New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Reference implementation for onnx operators (python runtime for onnx) #4483
Conversation
This pull request introduces 8 alerts when merging fce8bad into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 5018923 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 4e247d7 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 3682e11 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 7b53b81 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 452253b into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 4800f3d into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 4a9218a into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging fd14c7c into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging d5dd0ef into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 8707d24 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging b181f44 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 7a68ae8 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 176cbd0 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 504aedd into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 7 alerts when merging 4498dcb into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 8a1343c into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 4677bc4 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 6273ab9 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 1548e7a into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging acd69ed into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 6 alerts when merging 6534f54 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 9 alerts when merging cc8393f into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 9 alerts when merging fc68aa5 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 9 alerts when merging 5315f33 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 11 alerts when merging 83163e0 into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 15 alerts when merging d2827cd into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 15 alerts when merging a4fd14e into f2f1408 - view on LGTM.com new alerts:
|
This pull request introduces 15 alerts when merging c31556f into f2f1408 - view on LGTM.com new alerts:
|
Signed-off-by: xadupre <xadupre@microsoft.com>
onnx/reference/README.md
Outdated
|
||
This is a first attempt to provide an implementation for all operators | ||
defined by onnx. This is a pure python implementation. | ||
Mismatches may remain between the official specifications and the implementation here. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: "specifications" => "specification"
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
Signed-off-by: xadupre <xadupre@microsoft.com>
It seems that some tests in reference_evaluator_backend_test.py and reference_evaluator_test.py failed (see CI). One failed Windows CI uses x86 machine and another CI uses old NumPy version 1.16.6. Since other CIs are fine, is it because NumPy behaves slightly differently in Windows x86 machine and old NumPy version 1.16.6? |
…onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com>
…onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com>
* Fix enforce-style CI error: solve empty-body issue raised by mypy 0.990 (#4643) * solve empty-body issue raised by mypy 0.990 Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix black Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Add 32 and 64 bit unsigned integers as tensor types in IR (#4634) These types are already extensively used throughout in the operator definitions, but were missing in the definition of the intermediate representation. Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Reference implementation for onnx operators (python runtime for onnx) (#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * fix parse repo name Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * revert change Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: Chun-Wei Chen <jacky82226@gmail.com> Co-authored-by: Christian Bourjau <cbourjau@users.noreply.github.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Co-authored-by: Xavier Dupré <xadupre@users.noreply.github.com> Co-authored-by: sdpython <xavier.dupre@gmail.com>
I can either skip the test if numpy is too old and add in the documentation that the reference implementation only supports numpy >= 1.21.5 or fix the test one by one for numpy == 1.16.6. What do you prefer? |
Although ideally ONNX should always support the minimum required NumPy version (for now it's 1.16.6), I believe sometimes NumPy behaves quite differently in different environments (hardware, version). It might take a lot of time to fix NumPy behavior gap one by one and it's pretty close to the code freeze date (11/18). Thus, skipping them for now is fine to me. BTW, besides old NumPy version issue (1.16.6), NumPy package for Windows x86 seems to behave differently as well so the test failed in release _win CI for x86 machines. |
Just speak aloud: to solve this kind of failure, is it possible that we can have loose tolerance when comparing output result from onnx reference and expected? |
