GitLab

New compiler pass that simplifies dynamic shapes related operators to their static versions if possible
Will normally be used after a split_single_dyn_dim pass

* 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>

Implements a fill operator that sets the values in an output buffer to a given value
Will be used when parsing ONNX ConstantOfShape
Can also be used when a buffer needs to be filled with a value that is determined at runtime

New op that populates a shape with random numbers with a uniform distribution. The rand_uniform op. can implement the Onnx RandomUniform instruction, and can also create the random number sequence necessary to implement Multinomial. (At this time, our Onnx Multinomial parsing generates a random sequence of numbers when parsing as a workaround, so that the resulting program uses the same "random" set every time.)

Arguments: shape, seed. Shape is required; can be static or dynamic. Seed is still optional in this version. If it's not given at inference time, use the value in the creation attribute seed. Update: deleted A boolean use_auto_seed causes any given seed to be ignored.

Makes a version of allocate that takes in dimensions and allocates a buffer
Going to create a simplify_dynamic_ops compiler pass that will use the use_shape_attr flag
The ONNX op ConstantOfShape needs the buffer to be filled with a specific value, so going to make another PR for that and a fill operator

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

* use implicit deps for the sorting
* use BFS for sorting program

* Use shape of Instruction (instead of a default) in add_return()

* Instruction validation fix: not to use a default shape value for comparison

* Fix instruction::replace() to recompute shape for "@return"

* handle the case of missing shape in an Instruction related Test

* use compute_shape() to get op shapes + test case for tuple_type

* add test case shape_test/return_shape_tuple

* Add test for @return to check for half type

* Move @return unit-tests around..; Address review comments

* Broken comparison fix: comparison to a (default) shape of tuple_type

* Test cases: (add) return_shape_empty & (modify) return_shape_tuple

* modify the assert() statement

* Improve performance of pointwise/reduction kernels when using NHWC layouts

* Format

* Add nhwc test

* Format

* Remove inline namespace

* Add reduce test

The --py output can be loaded back in the driver. This will embed the migraphx interperter so we can execute the python directly. There is a migraphx_py library which will dynamically load the version of the library for python version is available on the system.

* Add dynamic input to prefix_scan_op

* Added a shape test.  The op should return the same dynamic shape as the input.

* add 2d shape test for prefix_scan

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

This adds a format.py file which will check or apply formatting using git clang-format or yapf.

Fixes #1957

Clamping was removed in #1853.

Turns out clamping as necessary to handle overflow/underflow cases. during downcasting, if it overflowed then without clamping it returned infinity.

* add multitarget evaluation
* Add evaluation for the multi-target unit-tests

throw on use_len with non-fixed dynamic dimensions
change normalize_attributes to use input shape rather than input dimensions

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.

Changes to the way Pooling operation calculates pooling when there's padding. Old code would clip off any padding values before computing; for instance if an Average pooling window contained 0 1 2 where the 0 is padding, the result was 1.5 instead of 1.0. See Issue 1766

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 is to help enable multi-target execution. We store a vector of targets and contexts. Currently this will only compile a single target, the PR #1672 is needed to enable multiple targets.

This will also serialize the targets and contexts.

When using the execution_environment or prog.get_context() it will always use the context from the first target assuming this is the "primary" target. Although, its unlikely a user would use execution_environment with a multi-target environment.

Fixes the failing test case in #1815. Added a test that would otherwise fail without the change.

Needed to run multi-targeted program where "main" isn't the only root module. There could be many root modules other than main.

Updates ref version of clip to work with dynamic shapes
Encountered in agentmodel