nihui
/
ncnn
Not watched
1
0
Fork 0
Code
Releases 46 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
2088 Commits
5 Branches
354 MB
1 Download
Tree: 882f114a8d
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '882f114a8d'
${ noResults }
Commit Graph

1286 Commits (882f114a8debc4c98a5b87d2a43c00a67ae5eb03)

Author SHA1 Message Date
  nihuini 9b5cb959b9 auto convert int8 to fp32 on extract 5 years ago
  nihuini 00b0094b2f copy-pasta makes msvc happy 5 years ago
  nihuini 9a12597dff build ncnn quantization tool even opencv not found 5 years ago
  nihui 49f3e1ea09
drawing api and stb_image (#2913) 5 years ago
  nihui ad37c34d25 disable NCNN_ARM82DOT whenever NCNN_ARM82 disabled 5 years ago
  FusionBolt 322812de2d
Fix typo (#2922) 5 years ago
  Cai Shanli 8cc8cd716a
Add get input and output names (#2890) 5 years ago
  nihuini afc02d57f9 runtime detect armv8.2 dotprod 5 years ago
  nihui 5f62fdec87 allow more concurrent gpu submits on device with low queue count 5 years ago
  nihui 5d6f03dbbd asm word for risc-v without v 5 years ago
  nihui a1b06baec8 fix build 5 years ago
  nihui 17936e9f54 fix packing risc-v test, add cpu_riscv_vlenb() 5 years ago
  nihui da40cc6ec8 risc-v v optimization for math functions of all LMUL 5 years ago
  nihuini d0d8120335 force lrn vulkan accumulator always use fp32, fix #2882 5 years ago
  nihuini 7a1e015d72 fix bias error in conv1x1s1 sgemm pack8to4 fp16sa, fix #2880 5 years ago
  nihui 11958424c2 runtime riscv v and zfh dispatch, riscv v optimization for cast 5 years ago
  nihui e87af6d020 riscv v optimization for log_ps sincos_ps tanh_ps, mish, swish, tanh, unaryop, enable fp16sa 5 years ago
  nihui 37aeba2369 riscv v optimization for exp_ps, sigmoid 5 years ago
  nihui 1c26291757 more verbose hint for find_blob_index_by_name failure 5 years ago
  nihui e8f92e4a04 riscv v optimization for packing, add rvv071 compatibility header 5 years ago
  nihuini 34bd5ef161 update eq quant info 5 years ago
  nihuini 72ef77a469 fix build with NCNN_STRING off and NCNN_VULKAN on 5 years ago
  zhiliu6 fb9d529487
fix compile error when NCNN_STRING is disabled (#2874) 5 years ago
  songqun ad1012bcda
add comment for alpha beta in hardswish and hardsigmoid compared to tf/pytorch implementation (#2859) 5 years ago
  Zhang Xianyi a1ece94f51
Use RVV spec 0.7.1 for C906. (#2868) 5 years ago
  nihui 45bf3cd779 add runtime riscv v detection function, the initial c906 riscv linux toolchain 5 years ago
  nihui bdb3aa5657 fix build 5 years ago
  nihui a61f03ec76 arm neon optimization for pixelshuffle scale 2 5 years ago
  nihui 81be8e235c
workaround macos intel dummy image readonly issue, fix #2548 (#2864) 5 years ago
  nihuini d6b2ea5aac arm neon optimization for convolution 3x3 on small channels 5 years ago
  nihuini 6a397716ca arm neon optimization for instancenorm 5 years ago
  nihuini 687cc857b1 some x86 sse2 optimization for convolution int8 5 years ago
  zhiliu6 ec0f904c16
improve x86 1x1 pack8 convolution performance (#2852) 5 years ago
  nihuini 68468dccbd arm neon assembly optimization for padding int8 pack8, convolution int8 out elempack 4 5 years ago
  nihuini 31d436c627 more verbose load failure, ncnn2int8 write int8 data properly 5 years ago
  nihuini 05d457c78f innerproduct int8 support all fused activation types 5 years ago
  nihuini 1bc0126302 fix crash when input cpu blob and extract the same from gpu, update vgg16 int8 model 5 years ago
  nihui 7e1aaa5828
cmake option NCNN_INT8 (#2839) 5 years ago
  nihui 66455c1b95
implement 2823 binary broadcasting type (#2827) 5 years ago
  nihuini 85efe132ff unroll inch 4 for convolution sgemm int8 5 years ago
  nihui c6cd5e8628 fix armv7 no-neon build 5 years ago
  nihuini e38a5fcbe6 fix build 5 years ago
  nihuini 01f5dcb700 arm neon optimization for convolution sgemm pack1 pack8to1 int8 5 years ago
  zhiliu6 c4700c52ca
optimize x86 1x1 pack8 convolution (#2820) 5 years ago
  nihui 0d1d5b66c5 fix arm64 asm build 5 years ago
  nihuini e9ab1acf27 arm neon optimization for convolution sgemm pack1to8 int8 5 years ago
  nihuini e975de1f36 better condition for arm82 conv3x3s1 winograd 5 years ago
  nihuini 41a4bea954 unroll size 8 for conv3x3s1 pack8to1 int8 arm64 5 years ago
  nihuini 3631c1933d non-inlined addref and release slows down overall speed, move them to header 5 years ago
  nihui e9cc637573
arm neon optimization for int8 packing kernels (#2809) 5 years ago