nihui/ncnn - ncnn - 开源协同云脑生态支撑系统

History

nihui f22e5d4a6d fix build		6 years ago
..
absval_arm.cpp	neon enable support_packing	6 years ago

absval_arm.h	absval batchnorm clip relu sigmoid support packing	7 years ago

batchnorm_arm.cpp	fix some boring compile warnings (#1510)	6 years ago

batchnorm_arm.h	absval batchnorm clip relu sigmoid support packing	7 years ago

bias_arm.cpp	implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469	7 years ago

bias_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

binaryop_arm.cpp	clip sigmoid tanh bf16s	6 years ago

binaryop_arm.h	binaryop bf16s	6 years ago

cast_arm.cpp	the use_bf16_storage option	6 years ago

cast_arm.h	cast pack4 arm neon	6 years ago

clip_arm.cpp	clip sigmoid tanh bf16s	6 years ago

clip_arm.h	clip sigmoid tanh bf16s	6 years ago

concat_arm.cpp	code clean, concat bf16s	6 years ago

concat_arm.h	code clean, concat bf16s	6 years ago

convolution_1x1.h	move platform includes out of namespace	7 years ago

convolution_1x1_bf16s.h	conv1x1s1 bf16s neon kernel	6 years ago

convolution_1x1_int8.h	replace nullptr to 0, fix build on old toolchain, fix #1541	6 years ago

convolution_1x1_pack4.h	fix potential out of write on unroll 12 remainder	6 years ago

convolution_1x1_pack4_bf16s.h	code clean, concat bf16s	6 years ago

convolution_1x1_pack4to1.h	arm neon assembly optimization for conv1x1s1 pack4to1	6 years ago

convolution_1x1_pack4to1_bf16s.h	code clean, concat bf16s	6 years ago

convolution_2x2.h	move platform includes out of namespace	7 years ago

convolution_3x3.h	Mat misc function accept option parameter, deconvolution pack4 arm neon	6 years ago

convolution_3x3_int8.h	fix some boring compile warnings (#1510)	6 years ago

convolution_3x3_pack1to4.h	unroll outch 2 for conv3x3s1 pack1to4	6 years ago

convolution_3x3_pack1to4_bf16s.h	conv1x1 pack4to1 bf16s neon kernel	6 years ago

convolution_3x3_pack4.h	fix group convolution 3x3s1 with size hit winograd bound, fix #1581	6 years ago

convolution_3x3_pack4_bf16s.h	conv3x3s1 pack4to1 bf16s neon kernel	6 years ago

convolution_3x3_pack4to1.h	fix group convolution 3x3s1 with size hit winograd bound, fix #1581	6 years ago

convolution_3x3_pack4to1_bf16s.h	conv3x3s1 pack4to1 bf16s neon kernel	6 years ago

convolution_4x4.h	move platform includes out of namespace	7 years ago

convolution_5x5.h	move platform includes out of namespace	7 years ago

convolution_5x5_pack4.h	arm neon assembly optimization for conv5x5s1 pack4 and conv5x5s2 pack4	6 years ago

convolution_5x5_pack4_bf16s.h	conv5x5s2 bf16s neon kernel	6 years ago

convolution_7x7.h	move platform includes out of namespace	7 years ago

convolution_7x7_pack1to4.h	fix build on armv7	6 years ago

convolution_7x7_pack1to4_bf16s.h	conv7x7s2 pack1to4 bf16s neon kernel	6 years ago

convolution_arm.cpp	conv1x1s1 bf16s neon kernel	6 years ago

convolution_arm.h	code clean, concat bf16s	6 years ago

convolution_sgemm.h	fix conv_im2col_sgemm_neon, second try	6 years ago

convolution_sgemm_int8.h	replace nullptr to 0, fix build on old toolchain, fix #1541	6 years ago

convolutiondepthwise_3x3.h	move platform includes out of namespace	7 years ago

convolutiondepthwise_3x3_int8.h	arm-hisim100-linux make error convolutiondepthwise_3x3_int8.h for asm (#1310)	6 years ago

convolutiondepthwise_3x3_pack4.h	convdw3x3 pack4 arm neon assembly optimization	6 years ago

convolutiondepthwise_3x3_pack4_bf16s.h	the use_bf16_storage option	6 years ago

convolutiondepthwise_5x5.h	move platform includes out of namespace	7 years ago

convolutiondepthwise_5x5_pack4.h	arm neon optimization for convdw5x5 pack4	6 years ago

convolutiondepthwise_5x5_pack4_bf16s.h	fix dwconv5x5s1 pack4 bf16s on aarch64	6 years ago

convolutiondepthwise_arm.cpp	dwconv5x5s1 pack4 bf16s neon kernel	6 years ago

convolutiondepthwise_arm.h	the use_bf16_storage option	6 years ago

crop_arm.cpp	crop arm bf16	6 years ago

crop_arm.h	neon enable support_packing	6 years ago

deconvolution_3x3.h	move platform includes out of namespace	7 years ago

deconvolution_4x4.h	move platform includes out of namespace	7 years ago

deconvolution_arm.cpp	fix build without neon	6 years ago

deconvolution_arm.h	the initial auto test (#1464)	6 years ago

deconvolutiondepthwise_arm.cpp	fix build without neon	6 years ago

deconvolutiondepthwise_arm.h	the initial auto test (#1464)	6 years ago

dequantize_arm.cpp	Fix chgemm (#1480)	6 years ago

dequantize_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

dropout_arm.cpp	neon enable support_packing	6 years ago

dropout_arm.h	neon enable support_packing	6 years ago

eltwise_arm.cpp	code clean, concat bf16s	6 years ago

eltwise_arm.h	the use_bf16_storage option	6 years ago

flatten_arm.cpp	the use_bf16_storage option	6 years ago

flatten_arm.h	convolution and pooling make padding helper, flatten innerproduct pooling bf16s neon	6 years ago

