This is not the only OpenCL fork of Tensorflow available.
it is developed on Ubuntu 16.04 (using Intel HD5500, and NVIDIA GPUs) and Mac Sierra (using Intel HD 530, and Radeon Pro 450). At least, StochasticGradientDescent trainer is working, and the others are commited, but not yet tested reductions, argmin, argmax, again using Eigen, as per earlier info and links. blas / matrix-multiplication, using Cedric Nugteren's CLBlast. per-element operations, using Eigen over OpenCL, (more info at ). for now, the following functionalities are implemented:. It's a very general goal, and a very general compiler cuda-on-cl targets to be able to take any NVIDIA® CUDA™ soure-code, and compile it for OpenCL 1.2 devices. it's based on an underlying library called 'cuda-on-cl',. 
It doesnt need OpenCL 2.0, doesnt need SPIR-V, or SPIR.
it targets any/all OpenCL 1.2 devices. This fork of tensorflow for OpenCL has the following characteristics: I'm writing an OpenCL 1.2 backend for Tensorflow at
0 kommentar(er)
