* regsiter part of opset13 cpu kernels; fix a bug in func impl; adjust reshapefusion order
* remove useless function
Co-authored-by: Cheng Tang <chenta@microsoft.com>
Adds EinSum operator (purely an EP kernel, not a dedicated DML operator), which takes an equation string and depending on the specifics is capable of representing: identity, diag, trace, transpose, reduce sum, dot product, matmul, elementwise multiplication, inner product, outer product.
The DML EP recognizes many of them (identity, transpose, reduce sum, 1D dot product, matmul, elementwise multiplication), but defers to CPU when not supported (extended inner product, outer product, diag, trace, arbitrary batch ellipsis).
https://github.com/onnx/onnx/blob/master/docs/Operators.md#Einsum
WindowsAI PR: https://microsoft.visualstudio.com/DefaultCollection/WindowsAI/_git/WindowsAI/pullrequest/5100608
Related work items: #27469790
* Optimized MatMulGrad for dB when B's shape is 2D
* Refactor for ConstantScalarNode
Co-authored-by: Sherlock Huang <bahuang@OrtTrainingDev3.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
* matching multiple choice between new and old apis
* update according to reviewer's comments
Co-authored-by: liqun <liqun@OrtTrainingDev4.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
* support Normalized_0_1 and Normalized_1_1
* add tests for Normalized_1_1
* fix build error
* fix imagetests failure
* support denterization and add more tests
* fix build
* remove added models
* disable gpu tests for CPU pipeline
* refactor based on comments and moved two added models
* merge normalizer and Denomalizer into NominalRangeConverter
* add comments
* little change
* fix build failure for amd64
* cuda kernel support
* on comments
* test UT
* test UT
* revert settings
* attempt to fix broken UT
* corrected UT fix
Co-authored-by: Ethan Tao <ettao@microsoft.com>
* Extend C++ API for Map/Sequence Type Info (#3517)
Expose functionality to view type information about sequences/maps
to C++ API.
- Add functions
- `TypeInfo::GetSequenceTypeInfo`
- `SequenceTypeInfo::GetSequenceElementType`
- `TypeInfo::GetMapTypeInfo`
- `MapTypeInfo::GetMapValueType`
- `MapTypeInfo::GetMapKeyType`
- Add structs
- `SequenceTypeInfo`
- `MapTypeInfo`
Co-authored-by: Dudeldu <mustermann.informatik@gmail.com>
Co-authored-by: Jonas-Heinrich <Jonas@JonasHeinrich.com>
* Extend tests to cover new type info functionality for sequences and maps
- two new test case in test_nontensor_types for maps and sequences
Co-authored-by: Jonas-Heinrich <Jonas@JonasHeinrich.com>
* support Normalized_0_1 and Normalized_1_1
* add tests for Normalized_1_1
* fix build error
* fix imagetests failure
* support denterization and add more tests
* fix build
* remove added models
* disable gpu tests for CPU pipeline
* refactor based on comments and moved two added models
* merge normalizer and Denomalizer into NominalRangeConverter
* add comments
* little change
* Export GPT-2 ONNX model without postion_ids and attention_mask inputs
* allow benchmark_gpt2 on user's model
* refactor: get_dummy_inputs returns a data class.
* check whether the node has been casted before
* check casted node logically
* better naming convention
* nit: extra space
* change to skip for Cast Node
* remove hasNodeBeenCast
* Add a Unit test
* Add test onnx file
* nit: naming convention and comments
* check CI: try to remove test
* move test to existing test file
* Copy samples to build folder and load models from there. Fix CI
* This PR also includes a fix to path validation for save_as_onnx API
* Add torchtext to CI for GPU training
* Remove new frontend tests from CI
Co-authored-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>
