jzhang533/Paddle

develop

Paddle

Limited to 5K files. Learn more.

Grade	Name	Complexity	Churn	Issues
B	paddle\phi\kernels\impl\matmul_kernel_impl.h	46	23	4
B	paddle\phi\kernels\gpu\flash_attn_v3_grad_kernel.cu		24	6
B	paddle\phi\kernels\gpu\depthwise_conv.h	271	20	11
B	paddle\phi\kernels\fusion\cutlass\memory_efficient_attention\iterators\predicated_tile_iterator_residual_last.h	163	2	6
B	paddle\phi\kernels\fusion\cutlass\cutlass_extensions\gemm\threadblock\dp_mma_multistage_percol.h	35	2	8
B	paddle\phi\kernels\fusion\cutlass\cutlass_extensions\gemm\threadblock\dp_mma_multistage_finegrained.h	40	2	8
B	paddle\phi\kernels\funcs\pooling.cc	289	7	21
B	paddle\phi\kernels\funcs\blas\blas_impl.hip.h	228	5	9
B	paddle\phi\kernels\funcs\blas\blaslt_impl.cu.h	105	8	6
B	paddle\phi\kernels\cpu\interpolate_grad_kernel.cc	202	15	12
B	python\paddle\base\dygraph\tensor_patch_methods.py	334	33	8
B	paddle\phi\kernels\funcs\blas\blaslt_impl.hip.h	102	7	6
B	paddle\phi\infermeta\ternary.cc	161	38	7
B	paddle\fluid\pybind\pybind.cc	509	78	4
B	paddle\fluid\platform\tensorrt\engine.cc	141	5	9
B	paddle\fluid\prim\api\composite_backward\composite_backward_api.h	329	15	12
B	paddle\fluid\pir\transforms\pd_op_to_kernel_pass.cc	379	19	10
B	paddle\fluid\pir\transforms\onednn\conv_concat_activation_onednn_fuse_pass.cc	77	5	7
B	paddle\fluid\primitive\codegen\decomp_vjp_gen.py	76	10	6
B	paddle\fluid\pir\transforms\onednn\cpu_bfloat16_pass.cc	287	4	17