Some records of personal CUDA kernel implementations. These implementations are not best optimized and mainly for learning purposes.
- Softmax
- ReLU
- GEMM
- LayerNorm
More kernels are coming...
-
Notifications
You must be signed in to change notification settings - Fork 0
KarhouTam/cuda-kernels
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Some common CUDA kernel implementations (Not the fastest).