2672 Commits (76849cede4d68fde66646ecdc1faa070daeb6dac)
 

Author SHA1 Message Date
  nihui 76849cede4
armv8.4 i8mm optimization for convolution gemm int8 (#4034) 3 years ago
  Jie Li 627be612c6
fix(examples): optimize nms_sorted_bboxes (#4030) 3 years ago
  nihui dd86cebab8
armv8.6 ci and coverage (#4025) 3 years ago
  nihui c5bb0e52ed
add ingenic-x2000 toolchain file 3 years ago
  nihui 9b39691cc8
pnnx handle unrecognized file format (#4028) 3 years ago
  nihui f1ea792b26
fix too many microtask error in old libomp runtime (#4002) 3 years ago
  Guo Haria c1e2ab7205
add yolov7_pnnx example (#4027) 3 years ago
  xuehao.ma 962a49069a
add the param file of fastestdet in benchmark (#4026) 3 years ago
  teng 3901b837e2
add example yolov7 (#4019) 3 years ago
  nihui 9b8272e86d
arm edsp and arm neon optimization for convolution int8 winograd (#4017) 3 years ago
  nihui a12cd7c212
mips msa and loongson mmi optimization for convolution int8 winograd f43 (#4014) 3 years ago
  陸 言 cae8d0f1d7
Add Loongson 2F toolchain support (refer to AOSC) (#3992) 3 years ago
  nihui 5725c028c0
arm dsp infrastructure and optimization for convolution gemm int8 (#4011) 3 years ago
  nihui ef216f732e
armv5 optimization for convolution gemm int8 (#4010) 3 years ago
  nihui 0a12f81a2d
fix data race in arm rnn/gru/lstm (#4008) 3 years ago
  nihui 322667a2ab
pnnx fix fused tensor_split operator insert order (#4006) 3 years ago
  Jie Li 434e55c0f8
fix(examples): load_param always failed (#4001) 3 years ago
  nihui 1892c25360
pnnx fuse megvii style shufflechannel slice (#3999) 3 years ago
  nihui 94786308bd
pnnx fuse binaryop eltwise as weighted sum (#4000) 3 years ago
  dependabot[bot] 8d28fc91b5
Bump pypa/cibuildwheel from 2.7.0 to 2.8.0 (#3995) 3 years ago
  nihui a5fb92db51
optimize innerproduct fp16s transform kernel (#3994) 3 years ago
  nihui d4a704de0e
pnnx eliminate noop upsample (#3991) 3 years ago
  nihui b4bae2c9e4
pnnx convert torch.tensor_split, fuse full dim size slice to tensor_split (#3988) 3 years ago
  sodo 3b3605eec4
add pkgconfig (#3984) 3 years ago
  Zhuo Zhang c04a1bccce
fix(examples): exit program if load_param, load_model failed for clear message (#3983) 3 years ago
  tpoisonooo 207ca0e0bb
Improve protobuf FAQ doc (#3973) 3 years ago
  nihui 044467b2c6
pnnx support torch 1.12 (#3981) 3 years ago
  nihui 0de869e8e3
ci release copy and zip preserve links (#3980) 3 years ago
  nihui d2fc473d8f
update qcom810 benchmark data 3 years ago
  nihui b9b5f9b119
ci release python fix, install strip (#3977) 3 years ago
  nihui eeb969ad2c
update qcom855plus benchmark data. set adreno gpu frequency. 3 years ago
  nihui f8c76e730a
fix ci release optimization with cmake >= 3.21 and ndk23 (#3976) 3 years ago
  nihui eeac683c70
fix release python (#3974) 3 years ago
  nihui 14588023b5
release ubuntu 22.04 package, fix ndk debug flag for r23+ (#3972) 3 years ago
  nihui 531506d602
improve pattern value match, always treat inplace operator as non-inplace version (#3970) 3 years ago
  nihui 8dbedf8a19
use cmake gnuinstalldirs for install destination (#3968) 3 years ago
  nihui 43912c7f42
more strict compiler avx512fp16 support detection (#3966) 3 years ago
  moozae 94ec06a8a5
delete unused variables (#3965) 3 years ago
  nihui f597619e94
pnnx export weight inside moduleop (#3902) 3 years ago
  nihui b85bfb6085
armv8.2 asimdfhm and armv8.4 bf16 i8mm and armv8.6 sve sve2 compiler flags and runtime detection functions (#3964) 3 years ago
  nihui be8e195630
add avx512 spr ci (#3960) 3 years ago
  nihui 5ae827c745
convert inplace relu6, match more hardswish pattern (#3952) 3 years ago
  nihui bc8e939c85
insert reshape for nn.Linear with 4d/5d inputs (#3959) 3 years ago
  Yoh a4ccad3325
Fix gcc4 simd conflict (#3957) 3 years ago
  dankernel 01655613d1
Create build-for-VisualStudio.en.md (#3956) 3 years ago
  Jianbo-Ning 4fad760a64
add faq.en.md (#3901) 3 years ago
  nihui 28e0c9132a
fix cmake find thread on mips 3 years ago
  nihui 27dc780005
mips msa optimization for innerproduct fp16s (#3953) 3 years ago
  nihui 706831f8a9
arm vfpv4 optimization for innerproduct (#3950) 3 years ago
  nihui 440bfdd2cc
x86 f16c optimization for innerproduct (#3944) 3 years ago