GitLab

Add support for the QLinearMatMul onnx operator

Makes ConstantOfShape work with a variable dimensions input

* Rectify flipped coordinate_transformation_mode logic in ROIAlign
* Handle both opset 10 and 16 versions
* Fix version check and clang tidy warning
Co-authored-by: Dino Musić <dino.music@htecgroup.com>

Add parsing support for value_float, value_floats, value_int, value_ints attributes
Disable failing tests
Resolves Test failures due to IndexError: _Map_base::at migraphx-benchmark/AMDMIGraphX#76

Reduce memory footprint by std::move of temporary (potentially very large) containers.
Minor cleanup for performance optimization: e.g. of Index() calculation -- which can get repeated millions of times in large tensors/vectors in a single Visit.

Adds support for dynamic input shape in pooling operator along with auto-padding. This combination requires that the padding (and therefore the output shape) can't be computed until runtime.

Allows slice to work with variable starts, ends, and axes input
Outputs a dynamic shape even with a static shape data input when the starts and ends are variable

Throwing on these calls catches dynamic shape errors earlier rather than having to backpedal from a bad call

Fixes #1746

BatchNorm only has x as the runtime input parameter for the following equation. All the other parameters are compile-time constants and related operations can be const-folded before quantizing to fp16 to preserve precision.

Renames deconvolution -> convolution_backwards to be more consistent with the literature
Note: this is not the cross-correlation operator (which is the adjoint of convolution). This is technically a standard convolution operator combined with an upsampling operator rather than a downsampling operator.
Adds unit tests for the padding, strides, dilations, and other op attributes.
Throws on auto_pad attribute since it has not been implemented
Previously it read the attribute and set it but then did nothing with it
Extended for dynamic shapes
Does not support using asymmetric padding (padding_L != padding_R) and output_shape with dynamic shapes.

Export API symbols for migraphx, migraphx_ref, migraphx_cpu, migrphx_gpu, migraphx_device, migraphx_tf, and migraphx_onnx. There is a separate PR for migrahx_c.

API symbol exporting affects only Windows. It is transparent on Linux.

This will show the instructions that are inserted for each onnx operator. This should help debugging where certain instructions come from.

Co-authored-by: kahmed10 <15948690+kahmed10@users.noreply.github.com>

* fix parse_instancenorm to create broadcast and multibroadcast instructions with two dynamic shape arguments instead of 1.  Their make_op() functions don't support dynamic shapes when called with one input.  This caused an error when parsing an ONNX 3duunet model

* Use add_common_op() to create multibroadcast op.

* add verification and parsing test for instance_norm with dynamic input.  Parse test doesn't pass.

* fix for test; still doesn't pass

* another fix for test; still doesn't pass

* work in progress, instance_norm_dyn_batch_test works but instance_norm_test doesn't

* fix onnx instancenorm tests to match parser changes.  Passes all check tests

* Updated comments explaining usage of add_common_op()

* hand-merged conflicts with develop

* fix instance_norm_half_test after merge

* add Onnx test instance_norm_dyn_batch_half_test

* add shape test cases broadcast_1in_dyn_error and multibroadcast_1in_dyn_error_0

Enable with MIGRAPHX_ENABLE_CK=1 and --exhaustive-tune tune flag

By converting to fp32 : fp16 3d-unet model accuracy comes out the same as FP32 accuracy.

By using reduce_sum method on Fp16 : accuracy comes out ~0.9% lower compared to fp32 while keeping entire model in fp16.

Move CI to support the rocm5.5 release

add option to print tf supported ops
sort both onnx and tf ops alphabetically

Fixes the above behavior
This needs to be changed to allow for setting static shapes with map_dyn_input_dims since you cannot also use map_input_dims

Converts can be inserted when the scales and input differ in the onnx file(we are already doing this implicit conversion in the ref implementation). This will also improve the compile-time of quantizelinear.hpp since we can remove the nested visit method.

Makes the optimals into a std::set<std::size_t>
Changes shape object functions to handle the opts change
Changes to convolution, flatten, pooling, and convolution in that they no longer calculate the output optimal dimensions. Instead returns empty opts. Will need to change this in the future if we want to support dynamic shapes fully.
Many changes to tests and shape calls with respect to the new optimals

Add dynamic shape support to slice operator.

First draft of this feature doesn't support ops slicing non-fixed, dynamic axes. Resulting shape in such cases is not guaranteed.* Also, onnx parsing doesn't support any arguments other than "axes".