GitLab

disable quant_dot test for CPU backend, it requires removing downcast converts in "ref"implementation. Inside the "ref" impelmentation, it downcasting one of the inputs of the GEMM from float to fp8. Which is a lossy conversion and while computing gemm, it upcasts it back to  float.  CPU backend removes converts completely since it is going from "float->fp8->float". due to lossy cast, results are coming out different.

Add min and max reduction modes
Implement duplicate index handling
Add corresponding reference, onnx, and verify tests.

Introduce dilations attribute to pooling operators reference implementation.

Since ONNX opset version 14, layout attribute has been introduced to LSTM operator which allows two predefined layouts for input and output shapes.  Add corresponding reference, onnx, and verify tests.

The inaccuracy was caused by ONNX round requires nearest integer rounding for halway (0.5) cases.
std::round rounds away from zero, thus giving wrong results with halfway cases.
Replaced std::round with std::nearbyint which uses the correct rounding by default.

Currently, we see MLIR partition candidates recieving non-standard shape due to not fusing in squeeze, flatten and unsqueeze ops. These ops could be canonicalized to reshape without introducing additional ops as long as MLIR backend is concerned.

* Fix scatter operator for nonstandard shapes

remove standard() shape check for scatter inputs.

* Add nostandard input tests for scatter

---------
Co-authored-by: Chris Austen <causten@users.noreply.github.com>

MIGraphX verification by default uses normalized RMS error as the basis for the verification.  This change adds some logic to allow migraphx to do "np.allclose" type of elementwise verification using atol and rtol.

Commit also includes changes to consistently pass "gold" or "expected" results as the second argument for "verify_range()" calls.  Default RMS tolerance inside driver is set to 0.001 which IMO is high for FP32 compared to what we had earlier. Need better defaults