-
Quansight
- Scotland
Highlights
- Pro
Block or Report
Block or report peterbell10
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
895 contributions in the last year
Activity overview
Contribution activity
June 2022
Created 6 commits in 1 repository
Created a pull request in pytorch/pytorch that received 10 comments
Exploit symmetry in comparison operators to reduce no. of kernels
Stack from ghstack (oldest at bottom):
-> #78990
#78989
gpu_kernel_with_scalars generates 3 gpu_kernel calls to compute
f(a, b) where either a or…
+52
−13
•
10
comments
Opened 9 other pull requests in 1 repository
pytorch/pytorch
7
open
2
closed
- Use cub::BlockRadixSort to improve medium length sort performance
- Improve small sort performance on CUDA
- Support non-standard bools in mode CUDA kernels
- Support non-standard bools in CUDA unique
- Support non-standard bools in masked_scatter CUDA
- Fix remaining CPU operators for non-standard bools
- Add OpInfo for torch.equal and fix support for non-standard bools
- Add symmetric version of gpu_kernel_with_scalars
- Accept non-standard bools in more CUDA kernels



