zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-09-03 22:34:43 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

[CUDA] Add GEMM-based fallback convolution kernels (#2511)

Browse Source 
* Add gemm_conv

* Add gemm_grouped_conv
This commit is contained in:
Cheng
2025-08-20 10:06:22 +09:00
committed by GitHub
parent 65d0d40232
commit ac85ddfdb7
8 changed files with 667 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
python/tests/cuda_skip.py
View File

@@ -15,14 +15,6 @@ cuda_skip = {
"TestOps.test_hadamard_grad_vmap",
# Convolutions NYI
"TestConv.test_1d_conv_with_2d",
"TestConv.test_conv_1d_groups_flipped",
"TestConv.test_conv_general_flip_grad",
"TestConv.test_torch_conv_2D",
"TestConv.test_torch_conv_depthwise",
"TestConv.test_torch_conv_general",
"TestConvTranspose.test_torch_conv_transpose_1D_grad",
"TestConvTranspose.test_torch_conv_transpose_2D_grad",
"TestConvTranspose.test_torch_conv_transpose_3D_grad",
# FFTs NYI
"TestFFT.test_fft",
"TestFFT.test_fft_big_powers_of_two",