Closed Erkaman closed 8 years ago
Also a few to add:
I am currently working on implementing some GPU primitives. Right now I am working on
Adding some references that may come in handy:
Adding some references for implementing Deep Learning:
Useful references for GPU sorting:
I shall give implementing deep learning a try. To do that, I will first need to implement these two components on the GPU
So I wrote a demo that does MNIST classification on the GPU: https://github.com/Erkaman/regl-cnn
It shows both how to implement convolution and matrix multiplication on the GPU.
We have some examples. I'm gonna close this since it is too vague.
I know some existing examples that I found useful https://github.com/sugeerth/pycuda
On Fri, Oct 7, 2016 at 3:32 PM, Mikola Lysenko notifications@github.com wrote:
Closed #164 https://github.com/regl-project/regl/issues/164.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/regl-project/regl/issues/164#event-816874891, or mute the thread https://github.com/notifications/unsubscribe-auth/AD8LYOgFh2f8XmXNNVdoA651f0GJ9fKwks5qxsiKgaJpZM4JY-Hx .
Right now, there are not many examples that shows how regl can be used for GPGPU. But since regl greatly simplifies the usage of fbos, easy GPGPU is a great selling point of the framework, so we need good examples for it. Some possible examples are:
And anyone who has implemented such an example is very much welcome to submit it, by a pull request!