GitLab

Revert "Handle broadcasts across dot and concat (#1689) (#1731)"

* Convert Fp16 instance-norm to FP32 temporarily (#1779)
* Conditionally enable GeLU approximation  (#1810)

This reverts commit a46f378e.

Use std::numeric_limits::min/max() functions plus the appropriate value to encode -inf/inf
Co-authored-by: Ted Themistokleous <107195283+TedThemistokleous@users.noreply.github.com>

* Use action to free space which uses apt  remove to remove all the dependencies as well (#1756)
* Docsupdate (#1748)
* adjust docker files to support new rocm 5.5 (#1729)
* update to v0.11.0 of rocm-docs-core (#1763)

* Removed split_single_dyn_dim compile flag (#1711)
* Update C/C++ API for dynamic batch (#1712)
* Python API update for dynamic batch (#1723)
* Dynamic batch C++ API example #1728
* Optimize file space of github runners (#1743)
Co-authored-by: Charlie Lin <charlie.lin@amd.com>

Improves the constant propagation for bert models. Larger batch size no longer use as large of constants.  Also improves the speed of model compilation
Co-authored-by: Paul Fultz II <pfultz2@yahoo.com>

update rocBLAS version check to support 3.0 and above with simplified logic

Bumps [tensorflow](https://github.com/tensorflow/tensorflow) from 2.9.3 to 2.11.1.
- [Release notes](https://github.com/tensorflow/tensorflow/releases)
- [Changelog](https://github.com/tensorflow/tensorflow/blob/master/RELEASE.md)
- [Commits](https://github.com/tensorflow/tensorflow/compare/v2.9.3...v2.11.1)

---
updated-dependencies:
- dependency-name: tensorflow
  dependency-type: direct:production
...

Updates the hip::copy_to_gpu and hip::copy_from_gpu operators to work with dynamic shapes

Allows for offload_copy to be used with dynamic batch

Changed assert in select_module because the argument might now be smaller with how offload_copy will work with dynamic batch. (maximum buffer size will be used)

This fixes #1700

Solves #1311

Expose get_shape and get_operator methods for instruction_ref object in the python API.

* Use hash for docker layer
* Remove `layer-` prefix. it gets added by action automatically
* Add requirements file to docker key hash

This will show whats being replaced with a constant. This is useful for debugging where a literal comes from.

Ensure that we don't have empty inputs when computing shape for pointwise function

CI changes to improve github cache management 

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

Expose the shape::type_t values to be used by the python api and is required by torch_migraphx to support torchbench models.

This removes the --cxx flags from the rbuild commands since it is not necessary. Also added a section about using rbuild to set up an environment for development.

Adds a matcher to split_single_dyn_dim to find all broadcast or multibroadcast with two static shape inputs and replaces the instruction with the one input version.
Sorts the get_output_parameters() list to ensure the correct ordering. (Was getting an error for some models.)

* Enable hiprtc by default

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.

Examples..

bin/driver verify /codes/onnx_models/resnet50-v1-7/resnet50-v1-7.onnx --split-single-dyn-dim --batch 3 --dyn-input-dim @data "[{min:1, max:4}, 3, 224, 224]"

bin/driver compile /codes/onnx_models/resnet50-v1-7/resnet50-v1-7.onnx --split-single-dyn-dim --default-dyn-dim "{min:1, max:10}" --output resnet50_batch1-10.mxr

bin/driver perf resnet50_batch1-10.mxr --batch 4

Automatically fuse multiple reductions and pointwise operations.

* Add MIGRAPHX_VALIDATE_MATCHES env variable to validate each matcher

This will replace conv(x+a, w) with conv(x, w) + conv(a, w) where a is a constant so conv(a, w) can be replaced with a constant.