We can. I had to do it. There are dictionaries I use to specify different tolerances for every test: https://github.com/onnx/onnx/blob/main/onnx/test/reference_evaluator_backend_test.py#L764. |
I agree that for this release, doing the simplest is best. Ok with me to see it currently supports only numpy >= 1.21.5, and add support for other numpy versions later on. |
See #4658, I disabled only three tests related to cast into/from bfloat16. |
…onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com>
* Fix enforce-style CI error: solve empty-body issue raised by mypy 0.990 (onnx#4643) * solve empty-body issue raised by mypy 0.990 Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix black Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Add 32 and 64 bit unsigned integers as tensor types in IR (onnx#4634) These types are already extensively used throughout in the operator definitions, but were missing in the definition of the intermediate representation. Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Reference implementation for onnx operators (python runtime for onnx) (onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * fix parse repo name Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * revert change Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: Chun-Wei Chen <jacky82226@gmail.com> Co-authored-by: Christian Bourjau <cbourjau@users.noreply.github.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Co-authored-by: Xavier Dupré <xadupre@users.noreply.github.com> Co-authored-by: sdpython <xavier.dupre@gmail.com>
I tried using this, and get an error about
|
Is it possible to share the file foo.onnx you used? |
I didn't expect that the model file was relevant, since the runtime is reporting |
I just tried with the main branch on github and it works. Which version of onnx are you using? The implementation for ml operators were introduced after the last published release on pypi. |
…onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com>
* Fix enforce-style CI error: solve empty-body issue raised by mypy 0.990 (onnx#4643) * solve empty-body issue raised by mypy 0.990 Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix black Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Add 32 and 64 bit unsigned integers as tensor types in IR (onnx#4634) These types are already extensively used throughout in the operator definitions, but were missing in the definition of the intermediate representation. Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * Reference implementation for onnx operators (python runtime for onnx) (onnx#4483) * add backend test Signed-off-by: xadupre <xadupre@microsoft.com> * fix backend Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add more unary operators Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix type issue in operator Constant Signed-off-by: xadupre <xadupre@microsoft.com> * fix misspelling Signed-off-by: xadupre <xadupre@microsoft.com> * switch to relative import Signed-off-by: sdpython <xavier.dupre@gmail.com> * apply isort Signed-off-by: sdpython <xavier.dupre@gmail.com> * pylint Signed-off-by: sdpython <xavier.dupre@gmail.com> * black Signed-off-by: sdpython <xavier.dupre@gmail.com> * less lint issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * make OpRun an abstractclass Signed-off-by: sdpython <xavier.dupre@gmail.com> * add and, or, xor, argmax, argmin Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GlobalAveragePool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BatchNormalization Signed-off-by: sdpython <xavier.dupre@gmail.com> * add BitShift Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator cast Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator CastLike Signed-off-by: sdpython <xavier.dupre@gmail.com> * add operator Celu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add the possibility to test custom operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Compress Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Concat Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConcatFromSequence Signed-off-by: sdpython <xavier.dupre@gmail.com> * add CumSum Signed-off-by: sdpython <xavier.dupre@gmail.com> * add ConstantOfShape, DepthToSpace, SpaceToDepth, Det, DequantizeLinear Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix string issue Signed-off-by: sdpython <xavier.dupre@gmail.com> * add DFT Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Div, DropOut Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Einsum, Elu, Equal, Erf Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Expand, EyeLike, Flatten, Gather, GatherElements, GatherND Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Floor, Gemm, GlobalAveragePool, GlobalMaxPool Signed-off-by: sdpython <xavier.dupre@gmail.com> * add GRU, HardSigmoid, Hardmax Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add IsInf, Isnan, LayerNormalization, LeakyRelu, Less, LessOrEqual Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Loop, LpNormalization, LogSoftmax, Softmax, LRN, LSTM Signed-off-by: sdpython <xavier.dupre@gmail.com> * add unit test on Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Slice, Unsqueeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * complete Loop Signed-off-by: sdpython <xavier.dupre@gmail.com> * sort test Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix