Tencent/ncnn

master

ncnn

Grade	Name	Complexity	Churn	Issues
F	src\layer\x86\convolution_3x3_winograd_bf16s.h	540	2	45
F	src\layer\arm\convolution_3x3_winograd_fp16s.h	529	1	19
F	src\layer\arm\convolution_3x3_winograd.h	484	1	69
B	src\layer\arm\gemm_arm_asimdhp.cpp	479	2	14
F	src\layer\x86\convolution_3x3_winograd_int8.h	476	2	25
F	src\layer\x86\convolution_im2col_gemm.h	450	3	26
A	src\layer\x86\deconvolution_packed_bf16s.h	450	2	6
A	src\layer\x86\deconvolution_packed.h	448	2	6
A	tools\modelwriter.h	445	2	4
B	src\layer\mips\convolution_mips.cpp	437	1	9
B	src\layer\loongarch\convolution_loongarch.cpp	437	1	9
A	src\layer\arm\convolution_arm_asimdhp.cpp	426	2	4
A	tools\pnnx\src\model_stat.cpp	426	1	5
C	src\layer\riscv\gemm_riscv.cpp	423	2	19
F	src\layer\x86\convolution_im2col_gemm_bf16s.h	405	4	33
A	src\mat.cpp	392	1	1
B	src\layer\arm\convolutiondepthwise_arm.cpp	380	1	12
A	src\layer\x86\crop_x86.cpp	376	2	3
B	src\layer\vulkan\convolution_vulkan.cpp	363	5	6
A	src\cpu.cpp	355	1	3