mirror of
https://github.com/ml-explore/mlx.git
synced 2025-07-28 21:21:21 +08:00
6f5874a2f2
* Link with cuDNN * Initial implementation * Remove backend apis * Fix recording cudnn conv * More unused backend apis * Fix C++ conv tests * include cudnn as python dep * Install libcudnn9-dev-cuda-12 in CI * cudnn only accepts contiguous inputs * Switch to backend apis * Plan needs to be kept alive * Turn off tf32 * Add cache * Test the native cuda graph api * Set cudnn stream before execution * Make LRUCache more like a normal container * Do error check for cublas handle * Zero-initilizing array * Use tf32 for conv * Skip TestConv.test_torch_conv_2D test --------- Co-authored-by: Awni Hannun <awni@apple.com>
78 lines
2.8 KiB
Python
78 lines
2.8 KiB
Python
cuda_skip = {
|
|
"TestLoad.test_load_f8_e4m3",
|
|
"TestLayers.test_quantized_embedding",
|
|
"TestOps.test_dynamic_slicing",
|
|
# Block masked matmul NYI
|
|
"TestBlas.test_block_masked_matmul",
|
|
# Gather matmul NYI
|
|
"TestBlas.test_gather_matmul",
|
|
"TestBlas.test_gather_matmul_grad",
|
|
"TestBlas.test_gather_mm_sorted",
|
|
# Segmented matmul NYI
|
|
"TestBlas.test_segmented_mm",
|
|
# Hadamard NYI
|
|
"TestOps.test_hadamard",
|
|
"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_conv_groups_grad",
|
|
"TestConv.test_torch_conv_1D_grad",
|
|
"TestConv.test_torch_conv_2D",
|
|
"TestConv.test_torch_conv_2D_grad",
|
|
"TestConv.test_torch_conv_3D_grad",
|
|
"TestConv.test_torch_conv_depthwise",
|
|
"TestConv.test_torch_conv_general",
|
|
"TestConvTranspose.test_torch_conv_tranpose_1d_output_padding",
|
|
"TestConvTranspose.test_torch_conv_transpose_1D",
|
|
"TestConvTranspose.test_torch_conv_transpose_1D_grad",
|
|
"TestConvTranspose.test_torch_conv_transpose_2D",
|
|
"TestConvTranspose.test_torch_conv_transpose_2D_grad",
|
|
"TestConvTranspose.test_torch_conv_transpose_2d_output_padding",
|
|
"TestConvTranspose.test_torch_conv_transpose_3D",
|
|
"TestConvTranspose.test_torch_conv_transpose_3D_grad",
|
|
"TestConvTranspose.test_torch_conv_transpose_3d_output_padding",
|
|
# FFTs NYI
|
|
"TestFFT.test_fft",
|
|
"TestFFT.test_fft_big_powers_of_two",
|
|
"TestFFT.test_fft_contiguity",
|
|
"TestFFT.test_fft_exhaustive",
|
|
"TestFFT.test_fft_grads",
|
|
"TestFFT.test_fft_into_ifft",
|
|
"TestFFT.test_fft_large_numbers",
|
|
"TestFFT.test_fft_shared_mem",
|
|
"TestFFT.test_fftn",
|
|
# Lapack ops NYI
|
|
"TestLinalg.test_cholesky",
|
|
"TestLinalg.test_cholesky_inv",
|
|
"TestLinalg.test_eig",
|
|
"TestLinalg.test_eigh",
|
|
"TestLinalg.test_inverse",
|
|
"TestVmap.test_vmap_inverse",
|
|
"TestLinalg.test_lu",
|
|
"TestLinalg.test_lu_factor",
|
|
"TestLinalg.test_pseudo_inverse",
|
|
"TestLinalg.test_qr_factorization",
|
|
"TestInit.test_orthogonal",
|
|
"TestLinalg.test_svd_decomposition",
|
|
"TestVmap.test_vmap_svd",
|
|
"TestLinalg.test_tri_inverse",
|
|
# Quantization NYI
|
|
"TestQuantized.test_gather_matmul_grad",
|
|
"TestQuantized.test_gather_qmm",
|
|
"TestQuantized.test_gather_qmm_sorted",
|
|
"TestQuantized.test_gather_qmm_grad",
|
|
"TestQuantized.test_non_multiples",
|
|
"TestQuantized.test_qmm",
|
|
"TestQuantized.test_qmm_jvp",
|
|
"TestQuantized.test_qmm_shapes",
|
|
"TestQuantized.test_qmm_vjp",
|
|
"TestQuantized.test_qmv",
|
|
"TestQuantized.test_qvm",
|
|
"TestQuantized.test_qvm_splitk",
|
|
"TestQuantized.test_small_matrix",
|
|
"TestQuantized.test_throw",
|
|
"TestQuantized.test_vjp_scales_biases",
|
|
}
|