gemm_symm_int8.h	fix arch64 build (#1633)	6 years ago

hardsigmoid_arm.cpp	hardsigmoid bf16s	6 years ago

hardsigmoid_arm.h	hardsigmoid bf16s	6 years ago

hardswish_arm.cpp	hardswish bf16s	6 years ago

hardswish_arm.h	hardswish bf16s	6 years ago

innerproduct_arm.cpp	code clean, concat bf16s	6 years ago

innerproduct_arm.h	convolution and pooling make padding helper, flatten innerproduct pooling bf16s neon	6 years ago

interp_arm.cpp	interp share coeffs and xofs	7 years ago

interp_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

lrn_arm.cpp	Mat misc function accept option parameter, deconvolution pack4 arm neon	6 years ago

lrn_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

neon_activation.h	fix build on armv7 without neon	6 years ago

neon_mathfun.h	tanh arm neon optimize	6 years ago

neon_mathfun_tanh.h	fix reinterpret u32 to f32, second try	6 years ago

packing_arm.cpp	conv7x7s2 pack1to4 bf16s neon kernel	6 years ago

packing_arm.h	the use_bf16_storage option	6 years ago

padding_arm.cpp	code clean, concat bf16s	6 years ago

padding_arm.h	code clean, concat bf16s	6 years ago

padding_pack4.h	padding pack4 bf16s neon kernel	6 years ago

padding_pack4_bf16s.h	padding pack4 bf16s neon kernel	6 years ago

pooling_2x2.h	move platform includes out of namespace	7 years ago

pooling_2x2_pack4.h	arm neon assembly optimization for pooling2x2s2 max pack4	6 years ago

pooling_3x3.h	move platform includes out of namespace	7 years ago

pooling_3x3_pack4.h	arm neon assembly optimization for pooling3x3s2 max pack4	6 years ago

pooling_arm.cpp	test bf16s (#1644)	6 years ago

pooling_arm.h	convolution and pooling make padding helper, flatten innerproduct pooling bf16s neon	6 years ago

prelu_arm.cpp	neon enable support_packing	6 years ago

prelu_arm.h	neon enable support_packing	6 years ago

quantize_arm.cpp	constraint input value to [-127, +127] (#1258)	6 years ago

quantize_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

relu_arm.cpp	code clean, concat bf16s	6 years ago

relu_arm.h	relu bf16s neon	6 years ago

requantize_arm.cpp	Fix int8 requant (#1499)	6 years ago

requantize_arm.h	move vulkan code to subdir, new layer interface create_pipeline and destroy_pipeline for post-loading works	7 years ago

reshape_arm.cpp	remove the ncnn namespace for in source Option	6 years ago

reshape_arm.h	reshape pack4	6 years ago

scale_arm.cpp	neon enable support_packing	6 years ago

scale_arm.h	neon enable support_packing	6 years ago

selu_arm.cpp	implemented selu activation (#1051)	7 years ago

selu_arm.h	implemented selu activation (#1051)	7 years ago

shufflechannel_arm.cpp	fix build	6 years ago

shufflechannel_arm.h	shufflechannel bf16s	6 years ago

sigmoid_arm.cpp	clip sigmoid tanh bf16s	6 years ago

sigmoid_arm.h	clip sigmoid tanh bf16s	6 years ago

slice_arm.cpp	slice bf16s	6 years ago

slice_arm.h	slice bf16s	6 years ago

softmax_arm.cpp	the use_bf16_storage option	6 years ago

softmax_arm.h	neon enable support_packing	6 years ago

tanh_arm.cpp	clip sigmoid tanh bf16s	6 years ago

tanh_arm.h	clip sigmoid tanh bf16s	6 years ago

unaryop_arm.cpp	shader shape specialization constant and basic local group size partition (#1523)	6 years ago

unaryop_arm.h	neon enable support_packing	6 years ago

No Description

C++ C Python Text Protocol Buffer other

nihuini@tencent.com shuizhuyuanluo@126.com 248857878@qq.com 314377460@qq.com khj.application@aliyun.com myhouseng@gmail.com gemfield@civilnet.cn youhanmir@gmail.com ShuangLiu1992@users.noreply.github.com ericliu612@gmail.com leo@nullptr.com.cn sungmann.cho@navercorp.com leeward.xie@gmail.com abdel.younes@nviso.ch mingzilaochongtu@gmail.com 379489343@qq.com lzx1413@live.cn xieydd@gmail.com caishanli25@gmail.com jxhzq1996@126.com bindog@outlook.com liushouda@qq.com jpwc@qq.com zhiliu6@gmail.com 791136190@qq.com

Contributors (25+) All

Contributors (25+)
All