a couple of issues Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Constant_1 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Where Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Max, Mean, Min, Mod Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Not, OneHot, NonZero, NLL Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Optional Signed-off-by: sdpython <xavier.dupre@gmail.com> * remove unnecessary constructor Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Pad, Pow, PRelu Signed-off-by: sdpython <xavier.dupre@gmail.com> * fix pad Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add reduce operators Signed-off-by: sdpython <xavier.dupre@gmail.com> * add quantize, range, reciprocal Signed-off-by: sdpython <xavier.dupre@gmail.com> * add dynamic quantize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RandomUniform Signed-off-by: sdpython <xavier.dupre@gmail.com> * add random Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Trilu, Selu Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * bugs Signed-off-by: sdpython <xavier.dupre@gmail.com> * add soft...3 Signed-off-by: sdpython <xavier.dupre@gmail.com> * add squeeze Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Resize Signed-off-by: sdpython <xavier.dupre@gmail.com> * add window functions Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add Scatter Signed-off-by: sdpython <xavier.dupre@gmail.com> * add RNN Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add last python ops Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * add domaine ai.onnx.preview.training Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * lint Signed-off-by: sdpython <xavier.dupre@gmail.com> * type Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add ReverseSequence Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix reverse sequence Signed-off-by: xadupre <xadupre@microsoft.com> * add tile Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceMap Signed-off-by: xadupre <xadupre@microsoft.com> * fix sequence operators Signed-off-by: xadupre <xadupre@microsoft.com> * add operator scan Signed-off-by: xadupre <xadupre@microsoft.com> * add op MelWeightMatrix Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add MatMulInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add Optional Signed-off-by: xadupre <xadupre@microsoft.com> * add instance normalization Signed-off-by: xadupre <xadupre@microsoft.com> * first step with standard functions Signed-off-by: xadupre <xadupre@microsoft.com> * implement method eval Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * overriden Signed-off-by: xadupre <xadupre@microsoft.com> * modified Signed-off-by: xadupre <xadupre@microsoft.com> * attributes Signed-off-by: xadupre <xadupre@microsoft.com> * extend support for linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * support linked attributes Signed-off-by: xadupre <xadupre@microsoft.com> * remove schema_domain Signed-off-by: xadupre <xadupre@microsoft.com> * add conv Signed-off-by: xadupre <xadupre@microsoft.com> * finalize conv Signed-off-by: xadupre <xadupre@microsoft.com> * add ConvInteger Signed-off-by: xadupre <xadupre@microsoft.com> * add CropAndPad Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add qlinearconv Signed-off-by: xadupre <xadupre@microsoft.com> * add im2col Signed-off-by: xadupre <xadupre@microsoft.com> * update im2col Signed-off-by: xadupre <xadupre@microsoft.com> * fix im2col Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add col2im Signed-off-by: xadupre <xadupre@microsoft.com> * replace numpy. by np. Signed-off-by: xadupre <xadupre@microsoft.com> * complete replace numpy into np Signed-off-by: xadupre <xadupre@microsoft.com> * List[List|np.ndarray] Signed-off-by: xadupre <xadupre@microsoft.com> * review comments Signed-off-by: xadupre <xadupre@microsoft.com> * refactor unit test Signed-off-by: xadupre <xadupre@microsoft.com> * fix col2im Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add conv_transpose Signed-off-by: xadupre <xadupre@microsoft.com> * add convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * finalize ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * modif Signed-off-by: xadupre <xadupre@microsoft.com> * add grid_sample Signed-off-by: xadupre <xadupre@microsoft.com> * add max pool Signed-off-by: xadupre <xadupre@microsoft.com> * one move for max_pool Signed-off-by: xadupre <xadupre@microsoft.com> * raise exception when not implemented Signed-off-by: xadupre <xadupre@microsoft.com> * add maxunpool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * add non max suppression Signed-off-by: xadupre <xadupre@microsoft.com> * add has_attribute Signed-off-by: xadupre <xadupre@microsoft.com> * add max_pool_2d Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * merge from main branch Signed-off-by: xadupre <xadupre@microsoft.com> * add roi align Signed-off-by: xadupre <xadupre@microsoft.com> * add RoiAlign Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * simplifies overriden attributes Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * fix tfidfvectorizer Signed-off-by: xadupre <xadupre@microsoft.com> * improve tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * rename runtime into fn Signed-off-by: xadupre <xadupre@microsoft.com> * rename