* update TVM
* get alignment constant from TVM
* update TVM_VM_SetInputs to upstream with TVM API
* fix CI issue: update TVM EP dependencies
* add sudo
* revert changes needed to install missing package
* add package for TVM EP CI
Co-authored-by: Valery Chernov <valery.chernov@deelvin.com>
Co-authored-by: KJlaccHoeUM9l <wotpricol@mail.ru>
This reverts commit 1f2c926. Because it makes our packaging pipeline crash
Error message:
[ RUN ] QLinearConvTest.Conv3D_S8S8_Depthwise
Test #1: onnxruntime_test_all ...................Subprocess killed***Exception: 838.24 sec
We haven't successfully reproduced the bug on a real ARM64 hardware. Currently we only saw it showed up with qemu. More investigations are on-going.
* Initial changes with comments on potential unit test changes.
* Update tests to disable TransposeOptimizer as that's simpler.
Add some extra comments.
Cleanup.
* Update comments in TransformGraph
* Add regression test.
Add limitation that transpose optimizer will ignore assigned nodes that do not match the context EP if that is set.
* Fix test. I removed a trailing Transpose after initial validation to simplify but that changed things so that the transpose optimizer didn't kick in, and the DQ -> Transpose -> Q was actually converted to a single Transpose by the CPU EP QDQ handling. Same end result in most builds so the subtle difference wasn't noticed, but in a build without contrib ops the CPU EP QDQ handling is disabled so the end result was different.
Update the test to re-instate the trailing Transpose so transpose optimizer alters the graph as desired.
* Don't run level 1 optimizers after partitioning as they don't guarantee to handle EP assignment for new nodes they create.
* Initiate Ort SNPE EP
* fix snpe ep windows build which is caused by the utility method (ToUTF8String) name change on master
* correct the source path for libonnxruntime.so while building for andorid package
* add AdditionalDependencies for amr64
* On MS-Windows, the patchfile must be a text file, i.e. CR-LF must be used as line endings. A file with LF may give the error: "Assertion failed, hunk, file patch.c, line 343," unless the option '--binary' is given.
* fix build failure if snpe is not enabled
* update doc for contrib op
* separate out snpe ep settings to onnxruntime_snpe_provider.cmake
* renaming according review comments
* update according review comments
* Implement XNNPACK support via an EP.
* Layout transform uses the GraphPartitioner infrastructure.
* Node fusion is supported.
* Conv and MaxPool implementations were ported from Changming's PR.
* Added optional mutex in InferenceSession::Run as we only want to allow sequential calls if xnnpack is enabled
* add missing version
* opset check
* fix format
* reject fusion if type not allowed
* per comments
* trigger new build
Co-authored-by: Ethan Tao <ettao@microsoft.com@orttrainingdev7.d32nl1ml4oruzj4qz3bqlggovf.px.internal.cloudapp.net>
* Add better message for subgraph output coming directly from outer scope value.
* Use regex to match value name as the test model is processed in a different order on different platforms.
* skeleton change
* adam compute kernels
* add rtol/atol for tests
* some clean up
* optional outputs
* more clean up
* add tests
* adamw mode=1 test pass
* clean up tests
* add HF AdamW test cases
* refactor adam test file
* make test pass
* all test pass, fix comments
* rename to adamw
* make test pass again
* fix cpplint
* minor fixes
* fix python lint
* Fix build and tests
* fix builds
* fix windows build
* fix win build
* minor fix
* Refine based on comments
* resolve comments
* formatting
* resolve comments
* add ut
* Implement BitmaskDropout and associated unit tests.
* Implement BitmaskDropoutGrad and associated unit tests.
* Implement Dropout -> BitmaskDropout rewrite rule and associated unit tests.
* Implement (Dropout,DropoutGrad) -> (BitmaskDropout,BitmaskDropoutGrad) rewrite rule.
This commit does not yet include unit tests for this rewrite rule.
This commit also introduces improved documentation for all changes which will be grouped
into this PR.
* bitmask dropout
* fix win build
* bugfix for rocm
* bugfix
* fix code format
* fix ut
* fix build break
* fix ut in win
* resolve comments
* fix ut in trt
* resolve comments
* fix rocm build error
* fix typo
Co-authored-by: Aidan Beggs <aidanbeggs@microsoft.com>
* Fix torch cpp ext build when CPU wheel is installed but GPU card is present
Also there is a minor improvement for ATen operator that allows both
"::op" and "aten::op" name for operators
* Fix flake8 false positive
This includes a series of unit test that exercise
the MatMul fusion. This is not an exhaustive list
of tests. The tests focuse on paterns seen in
in models, with additional tests to cover at least
one instance of each operator type that can be part
of the fusion.
Signed-off-by: George Nash <george.nash@intel.com>
* [UPDATE] update amd ci pipeline 2 rocm5.1.1
* [FIX] json format error
* [ERROR] disable unit tests
* [FIX] ucx error
* [FIX] cmake version
* [FIX] units test
* add so_folder option to TVM EP options. add TvmSoEP class and update TVM EP factory
* compilation from so_folder was implemented
* update TVMCompiler for default pipeline and compilation from shared lib
* filter excess so-file in so_folder
* clean Compile method and vm conditions
* implementation of TVMSoCompile on native side instead of python API
* cpplint fixes
* some fixes after review
* more cpplint fixes
* more fixes after review
* align TVMso EP with new API for compilation from #10632
* small fixes for cpplint
Co-authored-by: Valery Chernov <valery.chernov@deelvin.com>
