nihui
4494aadd74
deconvolution dynamic weight ( #5119 )
2 years ago
nihui
14e14a9ae8
slice with indices ( #5103 )
2 years ago
邓实诚
a1e3ebf8e5
implement simplemath ( #4905 )
* complete abs, fmod and sin function in simplemath.h
* remove some unused variables in simplemath.cpp
* modify test-coverage.yml and add some functions to simplemath.cpp
* modify erf.cpp which included math.h
* include platform.h for NCNN_SIMPLEMATH definition
* move utility constants and functions in simplemath.h to simplemath.cpp
* guard simplemath functions with extern "C"
* add NCNN_EXPORT macro in simplemath.h
* include plateform.h and guard all declarations with NCNN_SIMPLEMATH
* clean unused code in test_unaryop.cpp
* guard #include <vector> with NCNN_SIMPLEMATH in benchncnn.cpp
* add 'static' to guard functions that not declarated in header file
* modify sin and cos with better implementation
---------
Co-authored-by: HonestDeng <HonestDeng@users.noreply.github.com>
2 years ago
Yoh
3f437d3f3d
Grid sample op ( #4373 )
* pnnx support grid_sample op
* complete the permute and gridsample operator fusion
* spilt calculation into two stages and support permute fusion
2 years ago
nihui
7b02425246
x86 optimization for convolution int8 winograd unified elempack ( #5054 )
2 years ago
FhqTreap
1d7720efe8
fix test conv1d ( #5049 )
2 years ago
nihui
78aca88d67
elu 4d and selu 4d ( #5047 )
2 years ago
Beq Jal
019176c6b2
selu and shufflechannel on x86 ( #5017 )
2 years ago
Amir Ramezani
7e5fa3ade3
shrink operator ( #5022 )
2 years ago
nihui
c8662cce5e
arm optimization for convolution int8 gemm unified elempack ( #5016 )
2 years ago
Amir Ramezani
0ea587b8c7
celu activation vulkan and onnx conversion ( #5018 )
2 years ago
Beq Jal
bcfec1da33
Celu layer and export to ncnn ( #5019 )
2 years ago
Beq Jal
c851231832
add diag layer and its converter ( #4935 )
2 years ago
Amir Ramezani
695f770eab
erf implementation ( #5012 )
* added erf implementation
* added testcase for erf
* added onnx2ncnn support of erf
2 years ago
nihui
4abadd2ffb
binaryop implicit broadcast B with 1 dimension rank for outer axis ( #4930 )
2 years ago
nihui
c45c01c7c1
enable VK_KHR_cooperative_matrix ( #4823 )
* enable VK_KHR_cooperative_matrix
* add khr cm shader
* update glslang
* print matrix info
2 years ago
nihui
55709708e9
x86 optimization for convolution int8 packed unified elempack ( #4861 )
2 years ago
nihui
1283a19305
pnnx convert torch round trunc ( #4813 )
* update riscv qemu
* c906 test on qemu
* fix qemu aarch64
2 years ago
nihui
9022b7162a
implement all explicit binaryop broadcast types ( #4809 )
* simplify binaryop
* less gpu test
* update binaryop broadcast doc
* do not test atan2 zero
2 years ago
nihui
903ec7c2c9
fix overwrite builtin layer destruction ( #4732 )
* fix overwrite builtin layer destruction
* make modelbin class copyable
* test++
3 years ago
nihui
f893d2440d
innerproduct allow 1 height gemm ( #4730 )
3 years ago
nihui
249b264336
workaround moltenvk error on spec const composite op ( #4714 )
* workaround moltenvk error on spec const composite op
* workaround moltenvk crying on binding image with memory offset
3 years ago
nihui
a37a83d850
clip gelu mish tanh 4d ( #4695 )
3 years ago
nihui
cd5a6098a2
sigmoid and swish 4d ( #4692 )
3 years ago
nihui
c28c8c04a1
multiheadattention attn mask ( #4668 )
3 years ago
nihui
b640574b88
rough vulkan gemm and multiheadattention ( #4618 )
3 years ago
nihui
db628b1b99
allow overwriting built-in layer with custom layer ( #4616 )
3 years ago
nihui
1133a18ca8
x86 and arm optimization for convolution1d packed unified elempack ( #4615 )
3 years ago
nihui
85991e2e0e
test custom option, update ci ( #4609 )
* early return for cpu test
* make nvidia driver happy
* fix gemm x86 threading
3 years ago
nihui
6987efd950
fix scale avx512 ( #4580 )
3 years ago
nihui
dabc4c065f
arm convolution winograd unified elempack ( #4556 )
* update f43 coeffs
* arm convolution winograd unified elempack
* disable bf16s test atm
* test gnu inline asm off
3 years ago
WuJinxuan
ff80ac2955
[ARM] Multiheadattention ( #4463 )
3 years ago
nihui
d0c2738043
update riscv winograd f43 coeffs and fix some warnings ( #4537 )
* update winograd f43 coeffs
* rvv tanh rework
* fix warnings
* rebuild qemu
3 years ago
WuJinxuan
6572da3533
[x86] GroupNorm ( #4471 )
Co-authored-by: EdVince <EdVince@users.noreply.github.com>
3 years ago
nihui
1832da8292
concat 4d ( #4528 )
3 years ago
nihui
fb9cf7982d
eltwise 4d ( #4529 )
3 years ago
nihui
32e2de015e
slice 4d ( #4525 )
3 years ago
nihui
fc6ce4a641
copyto operator ( #4522 )
3 years ago
nihui
242e775d21
pnnx convert torch log10, pow 2 as square ( #4518 )
3 years ago
nihui
246e71c526
implement atan2 ( #4516 )
3 years ago
Fangjun Kuang
92e75105c9
Support torch.cumsum ( #4505 )
3 years ago
nihui
ab4cfbf5b0
enrich ncnn binary broadcast rules ( #4513 )
3 years ago
nihui
dfbcd3e69b
improve vulkan winograd f43 fp16 numerical stability ( #4492 )
3 years ago
nihui
fed99fd35b
gemm output transpose, prepack c ( #4479 )
* mha is now permute and reshape free
* gemm user defined tile mnk param
3 years ago
nihui
2e3e680d77
x86 optimization for packed convolution unified elempack ( #4469 )
3 years ago
nihui
88274827da
x86 optimization for winograd unified elempack ( #4456 )
3 years ago
nihui
15761fc1a6
arm vfpv4 asimdhp asimdfhm optimization for gemm ( #4432 )
3 years ago
nihui
c5640a16c3
gemm x86 multiply alpha beta in post gemm stage, enable one_blob_only ( #4407 )
* gemm x86 multiply alpha beta in post gemm stage, enable one_blob_only
* relax mnk multiple restrictions
* make square tiles in each thread
* sanitize num_threads changes
3 years ago
nihui
fd1ac3c7a0
x86 optimization for gemm unified elempack ( #4387 )
3 years ago
nihui
0736c5b658
Fix c api allocator ( #4360 )
* add some c_api interfaces related to allocator setup.
* fix errors in allocator parameters in c_api.
* test c api allocator
Co-authored-by: zhangtongshe <yuyuyezi@vip.qq.com>
3 years ago