[QNN EP] Update QNN SDK to version 2.17.0 #18684
adrianlizarraga
merged
26 commits into
main
from
adrianl/qnn-sdk-2.17
Start adding 2 tolerance variables for QDQ accuracy tests.
5eb55333
Increase tolerance for tests failing on QNN SDK 2.17. Print warning f…
71335225
Enable test fixed in QNN SDK 2.17: LayerNorm1D_LastAxis_DynamicScale
91263734
Enable test fixed in QNN SDK 2.17: GlobalMaxPool_LargeInput2_u8
3b739c9a
Add 16-bit QDQ tests for MaxPool
ab5069d5
Enable Softmax rank 5 tests that were fixed in QNN SDK 2.17
deff93ef
Enable Where op tests (large data) that were fixed in QNN SDK 2.17
e75c6479
Add unit test for LayerNorm A16,W8
f4b6b157
Enable fixed CPU MaxPool tests
1506b226
Replace useless Reduce op tests with version that test reducing only …
31fba4e3
Remove OverrideValueRange() since it's no longer needed
df60e8e4
Enable disabled Resize op tests (CPU)
335570fe
Update QNN SDK version of CI pipelines
f8277020
adrianlizarraga
commented
on 2023-12-03
Fix Gemm bug. Disable some CPU EP tests that fail
4b775cb8
Reduce changes
e1909718
Undo enabling of ResizeOpTest.ResizeOpLinearDownSampleTest_4DBilinear
a9094938
adrianlizarraga
commented
on 2023-12-04
Disable some ArgMax/ArgMin CPU EP tests
a8d0fe19
Merge branch 'main' into adrianl/qnn-sdk-2.17
62d38882
Temporarily disable win x64 qnn unit tests
6e66f27d
Experiment with matmul test
53e810eb
Disable matmul test for linux cpu backend
fc657890
Disabled wrong test
49732ab4
adrianlizarraga
commented
on 2023-12-05
adrianlizarraga
commented
on 2023-12-05
adrianlizarraga
commented
on 2023-12-05
Cleanup
64f2a0a5
adrianlizarraga
commented
on 2023-12-05
Do not unload QnnCpu.dll on Windows x64
0e4eebf7
Run linter
c970b776
Disable failing ONNX model test on Windows x64, QnnCpu.dll, QNN SDK 2.17
6707ca49
jywu-msft
approved these changes
on 2023-12-06
HectorSVC
approved these changes
on 2023-12-06
Login to write a write a comment.
Login via GitHub