Uploaded using `kernel-builder`.
Browse files
README.md
CHANGED
|
@@ -12,13 +12,12 @@ This is the repository card of kernels-community/layer-norm that has been pushed
|
|
| 12 |
from kernels import get_kernel
|
| 13 |
|
| 14 |
kernel_module = get_kernel("kernels-community/layer-norm")
|
| 15 |
-
|
| 16 |
|
| 17 |
-
|
| 18 |
```
|
| 19 |
|
| 20 |
## Available functions
|
| 21 |
-
- `layers`
|
| 22 |
- `dropout_add_ln_fwd`
|
| 23 |
- `dropout_add_ln_bwd`
|
| 24 |
- `dropout_add_ln_parallel_residual_fwd`
|
|
@@ -27,14 +26,3 @@ layers(...)
|
|
| 27 |
## Benchmarks
|
| 28 |
|
| 29 |
Benchmarking script is available for this kernel. Run `kernels benchmark kernels-community/layer-norm`.
|
| 30 |
-
|
| 31 |
-
### Performance
|
| 32 |
-
|
| 33 |
-
<img class="dark:hidden border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_light_animation.svg" />
|
| 34 |
-
<img class="hidden dark:block border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_dark_animation.svg" />
|
| 35 |
-
|
| 36 |
-
<img class="dark:hidden border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_light_latency.svg" />
|
| 37 |
-
<img class="hidden dark:block border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_dark_latency.svg" />
|
| 38 |
-
|
| 39 |
-
<img class="dark:hidden border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_light_throughput.svg" />
|
| 40 |
-
<img class="hidden dark:block border border-gray-200 dark:border-gray-700 rounded-lg" src="media/benches_dark_throughput.svg" />
|
|
|
|
| 12 |
from kernels import get_kernel
|
| 13 |
|
| 14 |
kernel_module = get_kernel("kernels-community/layer-norm")
|
| 15 |
+
dropout_add_ln_fwd = kernel_module.dropout_add_ln_fwd
|
| 16 |
|
| 17 |
+
dropout_add_ln_fwd(...)
|
| 18 |
```
|
| 19 |
|
| 20 |
## Available functions
|
|
|
|
| 21 |
- `dropout_add_ln_fwd`
|
| 22 |
- `dropout_add_ln_bwd`
|
| 23 |
- `dropout_add_ln_parallel_residual_fwd`
|
|
|
|
| 26 |
## Benchmarks
|
| 27 |
|
| 28 |
Benchmarking script is available for this kernel. Run `kernels benchmark kernels-community/layer-norm`.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|