-
Hello, I found Roland's paper to be quite insightful and have been exploring the accompanying code.
I have a question regarding **"Algorithm 3"** mentioned in the paper, specifically about the ins…
-
I have tried some of the algorithms like svm and knn but it seems too slow to run on my computer.
Are there any ways to make those algorithms running faster?
-
A number of examples:
- [ ] disable algorithms not yet implemented in GPU when the GPU flag is checked
- [ ] disable algorithms that do not support multi-mode (or multi-slice) if the corresponding…
-
The test_gpu.cpp example included in the testsuite builds with a non-gpu variant of hpx, but it fails with both rocm and cuda variants. There are cuda specific api calls in the code so I'm not surpris…
-
Make a new library above RAJA with common views, containers without dynamic storage, and sequential algorithms that work on GPUs, a lot like this https://github.com/nvidia/libcudacxx.
These are mainl…
-
When tuning is active during Multi-GPU runs each GPU independently tunes each Kernel. This results in different GPUs using different launch configurations for the final Kernel launch and finally makes…
-
### Description
I'm training learned optimizers using Jax and a custom version of https://github.com/google/learned_optimization. I get the following warnings when training on Ampere GPUs (tested f…
-
# Creation of benchmarks for the benchmarks server
This document is a compilation of benchmarks suggested by maintainers and contributors from all areas of the engine.
Our goal is to create a li…
reduz updated
4 months ago
-
Developers should be able to specify the requirements for their algorithms. One should be able to specify the minimum memory, number of GPUs needed, as well as input and output requirements for their …
-
### Describe the request
Add the ability to render in, for example, 1080p and upscale to 4k.
This will be useful for those wishing to use more eye-candy settings on less-powerful GPUs. In my cas…