Shape Inference and Model Optimization before Quantization
Model quantization with QDQ format, i.e. inserting QuantizeLinear/DeQuantizeLinear on
the tensor, requires tensor shape information to perform its best. Currently, shape inferencing
works best with optimized model. As a result, it is highly recommended to run quantization
on optimized model with shape information.
This change adds code for model optimization and shape inferencing of the following three steps:
1. Symbolic shape inference.
2. Model optimization
3. ONNX shape inference
At the same time we should recommend model optimization should be turned off during quantization.
As the optimization might change the computation graph, making it harder for the QDQ debugger
to locate matching tensors between original and the quantized models.
* fix shape mismatch in FuseConv
* remove zeroed bias
* offset Z dim
* append UT
* add testing model
* remove output
* remove commented
* fix comments
* refactor output msg
* narrowly restrict the use of cudnn...ActFwd
* reset changes in cudnn_common
* add test cases covering all path
* move cases to conv test
* remove extra space
* fix build err
Co-authored-by: Randy Shuai <rashuai@microsoft.com>
* Add first pass of rocm kernel profiler
* Clean up rocm_profiler. Format args. Demangle kernel names.
Add Api EventRecords
* Remove debug output
* Temporarily disable profiling unit test 'api record check' for cupti
* Fix compile error for non-gpu builds
* Use common file for demangle and pid/tid. Namespace ThreadUtil. Fix gpu buffer clearing.
* Merge demangle into profiler_common
* Merge demangle into profiler_common part 2
* Style cleanup
* Resolve linking issues via ProviderHost interface
* Demangle cuda kernel names
* Clean up comments
* Fix formatting
* Fix anal retentive formatting
* upgrade cuda version on ci pipelines
* keeping folder name same
* keeping folder name same
* setting manual seed for primitive test case
* resolving comments
* changing atol and rtrol only for test case
Co-authored-by: Adam Louly <adamlouly@microsoft.com@orttrainingdev7.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>
* sync graph proto in node's attributes
* Don't fuse nodes of control flow op until later in control flow op level
* remove unnecessary ep funtions
* remove unnecessary ep funtions
* remove unnecessary ep funtions
* missing 'override' keyword which makes MacOS/Web CI fail
* Add one more test run for Test3LayerNestedSubgraph with disabling graph optimization
* Update the comments to better understand the 4 cases
* moving training pipelines from cuda 11.5 to 11.6 and deprecating cuda 11.3
* change to cuda 11.6.2
* change pytorch's & torchvision's cuda version to 11.6
* specify deps version to 11.6.2
* update pytorch and torch text version
* torch 1.12.1
* change torchvision and torchtext version to be compatible with torch 1.12.1
* change cuda to 11.6 for cuda_home comaptibility
Co-authored-by: Adam Louly <adamlouly@microsoft.com@orttrainingdev7.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>
* Add asm statement to model.mm to force linker to link against CoreML.Framework.
Update targets.xml as per Rolf's suggestions
* Remove explicit numpy version from macos build. We don't specify it for other CIs and the version specified doesn't have a pre-built 3.10 wheel. This leads to the CI attempting to build numpy which fails.
Fix arithmetic overflow warning. Suggested fix by static analysis tool
Arithmetic overflow: Using operator '+' on a 4 byte value and then casting the result to a 8 byte value.
Cast the value to the wider type before calling operator '+' to avoid overflow (io.2).
* Fix SAME_UPPER/SAME_LOWER (auto_pad attribute) in ConvTranspose
* Bump ONNX 1.10.2 globally
* load ONNX_VERSION from VERSION_NUMBER
* /
* revert deprecate warning in ORT 1.12
* add a comment about why removing cntk_simple_seg
* correct the implem in DML as well
