nihui
9b91fe5153
implement flip layer and pnnx torch.flip conversion ( #6233 )
Co-authored-by: 佰阅 <43716063+Baiyuetribe@users.noreply.github.com>
9 months ago
Copilot
4644540ea4
Add Windows XP support merging PRs #6176 and #6177 ( #6204 )
Co-authored-by: Sugar-Baby <87747602+Sugar-Baby@users.noreply.github.com>
Co-authored-by: AtomAlpaca <66774326+AtomAlpaca@users.noreply.github.com>
10 months ago
GIBEREZ
44982d0d23
About the update to the GLSL documentation after the image functions are deprecated ( #6173 )
10 months ago
Christopher
260f493ada
add cross toolchain cmake config of AK3918(AK) and SS928(hisi) ( #6164 )
10 months ago
chri321
4c72f52954
docs: update Chinese glsl-extension documentation ( #6162 )
- synchronize the latest English content to the Chinese documentation
- correct spelling errors in the English version of glsl-extention
- Fix spelling 'enable_validation_layer' in src/gpu.cpp
10 months ago
nihui
bd0b111775
vulkan tight fp16p pack1 ( #6127 )
11 months ago
nihui
abf0de4488
update ruapu to detect zfh zvfh xtheadvector ( #5841 )
* always prefer xtheadvector
* update ci toolchain
11 months ago
nihui
1f6f01ca20
fix apple glslang package, drop glslang-default-resource-limits ( #6022 )
1 year ago
nihui
80da741307
glsl define ncnn_glsl_version macro ( #6003 )
1 year ago
nihui
ef0b0e631c
interp output size expression ( #5994 )
1 year ago
nihui
39c055d7f2
crop axes starts ends expression ( #5976 )
* skip dynamic tensor index
* handle clone oom
1 year ago
nihui
84970eed4d
vulkan validation layer enables NCNN_LOGE in shader source ( #5963 )
* NCNN_LOGE in glsl
* Update glsl-extension.md
1 year ago
Zhuo Zhang
23cdc817ca
refactor qnx cmake toolchain files ( #5961 )
* Support multiple QNX SDP versions: 700, 710, 800, etc
* Support windows and linux hosts when find programs like C/C++ compilers
* Better hint message by recommend users invoking QNX SDP bundled script
* Precise to aarch64 in file naming
* Update how-to-build.md QNX section
1 year ago
nihui
092200a966
faq vulkan in docker ( #5884 )
1 year ago
shacon wang
701f53794d
add read npy for ncnn2table ( #5930 )
* add read npy for ncnn2table
* optimize npy to mat to avoid non-contiguous memory data
* add create the calibration table from npy
* add catch exception
1 year ago
nihui
6396a732ef
reshape shape expression, drop reshape permute, test reshape oom ( #5918 )
1 year ago
nihui
1e3fcb9dda
paramdict value string type, natural array representation ( #5915 )
1 year ago
nihui
2535a6d5b7
update doc about vulkan-sdk ( #5911 )
1 year ago
nihui
19caca3140
port rvv intrinsic 1.0+ ( #5642 )
* zfh zvfh xtheadvector infra
* dispatch for rvv and xtheadvector
* dispatch for non-vector zfh
* port xtheadvector recp rsqrt trunc
* general rvv gemm
* c906 and c910 ci
* old tuple code clean
* update riscv64 ci
* update build doc
* drop old th1520 toolchain
1 year ago
nihui
0734b657d9
spectrogram and inverse spectrogram ( #5779 )
* only supports hann, hamming and all-one window
* inverse spectrogram does not support length parameter
* spectrogram always returns torch.view_as_real(out) as ncnn does not support complex typed mat yet
* inverse spectrogram always accepts torch.view_as_complex(in) as ncnn does not support complex typed mat yet
1 year ago
nihui
66b54cbea2
multiheadattention int8 quantization ( #5733 )
* x86 vulkan fallback
* comment about bf16s
1 year ago
nihui
1c7af00499
gemm int8 quantization ( #5706 )
* quantize gemm
* write gemm quantize scales
* update doc
* less openmp args
* x86 riscv fallback
* skip gemm vulkan int8
* fix noint8 test, fix arm bf16 test
* enable vfpv4 on neon build only
* fix gemm vulkan without C
* fp16 pack8 output
* enable elempack=8 only for asimdhp+
* tiled gemm int8 test
* opt arm64 tiles, fix asimdhp dispatch
1 year ago
Ankush Goel
9b5f6a39b4
fix: typo ( #5709 )
1 year ago
nihui
5df5413c81
embed int8 quantization and add embed test ( #5667 )
1 year ago
nihui
fdf0df3079
RMSNorm ( #5630 )
1 year ago
張小凡
051b04ffb4
Updated use-ncnn-with-pytorch-or-onnx document ( #5557 )
1 year ago
luxincn
02327ba96f
add esp32 build document and ci Refs #5536 ( #5567 )
1 year ago
TianZer
fc6b753d31
Add mingw ci and building document ( #5547 )
1 year ago
nihui
4c3debae2d
multiheadattention scale param ( #5526 )
* update swiftshader
* skip vs2017 swiftshader
1 year ago
村长大人
1e75a2df21
add harmonyos how to build with vulkan ( #5475 )
2 years ago
RoachZhao
d4292e9a65
Update vulkan-notes.md ( #5472 )
`compute_queue_count` is a function.
