huangyanpei/VLA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[lint] Update lint (#16)

Browse Source 
* update lint

* update readme

* update ruff lint
This commit is contained in:
Lufang Chen
2025-09-11 13:18:33 +08:00
committed by GitHub
parent a89dce95aa
commit e9332a283d
28 changed files with 2406 additions and 1074 deletions
Download Patch File Download Diff File Expand all files Collapse all files
csrc/README.md
+2 -2
View File
@@ -10,7 +10,7 @@ High-performance CUDA kernels for accelerating model training.
### Token Permutation
- `permute`: Token permutation for MoE routing
- `unpermute`: Token recovery after expert computation
- `unpermute`: Token recovery after expert computation
- `unpermute_bwd`: Backward pass for token recovery
### Multimodal RoPE
@@ -20,4 +20,4 @@ High-performance CUDA kernels for accelerating model training.
## Acknowledgments
The `permute` and `unpermute` operators are adapted from [fanshiqing/grouped_gemm](https://github.com/fanshiqing/grouped_gemm). Thanks for their open-source contributions.
The `permute` and `unpermute` operators are adapted from [fanshiqing/grouped_gemm](https://github.com/fanshiqing/grouped_gemm). Thanks for their open-source contributions.
csrc/ops.cu
-1
View File
@@ -13,4 +13,3 @@ PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("rope", &launch_multimodal_rope_forward, "Multimodal RoPE forward kernel");
m.def("rope_bwd", &launch_multimodal_rope_backward, "Multimodal RoPE backward kernel");
}
csrc/rope.cu
-1
View File
@@ -549,4 +549,3 @@ void launch_multimodal_rope_backward(
break;
}
}