issues
search
onnx
/
optimizer
Actively maintained ONNX Optimizer
Apache License 2.0
650
stars
90
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
fix fuse_concat_into_reshape when concat has multi consumers
#103
daquexian
closed
2 years ago
0
fix shape gather
#102
HSQ79815
closed
2 years ago
0
why using the same cmake but got error
#101
lucasjinreal
opened
2 years ago
1
-1 mean dynamic shape or Static shape?
#100
xiaobochen123
closed
1 year ago
2
No module named onnxoptimizer.__main__; 'onnxoptimizer' is a package and cannot be directly executed
#99
KOLANICH
closed
1 year ago
4
It builds dependencies instead if reusing installed in the system.
#98
KOLANICH
opened
2 years ago
1
`onnxoptimizer` is missing support for python 3.10
#97
bcm-at-zama
closed
1 year ago
5
Is a lower version of cmake be compatible with onnxoptimizer?
#96
gglin001
closed
2 years ago
2
eliminate nop concat and eliminate nop split
#95
HSQ79815
closed
2 years ago
0
simplify can't find graph
#94
ARozental
closed
2 years ago
1
CI publish to pypi
#93
daquexian
closed
2 years ago
0
Fix eli slice after shape opset9
#92
daquexian
closed
2 years ago
0
fix bug of eliminate_if_with_const_cond and subgraph param
#91
daquexian
closed
2 years ago
0
abstract function to pass_util.h
#90
HSQ79815
closed
2 years ago
0
add replace passtype
#89
HSQ79815
closed
2 years ago
0
upgrade fuse_concat_into_reshape to support concat->cast->reshape pattern
#88
HSQ79815
closed
2 years ago
0
How to add empty input to a node
#87
wjj19950828
closed
2 years ago
8
return sorted passes name when using GetAvailablePasses or GetFuseAnd…
#86
HSQ79815
closed
2 years ago
0
replace einsum with matmul
#85
HSQ79815
closed
2 years ago
0
fix FuseConcatAndReshape bug and add tests
#84
daquexian
closed
2 years ago
0
Use is_constant_initializer
#83
daquexian
closed
2 years ago
0
eliminate slice shape
#82
HSQ79815
closed
2 years ago
0
rename and minor update on shape gather pass
#81
daquexian
closed
2 years ago
0
eliminate nop reshape
#80
HSQ79815
closed
2 years ago
0
Upgrade cibuildwheel to latest
#79
daquexian
closed
2 years ago
0
Upgrade minimum cmake and update ci
#78
daquexian
closed
2 years ago
0
fuse constant shape for reshape
#77
HSQ79815
closed
2 years ago
0
eliminate nop gather shape
#76
HSQ79815
closed
2 years ago
0
set unique name for node
#75
HSQ79815
closed
2 years ago
0
Drop py36 and upgrade c++ standard to 17
#74
daquexian
closed
2 years ago
0
Support large model
#73
HSQ79815
closed
2 years ago
0
rename input/output
#72
HSQ79815
closed
2 years ago
0
Eliminate nop compute with unit
#71
HSQ79815
closed
1 year ago
0
fix eliminate_duplicate_initializer
#70
HSQ79815
closed
2 years ago
1
add eliminate_nop_expand pass
#69
HSQ79815
closed
2 years ago
1
Fix docstring
#68
maekawatoshiki
closed
2 years ago
3
execute "python3 -m onnxoptimizer model.onnx output.onnx" error
#67
try1995
closed
2 years ago
3
fuse_matmul_add_bias_into_gemm not working with batch size
#66
erelon
closed
1 year ago
2
fuse pad into avgpool and maxpool
#65
luliang2luliang
closed
2 years ago
3
upload sdist
#64
daquexian
closed
2 years ago
0
Upload sdist and m1 wheel
#63
daquexian
closed
1 year ago
1
Add a new fusion pass with pattern of constant and reshape
#62
jiangjiajun
opened
2 years ago
1
skip custom ops
#61
minrui-hust
closed
2 years ago
3
does this repo have some records of inference acceleration?
#60
MasterGG
opened
3 years ago
0
Unable to install on Mac M1
#59
hossein1387
closed
2 years ago
12
fuse_matmul_add_bias_into_gemm: not working for just MatMul
#58
AlexMontgomerie
closed
1 year ago
3
Is there a specific document for each .h header file in onnxoptimizer / passes?
#57
yeyuxmf
opened
3 years ago
0
optimize passes of mean ？
#56
kugua-FYZ
opened
3 years ago
1
why default pass only have 'eliminate_nop_transpose', 'eliminate_nop_pad','fuse_consecutive_transposes', 'fuse_transpose_into_gemm'?
#55
Veiiling
closed
1 year ago
0
[Pass request] Fuse Pad - Pooling (Avg/MaxPool)
#54
phager90
opened
3 years ago
0
Previous
Next