into funconnx Signed-off-by: xadupre <xadupre@microsoft.com> * rename Inference into ProtoRun Signed-off-by: xadupre <xadupre@microsoft.com> * fix Tfidf Signed-off-by: xadupre <xadupre@microsoft.com> * style Signed-off-by: xadupre <xadupre@microsoft.com> * add bitwise_not Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * add operator upsample Signed-off-by: xadupre <xadupre@microsoft.com> * improve code Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * remove one warning Signed-off-by: xadupre <xadupre@microsoft.com> * support bfloat16 in cast operation Signed-off-by: xadupre <xadupre@microsoft.com> * refactoring of backend test Signed-off-by: xadupre <xadupre@microsoft.com> * add annotation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * improve support Signed-off-by: xadupre <xadupre@microsoft.com> * add more tests Signed-off-by: xadupre <xadupre@microsoft.com> * add SequenceErase Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix bugs in Conv Signed-off-by: xadupre <xadupre@microsoft.com> * fix verbose test Signed-off-by: xadupre <xadupre@microsoft.com> * implement group for conv Signed-off-by: xadupre <xadupre@microsoft.com> * add conv 3d Signed-off-by: xadupre <xadupre@microsoft.com> * more comments Signed-off-by: xadupre <xadupre@microsoft.com> * complete max pool Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * resize Signed-off-by: xadupre <xadupre@microsoft.com> * roi Signed-off-by: xadupre <xadupre@microsoft.com> * fix rnn, gru Signed-off-by: xadupre <xadupre@microsoft.com> * improve stft, still not accurate Signed-off-by: xadupre <xadupre@microsoft.com> * rename into reference Signed-off-by: xadupre <xadupre@microsoft.com> * remove unnecessary type ignore Signed-off-by: xadupre <xadupre@microsoft.com> * reorder import Signed-off-by: xadupre <xadupre@microsoft.com> * loop Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * use absolute import Signed-off-by: xadupre <xadupre@microsoft.com> * import Signed-off-by: xadupre <xadupre@microsoft.com> * rename into ReferenceEvaluator Signed-off-by: xadupre <xadupre@microsoft.com> * rename files Signed-off-by: xadupre <xadupre@microsoft.com> * fix conv Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix maxpool Signed-off-by: xadupre <xadupre@microsoft.com> * improves scatter elements Signed-off-by: xadupre <xadupre@microsoft.com> * remove duplicated function bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * annotation Signed-off-by: xadupre <xadupre@microsoft.com> * remove pprint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * fix ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * improves conv transpose Signed-off-by: xadupre <xadupre@microsoft.com> * black Signed-off-by: xadupre <xadupre@microsoft.com> * fix pad in ConvTranspose Signed-off-by: xadupre <xadupre@microsoft.com> * fix convtranspose Signed-off-by: xadupre <xadupre@microsoft.com> * better documentation for bfloat16_to_float32 Signed-off-by: xadupre <xadupre@microsoft.com> * extend annotation Signed-off-by: xadupre <xadupre@microsoft.com> * improves resize Signed-off-by: xadupre <xadupre@microsoft.com> * fix Resize Signed-off-by: xadupre <xadupre@microsoft.com> * complete momentum Signed-off-by: xadupre <xadupre@microsoft.com> * fix stft Signed-off-by: xadupre <xadupre@microsoft.com> * better comment Signed-off-by: xadupre <xadupre@microsoft.com> * update the number of mandatory tests to pass Signed-off-by: xadupre <xadupre@microsoft.com> * documentation Signed-off-by: xadupre <xadupre@microsoft.com> * merge Signed-off-by: xadupre <xadupre@microsoft.com> * fix context dependent implementation Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * lint Signed-off-by: xadupre <xadupre@microsoft.com> * misspelling Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: sdpython <xavier.dupre@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * fix parse repo name Signed-off-by: Yi Zhang <zhanyi@microsoft.com> * revert change Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Signed-off-by: Yi Zhang <zhanyi@microsoft.com> Signed-off-by: Christian Bourjau <christian.bourjau@quantco.com> Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: sdpython <xavier.dupre@gmail.com> Co-authored-by: Chun-Wei Chen <jacky82226@gmail.com> Co-authored-by: Christian Bourjau <cbourjau@users.noreply.github.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Co-authored-by: Xavier Dupré <xadupre@users.noreply.github.com> Co-authored-by: sdpython <xavier.dupre@gmail.com>
Description and Motivation
This PR addresses issue #4432. It proposes a reference implementation for onnx operators through a runtime following the same API as onnxruntime. This python runtime can also be used to test function body. Currently, there is no way to check the body of a function validates the backend tests. This can also be used to validate other runtime or add more backend tests.
This work comes from a python runtime I wrote to easily check the onnx model built with a converter. onnxruntime is a fast runtime but it cannot be easily use to debug a model, like display intermediate results. This is another goal of this runtime. The runtime only depends on python and numpy.
The current implementation covers 99.4% of the backend tests. It skips the following tests: