2020-04-28 13:58:21 -0600 | commented answer | Offloading ANN GEMM operations to GPU I appreciate the quick response. I was not familiar with the cmake stucture in OpenCV so this helps! I agree, probably |
2020-04-28 13:55:20 -0600 | marked best answer | Offloading ANN GEMM operations to GPU I am trying to add functionality into the ANN_MLP object to offload the gemm operations in the predict function onto the GPU. I have a confirmed working build with NVIDIA CUDA able to query the GPU from OpenCV. Additionally, the ANN_MLP predict functionality is working properly. What I did was add an identical function to "predict" called "predict_cuda" in the ANN_MLP source code. I then replaced the gemm(...) call with cv::cuda::gemm(...) and included the header to the cv::cuda::gemm reference (cudaarithm.hpp). In the ml.hpp header, I added the predict_cuda signature to the ANN_MLP object reference. When running make in the build folder, I run into the issue: ../../lib/libopencv_ml.so.4.3.0: undefined reference to since the CUDA object is built later on and not linked with this target. Is there an easy way to configure cmake/make to build with this additional reference? Or am I missing something completely different? My goal is to accelerate the ANN_MLP I have with NVIDIA CUDA; however, there is no direct ANN_MLP on GPU support. So, I have decided to try and integrate the cv::cuda::gemm operations into the OpenCV source. |
2020-04-28 13:55:20 -0600 | received badge | ● Scholar (source) |
2020-04-28 02:38:14 -0600 | asked a question | Offloading ANN GEMM operations to GPU Offloading ANN GEMM operations to GPU I am trying to add functionality into the ANN_MLP object to offload the gemm opera |