nihui
|
9cdc02bb7a
|
unified vulkan khr/nv cooperative matrix shader (#6116)
|
1 year ago |
nihui
|
b9f98f0d3a
|
always allocate aligned size for 1d/2d mat and vkmat (#6104)
* fix sub mat cstep
* fix embed
* rnn/lstm/gru int8 test without rounding diversity
|
1 year ago |
nihui
|
8a2eab1114
|
set localsize as multiple of subgroup size (#2483)
* fix innerproduct gemm vulkan
|
1 year ago |
nihui
|
87e8b5f4c1
|
use combine_x for sse/avx vector combination (#6113)
|
1 year ago |
nihui
|
23b64c9cf9
|
fix vulkan validation error, do not enforce local_size_x be multiple of subgroup size (#6111)
|
1 year ago |
nihui
|
4c4ecdf118
|
dequantize pack8 for all datatypes, fix convdw int8 dequant pack8 (#6109)
|
1 year ago |
hanzh
|
78b2e68728
|
arm unified elempack optimization for groupnorm (#4080)
Co-authored-by: mmyyy22 <mmyyy22@users.noreply.github.com>
Co-authored-by: nihui <nihuini@tencent.com>
|
1 year ago |
nihui
|
6f175fb622
|
move ci gpu swiftshader to ubuntu25 (#6107)
|
1 year ago |
nihui
|
1d84e987ce
|
x86 unified elempack optimization for groupnorm, simplify c groupnorm (#6106)
|
1 year ago |
nihui
|
1365eed05e
|
simplify c layernorm (#6105)
|
1 year ago |
zhuzeitou
|
7b63f5d682
|
Switch to putenv to fix compilation with llvm-mingw (#6101)
Replaces `setenv` with `putenv` because `setenv` is not available on
some toolchains (e.g., llvm-mingw, clang-cl), leading to compilation
errors.
|
1 year ago |
nihui
|
ebc041cc56
|
force subgroup 32 for cooperative matrix shader atm (#6100)
|
1 year ago |
Lfalive
|
2ef954b672
|
unaryop avx512 mask optimization (#6098)
|
1 year ago |
nihui
|
6af6e8a96b
|
ppocr-v5 example (#6092)
* warn puttext
* split dict
|
1 year ago |
Lfalive
|
7fd167fd25
|
tanh avx512 mask optimization (#6096)
|
1 year ago |
nihui
|
5cd76533cc
|
pnnx generate test_inference with valid input shapes anyway (#6095)
|
1 year ago |
Yexuan Wu
|
73d8500988
|
sigmoid avx512 mask optimization (#6093)
|
1 year ago |
tpoisonooo
|
25824dc554
|
add a new FAQ link for ncnn deepwiki (#6083)
|
1 year ago |
nihui
|
ff11080db6
|
pnnx convert onnx MaxPool auto_pad same, torch reshape_as (#6089)
|
1 year ago |
tpoisonooo
|
dddfc50282
|
raise max gpu count to 32 (#6084)
|
1 year ago |
nihui
|
c5c4cdfd88
|
pnnx ppocrv5 onnx update, eliminate squeeze-unsqueeze pair, convert onnx shape+slice, fix param list match (#6081)
* fix memory leak
* fuse conv+bias early
|
1 year ago |
Yingxi
|
46b3de348c
|
Add MUSE Pi Pro Spacemit M1 benchmark. (#6074)
|
1 year ago |
nihui
|
c001f5ee6c
|
pnnx set windows codepage utf8 (#6080)
|
1 year ago |
nihui
|
a8e4db713b
|
initialize layer featmask zero (#6078)
|
1 year ago |
nihui
|
abb7de9a66
|
pnnx transformers update part2 (#6072)
* pnnx update batch 1 mask attention, add more clip test
* fuse same qkv input
* skip onnx clip sdpa
* maintain cache
|
1 year ago |
nihui
|
d9d9cc403d
|
pnnx torch logical ops test (#6069)
|
1 year ago |
nihui
|
6510fe6125
|
discover VK_KHR_shader_float_controls2 (#6068)
|
1 year ago |
nihui
|
ca045ac579
|
gpu info query cooperative matrix properties (#6067)
|
1 year ago |
nihui
|
7f899f2e94
|
update glslang, discover VK_KHR_shader_integer_dot_product and VK_KHR_shader_bfloat16 (#6066)
|
1 year ago |
nihui
|
5f056b8fee
|
pnnx transformers update, part1 (#6054)
* local cache
* skip transformer test on torch<2.1
* zip pnnx
* bool8 to bool
* initialize xlnet weights
* onnx einsum
* onnx reducelogsumexp
* skip tensor as scalar attribute
* skip longformer
|
1 year ago |
nihui
|
6fa649f103
|
clip x86 avx512 mask optimization (#6059)
|
1 year ago |
nihui
|
62ddee3873
|
limit pnnx onnxruntime threads by OMP_THREAD_LIMIT env (#6049)
|
1 year ago |
nihui
|
1d9c78c526
|
update pnnx ci (#6048)
|
1 year ago |
nihui
|
572b6198ad
|
fix subgroup min max size without subgroup size control extension (#6047)
|
1 year ago |
nihui
|
9b7a6eddc8
|
Update README.md
|
1 year ago |
nihui
|
401f59336f
|
add yoloworld example (#6043)
|
1 year ago |
nihui
|
5efea1b014
|
ci x64 coverage update (#6042)
|
1 year ago |
nihui
|
b969f3978d
|
update test-coverage ci (#6040)
|
1 year ago |
nihui
|
27714419db
|
move ci test-coverage to ubuntu25 (#6039)
|
1 year ago |
nihui
|
f371ecd16e
|
fix x86 conv int8 avx512vnni crash (#6038)
|
1 year ago |
nihui
|
bb021da807
|
update release url
|
1 year ago |
nihui
|
7570139879
|
add yolo11 examples (#6037)
|
1 year ago |
nihui
|
90b84cd50e
|
simplevk support nvidia headless vulkan (#6036)
|
1 year ago |
dependabot[bot]
|
8ccf720530
|
Bump pypa/cibuildwheel from 2.23.2 to 2.23.3 (#6027)
Bumps [pypa/cibuildwheel](https://github.com/pypa/cibuildwheel) from 2.23.2 to 2.23.3.
- [Release notes](https://github.com/pypa/cibuildwheel/releases)
- [Changelog](https://github.com/pypa/cibuildwheel/blob/main/docs/changelog.md)
- [Commits](https://github.com/pypa/cibuildwheel/compare/v2.23.2...v2.23.3)
---
updated-dependencies:
- dependency-name: pypa/cibuildwheel
dependency-version: 2.23.3
dependency-type: direct:production
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
|
1 year ago |
nihui
|
305837fd4a
|
fix vulkan driver blacklist (#6034)
|
1 year ago |
nihui
|
2d01465999
|
update rvm example (#6030)
|
1 year ago |
nihui
|
d04d4a0297
|
pnnx ncnn adjust dynamic slice axes for dropping batch index (#6032)
|
1 year ago |
nihui
|
ad0d603623
|
update release url
|
1 year ago |
nihui
|
205ca50e52
|
fix shader compile failure on nvidia (#6029)
|
1 year ago |
nihui
|
b8602a6587
|
pnnx test_inference with torch.no_grad() (#6028)
|
1 year ago |