2 years ago
nihui
08b7d99a75
rnn/lstm/gru dynamic quantization ( #5435 )
2 years ago
Tabbleman
be15dbe421
add riscv-gnu-toolchain build guide;-) ( #5446 )
2 years ago
nihui
db035d602d
update ncnnoptimize layers, lightmode=false keeps original weight ( #5414 )
2 years ago
lll143653
342faf2e79
Update how-to-build.md ( #5389 )
modify the "bulid for macOS"
2 years ago
Galasnow
964ed7a56a
Add implementation of build for protobuf>=22.0 on Windows ( #5359 )
And fix a missing word.
2 years ago
luqiang-guo
8ddc85f4dd
fix doc dual issue ( #5342 )
2 years ago
afredooo
96d073d541
Some typo fixes ( #5339 )
2 years ago
hugo-syn
7d8019d577
chore: add markdown code highlight ( #5302 )
Signed-off-by: hugo-syn <hugo.vincent@synacktiv.com>
2 years ago
hugo-syn
f35eb4b3b8
chore: Fix multiple typos ( #5301 )
Signed-off-by: hugo-syn <hugo.vincent@synacktiv.com>
2 years ago
nihui
5329d32e74
check vulkan fp16 uniform support and implement lfp conversion without fp16u ( #5287 )
2 years ago
nihui
c222208cc9
feat mask for disable threading, make some extractor setter no-op, update doc ( #5270 )
2 years ago
Ikko Eltociear Ashimine
5581d27d4d
docs: update FAQ-ncnn-vulkan.md ( #5268 )
plase -> please
2 years ago
JeremyRand
ed22eb44cc
Document libomp-dev dependency ( #5228 )
Co-authored-by: Jeremy Rand <jeremyrand@danwin1210.de>
2 years ago
lll143653
d4dcb3a2f0
Update FAQ-ncnn-produce-wrong-result.md ( #5220 )
2 years ago
JeremyRand
c1d952da7e
Fix "Rasberry" typo ( #5182 )
Co-authored-by: Jeremy Rand <jeremyrand@danwin1210.de>
2 years ago
JeremyRand
765ac7aef6
Update Vulkan dependency docs ( #5178 )
Vulkan deps are optional. vulkan-utils is replaced with vulkan-tools
since Debian 10.
Co-authored-by: Jeremy Rand <jeremyrand@danwin1210.de>
2 years ago
JeremyRand
d1f6193250
Update POWER Clang version docs ( #5174 )
Clang prior to 13 no longer fails to build ncnn since #4845 .
Clang 18 fixes SSE4.1 translation, which yields a major speedup.
Co-authored-by: Jeremy Rand <jeremyrand@danwin1210.de>
2 years ago
張小凡
2ecaf37a3e
Fix find GPU driver dll path in windows ( #5141 )
2 years ago
nihui
b4f26237cb
in-house vulkan loader ( #5130 )
* vulkan-driver-loader.md
* static vulkan on apple
2 years ago