* Use local ort python package in server model tests
* Create symlink for onnxruntime during test
* Using generated _pb2.py in the build folder
* Generate onnx_ml_pb2.py from ONNX CMakeList.txt
* Update model tests for python package path
* Only use onnx python package from build
* Revert some changes for pb2.py generation
* Refactor C# to handle x86
* update run script
* Add Native win x86 tests
* Add native x86 tests for Linux
* Update linux tests scripts to control which tests are run
* update linux image name for x86 to prevent using cached image
* update to not run unit python unit tests unless pybind is specified
* remove --build_wheel as a core common arg. Python cannot run on x86 build
* update OrtGetNumOfDimensions to OrtGetDimensionsCount in rest of C#
* Change function signature
* Convert compute to use custom op style APIs
* Remove dead CustomOp function
* Use CustomOp API in TensorRT EP
* Switch to new API in ngraph
* fix ngraph due to incompatibility with PR1019
* temp disable onnx backend test test_operator_repeat_dim_overflow_cpu due to failure in ngraph ep
* disable tiny_yolov2 for ngraph
* temporarily disable operator_repeat_dim_overflow for ngraph due to graph resolve failure
* initial version. there's a bug
* allow nodes with multiple outputs to be removed if only one output is used by downstream ops
* move node output checks into their own methods
* add test data file
* address comments by @kkaranasos and @skottmckay
* address more comments by @kkaranasos
* a comment to clarify the position of the mask output
* Finer control over when Python tests are run
* add --build_wheel to linux pipeline, instead of run_build.sh
* add --build_wheel to all ci configurations
* update per review comments
* Enable model tests for ORT Server in build script
* Nightly build pipeline definition
* Force prune docker images
* Clean up containers before prune images
* More C++ API improvements and conversions
* Mark more constructors as explicit
* Fix CSharp function name changes
* Change more test cases to use C++ API
* Intial commit
* Rename DynamicPad to Pad
* More changes
* Add Unique operator
* Revert accidental check-in
* Fix CUDA Pad to align with changes
* More changes
* Fix more CUDA pad source files
* More fixes
* More changes
* More changes
* Avoid vector copy
* Update vector validation logic
* Fix build failures
* Fix build
* Fix build failure
* Fix tensorrt build
* Fixes memory leaks in executor and adds executor tests
* Remove logging test from executor_test.cc
* Uses RAII to allocate and freee MLValue buffers
* Change private member buffer to buffer_
