nihui
/
ncnn
Not watched
1
0
Fork 0
Code
Releases 46 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
931 Commits
5 Branches
354 MB
1 Download
Tree: 70f3198703
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '70f3198703'
${ noResults }
Commit Graph

931 Commits (70f3198703b652bfaee3e12e85a6643d2ca2f9cd)
 

Author SHA1 Message Date
  nihuini 70f3198703 conversion between ncnn Mat and android Bitmap makes life easier :D 6 years ago
  nihuini 8f72256d1f mat pixel rgb2rgba bgr2rgba gray2rgba 6 years ago
  nihuini 91ef4eea4f fix unaryop arm, fix #1241 6 years ago
  nihuini db198be962 bump android platform version to 19, that makes ndk-r19c happy 6 years ago
  nihuini 56acc25df8 set QuantNet option, fix #1223 6 years ago
  nihuini 4ff7cc8930 fix copy header 6 years ago
  nihuini 3e3189736c fix msvc build, fix #1237 6 years ago
  Xu Yang 31cf7f3c5b fix ConvolutionDepthWise int8_requantize (#1233) 6 years ago
  nihuini c4bebc6371 x86 conv3x3s1 winograd43 produce wrong result, revert to the good-old winograd23 version 6 years ago
  nihuini 7e8cbe6170 add FAQ about pytorch and mxnet-gluon preprocess 6 years ago
  CnybTseng d11c4c1d42 修改最新版ncnn/src/layer/vulkan/shader目录下的几个文件，以适配最新版的glslang，本次修改已在大疆Manifold2-G平台上验证通过 (#1231) 6 years ago
  nihui 2556ece1ca advertise the new squeezencnn android example project 6 years ago
  nihuini 8768941223 rename to how-to-build 6 years ago
  nihuini 19efe499c2 update doc links 6 years ago
  tpoisonooo e16b338b13 build docs: (#1204) 6 years ago
  nihui 46e7ac76ab apply sgemm-like dot in winograd pack4 neon 6 years ago
  nihui d6860d93f2 fix batchnorm pack4 neon multithreaded 6 years ago
  nihui b90a6cf69c split always support packing :) 6 years ago
  nihui 6b5588ebbe fix batchnorm pack4 out of range write 6 years ago
  nihui 4dc98ffaab conv1x1s1 and conv3x3s1 winograd pack4 neon optimization, first try 6 years ago
  nihui 02c811e829 fix padding pack4 elempack 6 years ago
  nihuini 5d1a94826d crop border param, try to make life eaiser 6 years ago
  nihui af646b6577 workaround for the adjacency same pipeline issue 6 years ago
  nihui cb82ccbe20 convolution vulkan winograd63 shader compilation is too slow and runs slower in many cases, ok let's drop it 6 years ago
  nihuini d04b996cc4 add human pose estimation example 6 years ago
  MambaWong 997e96a490 add swapRB flag (#1211) 6 years ago
  nihuini 6db731408c crop vulkan pack1to4 and pack4to1 6 years ago
  nihui 35990bb5d5 instancenorm vulkan 6 years ago
  nihuini 4b8fb744bc fix potential to_pixels_resize out of range write, fix #1201 6 years ago
  nihui c2bc0d1b88 padding vulkan reflect mode 6 years ago
  MambaWong f5004f954b Update ncnn2int8.cpp (#1200) 6 years ago
  nihui 3eb30f4cdb write deconvolution output pad and output shape, write instancenorm weight, fix #1053 fix #1122 6 years ago
  nihuini 296e0022df deconvolution output adj and output shape 6 years ago
  BUG1989 f79fa87dca Add the readme file of quantization tools (#1198) 6 years ago
  nihuini e4b44d293e more autopad SAME_LOWER 6 years ago
  nihuini 0e26e3094e autopad SAME_LOWER 6 years ago
  nihuini 9c1eeb5d5c raise NCNN_MAX_PARAM_COUNT from 20 to 32 6 years ago
  nihuini 18833c9286 convert asymmetric pad parameter 6 years ago
  nihuini 9a6ee37eef asymmetric padding parameter for convolution and deconvolution family 6 years ago
  nihui 0e83b0f3a0
mention mobilenetv3 and unet 6 years ago
  nihui 49206ea03c eliminate useless reshape for mbv3 6 years ago
  nihuini 9d4255a4a4 add mobilenet_v3 benchmark 6 years ago
  nihuini f5ca997a66 fuse x * hardsigmoid to hardswish 6 years ago
  nihuini 46bfb4bca9 write hardswish parameter 6 years ago
  nihui 29324771b1
Implemented hard swish layer (#1195) 6 years ago
  nihuini da6a42ed96 fix onnx2ncnn build 6 years ago
  nihuini 012fdd46de convert onnx hardsigmoid hardswish, op chain fusion code reconstructed 6 years ago
  nihuini eced73c186 convert mxnet onnx pooling count_include_pad 6 years ago
  nihuini 3d5b7f20ff avgpool count_include_pad 6 years ago
  Hao Zeng 1f6919fd40 Implemented hard swish layer 6 years ago