b9af93c74b
Add percentile PTQ method in ncnn2table ( #6854 )
2026-10-09 10:33:34 +08:00
f74639deeb
gemm block quantization infrastructure for llm decoder style ( #6814 )
...
* another mha oob read
2026-07-13 10:13:39 +08:00
3724d10fc8
tools:Move Embed/MHA/RNN/LSTM/GRU weight scale generation to ncnn2table ( #6688 )
...
Co-authored-by: nihui <nihuini@tencent.com >
2026-05-18 19:38:07 +08:00
4699b26933
Using snprintf instead of sprintf for a tools ( #6554 )
...
Signed-off-by: Evgeny Proydakov <e.proydakov@gmail.com >
2026-02-26 10:29:28 +08:00
641a3e5be8
ncnn sdpa layer ( #6397 )
...
* fuse gqa
2025-11-11 11:14:04 +08:00
171b9d1bba
use spdx license header, copyright Tencent ( #6152 )
2025-07-08 11:01:29 +08:00
9afeb17cd4
ncnn2int8 use fp16 weight type by default ( #5907 )
2025-02-17 17:26:25 +08:00
66b54cbea2
multiheadattention int8 quantization ( #5733 )
...
* x86 vulkan fallback
* comment about bf16s
2024-10-15 16:38:00 +08:00
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
2024-10-12 19:02:55 +08:00
5df5413c81
embed int8 quantization and add embed test ( #5667 )
2024-09-02 18:48:01 +08:00
08b7d99a75
rnn/lstm/gru dynamic quantization ( #5435 )
2024-05-08 19:25:51 +08:00
e24a2e6c40
[fix] fix typo ( #3377 )
2021-11-23 21:44:36 +08:00
1ad37fae1e
generate random weight when null bin passed
2021-05-24 15:54:12 +08:00
56d74580a0
fix uninitialized input and weight
2021-04-15 17:45:12 +08:00
2b1f8bea25
fix float locale issue in ncnn2mem, fix long line parsing in ncnn2int8
2021-04-14 18:59:28 +08:00
5fe75f19ef
architecture changes for int8 packing ( #2771 )
...
* quantize and dequantize tests
* unify activation and usability function
* drop NCNN_REQUANT cmake option, test dequantize requantize pack8, fix webassembly build
* benchmark use requantize int8 model
2021-03-27 18:49:13 +08:00
f3c48cd063
Fix bug of adaptive pool in int8 quantization model ( #2637 )
2021-01-28 12:48:31 +08:00
30a4482506
fix typo in comment in ncnn2int8.cpp ( #2627 )
2021-01-26 21:48:01 +08:00
54c0a13b9f
build shared library ( #2525 )
...
* build shared lib and enable lto
* reserved for layer and option
* allocator pimpl
* datareader pimpl
* paramdict pimpl, disable copy assign for allocator and datareader
* modelbin pimpl
* net extractor pimpl
* gpu pimple
* disable copy assign vulkandevice, code format
* command pimpl, dummy image readonly
* pipeline pipelinecache pimpl, export platform class
* code format, export simple family
* update ci
* disable lto on android armv7, merge webassembly ci
* link libgcc, fix macos dylib version
* pipeline pimpl, gpu info pimpl
* destroy gpu info after vulkan device
* ignore msvc stl class warning
* fix ncnn_paramdict_get_float return type
* fix vktransfer upload fp16 without flatten, add command test
2021-01-01 22:55:10 +08:00
3ef995ed1e
format code style and setup restyled.io ( #1840 )
2020-06-13 22:17:51 +08:00
ce2251db05
Improve ROIAlign (accelerate ROIAlign, support sampling ratio and aligned ROIAlign) ( #1820 )
...
* update roialign
* update tool for roialign
* fix ceil for roialign
* fix ios build
* fix
* fix
* make it compatibile with the old version
* fix ios build
* trigger CI
* fix test
* order
* build
* remove code
* merge roialign
* accelerate ROIAlign
* update note
* rename func
* roialign version
* trigger CI
* fix roialign
* use ref for pre-calc in roialign
* retrigger CI
* pre-alloc
2020-06-08 21:40:24 +08:00
8a84077429
ncnnoptimize lstm
2020-03-15 11:56:23 +08:00
dba992c15c
ncnnoptimize add pixelshuffle, fix #1608
2020-03-12 20:54:50 +08:00
05ab6813f5
c++03 compiler happy
2020-01-23 14:36:47 +08:00
6916f45237
fix quantize tools cross platform issues ( #1505 )
...
* 1. for cross platform, use cv::CommandLineParser instead of getopt.h
2. use cv::glob to search folder instead of dirent.h
3. fix some other warnings of function from cstdlib
4. add some const and some auto
* 1. fix some other warnings of function from cstdlib
2. add some const and some auto
3. i forgot...
* switch quantize tools default on
* check OpenCV first, if not found, disable quantize tools
* add _CRT_SECURE_NO_DEPRECATE for msvc and remove *_s functions
* add _CRT_SECURE_NO_DEPRECATE for msvc and remove *_s functions
* keep the one line style block
* folding bracket
* dynamic_cast -> static_cast
2020-01-22 16:31:44 +08:00
5c08ce7db9
fix Reduction Op attr bug ( #1363 )
...
* fix onnx2ncnn MemoryData dims problem #1330
* fix ncnn2int8 bug (see #1336 )
* fix bug that MemoryData as multi operations input
* fix ncnn2table find_blob_index_by_name data failed bug, see #1338
* fix ncnn2table find_blob_index_by_name data failed bug, see #1338 , fix change requested
* fix Reduction operation attribution bug
* fix #1363 request changes
2019-11-08 14:09:59 +08:00
31c6cefc44
fix ncnn2int8 bug, see issue #1336 ( #1341 )
...
* fix onnx2ncnn MemoryData dims problem #1330
* fix ncnn2int8 bug (see #1336 )
2019-10-29 15:25:21 +08:00
6176ada9f0
update ncnn2int8.cpp ( #1315 )
...
Fix compile issue with ncnn2int8
2019-10-22 18:47:30 +08:00
f5004f954b
Update ncnn2int8.cpp ( #1200 )
...
asymmetric padding parameter for convolution and deconvolution family
2019-08-22 11:37:32 +08:00
f79fa87dca
Add the readme file of quantization tools ( #1198 )
2019-08-20 20:46:04 +08:00
bcfe9f453f
initial the ncnn post training quantization tools ( #1067 )
...
* initial the ncnn post training quantization tools
* clear some comments of tools
* fix the Travis ci compiler error
2019-06-26 21:33:02 +08:00
Layla