1. Parallel all the activations ops.
2. Parallel the performance critical path of the LRN op, which makes the ONNX model zoo googlenet model runs 60% faster(latency reduced from 21ms to 13ms).
3. Make the Gemm-Activation fusion support with all the activations ops. Before this change, it only supports LeakyRelu/Relu/Sigmoid/Tanh.
4. Delete onnxruntime/test/framework/op_kernel_test.cc because the file is almost empty.
5. Remove the loggings in KernelRegistry::TryFindKernel, return Status with error message instead.
* Add a helper script to more easily create a test directory for use with onnx_test_runner or onnxruntime_perf_test.
Add example script that can be used as a base for performance testing a model with a variety of input sizes.
Add __init__.py so files in this directory can be imported in other scripts.
* Fix some flake8 warnings.
Add example of specifying attribute for op.
* Add ability for test dir creation to fill in all missing input data with random values.
Add example of using test dir creation this way
* Fix a crash in Reshape
Reshape doesn't handle 0 input dimension properly, which leads to a
division by zero
* Fix reshape fusion
https://github.com/microsoft/onnxruntime/pull/3554 introduced a bug:
initializers can now come before Shape->Gather->Unsqueeze chains; if
those initializers have more than 1 element, expected dimensions in the
chains are now incorrect.
Authored-by: Max Kalinin <makalini@microsoft.com>
* Fix first instance where the calculation for TryBatchParallelFor was incorrect. Other usages need to be validated.
* Fix some other usages of the threadpool.
Increase test comparison tolerance. Add output of random seed value for easier debugging later. Unify RandomValueGenerator::Uniform() to consistently use [min, max) interval.
Revert softmax kernel implementation to the previous version.(commit 4d26f2ce86)
For
1. Reducing the binary size. Currently the single kernel is 500KB.
2. Removing dependency on Eigen::ThreadPoolDevice
* initial change to transformer.py
* prepare e2e transformer tests
* refactor transformer tests
* put test python files in a flat folder
* fix typo pip install transform(s)
* python 3.6
* python version to 3.6 in install_ubuntu.sh
* remove argparser
* to use opset ver 12
* workaround loss_scale naming patch in case of loss_fn_
* assign self.loss_fn_ so it can be checked
* skip a few un-needed post-process steps
* fix loss_scale_input_name, clean up post process steps
* skip non-frontend tests
* move cpu/cuda related files to coresponding cpu/cuda folder (#3668)
Co-authored-by: Weixing Zhang <wezhan@microsoft.com>
* type cast for ratio is not necessary for dropout (#3682)
Co-authored-by: Weixing Zhang <wezhan@microsoft.com>
* thrustallocator is not needed since cub is used directly for gather now. (#3683)
Co-authored-by: Weixing Zhang <wezhan@microsoft.com>
* GatherND-12 Implementation (#3645)
* Renamed, UT passing
* Move GatherND CUDA Kerenl into onnxruntime
* Merge GatherNDOpTest
* Refactor Test code
* Merge CPU Kernel Impl
* Handle Negative Indice, Fix UT
* Improve CUDA kernel to handle negative index
* Minor Fixes
* Preserve GatherND-1 Cuda kernel
* Fix Mac build
* fix UT
* Fix Build
* fix GatherNDOpTest.double > CUDA error cudaErrorInvalidDeviceFunction:invalid device function
Co-authored-by: Sherlock Huang <bahuang@OrtTrainingDev3.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
Co-authored-by: Peng Wang (pengwa) <pengwa@microsoft.com>
* update with reviewers' comments
* testBertTrainingGradientAccumulation was not using rtol and may fail occasionally with small (e-06) difference
* fix merge mistakes
Co-authored-by: liqun <liqun@OrtTrainingDev4.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
Co-authored-by: Weixing Zhang <weixingzhang@users.noreply.github.com>
Co-authored-by: Weixing Zhang <wezhan@microsoft.com>
Co-authored-by: Sherlock <baihan.huang@gmail.com>
Co-authored-by: Sherlock Huang <bahuang@OrtTrainingDev3.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
Co-authored-by: Peng Wang (pengwa) <pengwa@microsoft.com>
* Add TrySimpleParallerFor so that there's a path with OpenMP awareness for SimpleParallelFor. Makes it consistent with [Try]BatchParallelFor and [Try]ParallelFor.
Update TopK to check for the number of threads better, and to use TrySimpleParallelFor.
* Update doco to mention TrySimpleParallelFor
* Refactor GatherND CPU Kernel (Renaming & Simplify)
* Add batch_dim=1 or 2, negative slices tests
* Rename gather_nd_gard_impl.cu
* Use dispatcher to refactor CUDA GatherND/GatherNDGrad
* Change GatherNDBase::CommonComputeKernel --> GatherNDBase::PrepareCompute
* Use HasCudaEnvironment instead of __CUDA_ARCH__ for some double type tests
* Change naming of moments to Moment_x_<weight_name>
* Add checkpointing code and zero checkpoint aggregation
* Correct aggregation for LAMB, cleanup
* Add simple checkpointing test
* Add test for zero checkpoint aggregation
* Fix tests
* fix test
* Review changes
* Fix test after review comment fix
* Fix API, test
* Fix test after API change
* Decouple save load from ORTTrainer
* Add flag to not break checkpointing with ORTModel'
Co-authored-by: aishwarya bhandare <aibhanda@OrtTrainingDev3.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
