Hi, today I have a C and OpenCL code fine-tuning problem to think about. This problem is quite complex to solve and truth be told I do not know how and why it does not work. The example is from real project at https://github.com/sowson/darknet. Let me first show you slow to compute version of example method in OpenCL. To give you all the aspects of computation you need to see also C code that invokes this one Continue ReadingGPU OpenCL Fine-Tuning Problem
Hi, today I will show you some measurements results for my PhD. I am working of the first publication about DarkNet on OpenCL, the source code of this project you can find at https://github.com/sowson/darknet. The IEEE publication has to be consistent and smart. I cannot put on it too much graphics and big tables… but waits, I have a public blog site. So, I can post it here. First things first the battle heroes come on the Continue ReadingDarkNet CUDA vs OpenCL and CPU vs NVIDIA vs AMD
Hi, today I have something very nice for all readers. The subject is the GPU Computing on OpenCL for all readers. Abstract: Have you ever wanted to sum 10 million random numbers in array in C? Do you know that on GPU it can be faster about 20k times than on CPU? This post is about GPU Computing and all examples were written in C language with OpenCL 1.2+. Basic understanding of C language is necessary to Continue ReadingGPU Computing on OpenCL
Hi, thanks to Dr. Filip I realized that processing of video stream and write to the Console the same time is wrong. Because Console is slow :D… Thanks, Dr. Filip! p ;).
Hi, today I would like to share with you the setup of DarkNet in OpenCL that works fine on Asus Thinker Board S. Here it is the setup. mkdir github cd github git clone https://github.com/krrishnarraj/libopencl-stub cd libopencl-stub make sudo cp -r include/* /usr/include sudo cp libOpenCL.a /usr/lib cd .. git clone https://github.com/sowson/darknet.git cd darknet # change in Makefile OPENCV=0 and RPI=1 make As an example let me show to you Yolo3-Tiny computation And here it is the Continue ReadingDarkNet in OpenCL on Asus Thinker Board S
Hi, today I would like to share with you a great talk given by Lisa McInnes-Smith on the TEDx Melbourne in 2015 I believe. I received a link for her performance from a very good friend who wants to point out that you only learn when the pain is involved in the process. Thanks! p ;).
Hi, today again I advise my good friend what to do with his career journey. Again, and it seems to me that I am becoming more and more the Coach and the Mentor to people than just software creator and technical expert. I want to write about the career advisor role I sometimes got to some of my friends. The thing is that when I was at the beginning of my career I was looking for the Continue ReadingCareer Advisor
Hi, Today I would like to announce that my GitHub fork at https://github.com/sowson/darknet has a new update, the fork is an advanced port of DarkNet CNN from CUDA to OpenCL and tested on macOS with eGPU from Sonnet named Breakaway RX 570 Puck and on my GreenPC. Let’s get started the training by the inspiration from the solution original author Joseph Redmon given the TED talk. Interested how it works and how to rebuild? Why I spent some Continue ReadingDarkNet Training
Hi, Today I would like to announce that my GitHub fork at https://github.com/sowson/darknet has a new update, the fork is a advanced port of DarkNet CNN from CUDA to OpenCL and tested on macOS with eGPU from Sonnet named Breakaway RX 570 Puck. You can enjoy my coding work on this subject by cloning the GIT repository. I only want to show you two pictures of detections on my GreenPC. YoloV2 [gamer@dreampc darknet]$ ./darknet detect cfg/yolov2.cfg ../../weights/yolov2.weights data/dog.jpgDevice Continue ReadingDarkNet CNN in OpenCL on macOS with Yolo3
Yes, I know that it is impossible to predict lottery results. But if the results would be predictable? Of course I know they are not, but just hypothetically for a play with Perceptron Neural Network and prediction algorithms I want to show you that in 256 ;-) lines of code you are able to play with prediction thanks to EnCog 3.4 library. I bought recently 2 Jeff Heaton’s books about neural networks and C# and his library. Continue ReadingEuroJackpot with EnCog ANN and Stats