Using CMDSTAN with OPENCL/GPU for prophet

[rishabh1212]

As of now prophet uses CPUs I have been successfully able to use GPU/OPENCL compiled model for inferences https://github.com/rishabh1212/stan-gpu-docker My sugestion is to make new installation option which make use of OPENCL, I want to make PR for it if you think it's a good idea

[rishabh1212]

@bletham Can you check this and recommend?

[tcuongd]

Nice one! I guess we can allow the user to pass a command line flag to use GPU / OpenCL for compilation. There's really no downside to this if it's optional (provided all the tests still pass with the GPU-compiled binary), so happy to review the PR if you'd like to work on it :)

[rishabh1212]

Ok, I will make a PR

[xiaomaomaomm]

Any updates about the PR？

[ValentinPeron]

Any updates about the PR？

[donggu-kang]

PR에 대한 업데이트가 있나요?

+1