel configuration cache directory
via OPENCV_OCL4DNN_CONFIG_PATH parameter.
OpenCL program build log: dnn/dummy
Status -11: CL_BUILD_PROGRAM_FAILURE
-cl-no-subgroup-ifp
Error in processing command line: Don't understand command line argument "-cl-no-subgroup-ifp"!
I've got GeForce GTX 1050 and Cuda 10