NXP backend: added aten.slice support

[novak-vaclav]

Summary

adds support for aten.slice operator

Test plan

tests can be manually run using pytest -c /dev/null backends/nxp/tests/

cc @robert-kalmar @JakeStevens @digantdesai

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15889

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (1 Unrelated Failure)

As of commit aaeddc1 with merge base 670bc11 ():

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

• pull / android / run-emulator (gh) (#16137)
  Timeout waiting for emulator to boot.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[novak-vaclav]

@pytorchbot label "release notes: nxp"

[Copilot]

Pull request overview

This PR adds support for the aten.slice operator to the NXP backend, enabling tensor slicing operations to be delegated to the Neutron hardware accelerator.

Key changes:

• Implements SliceTensorConverter to convert PyTorch slice operations to TFLite slice operations with dimension transpositions when needed
• Adds SliceTensorPattern quantization pattern for the slice operator
• Registers the converter in both partitioner and edge program converter
• Includes comprehensive test coverage for various slicing scenarios

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
backends/nxp/tests/models.py	Adds SliceTensorModule and SliceTensorConvModule test models
backends/nxp/tests/ir/converter/node_converter/test_slice_tensor_converter.py	Comprehensive test suite for slice tensor conversion with various configurations
backends/nxp/quantizer/patterns.py	Defines SliceTensorPattern quantization pattern
backends/nxp/quantizer/neutron_quantizer.py	Registers the slice tensor pattern with the quantizer
backends/nxp/neutron_partitioner.py	Maps slice_copy.Tensor operation to SliceTensorConverter
backends/nxp/backend/neutron_converter_manager.py	Excludes HoistSliceAboveTranspose graph pass
backends/nxp/backend/ir/converter/node_converters/ops_converters/slice_tensor_converter.py	Core implementation of slice tensor conversion logic
backends/nxp/backend/ir/converter/node_converters/ops_converters/__init__.py	Exports SliceTensorConverter
backends/nxp/backend/edge_program_converter.py	Registers slice converter in edge program converter

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[novak-vaclav]

Thank you for the review, fixed found issues.
Please check the changes @MartinPavella

[Copilot]

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 2 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[novak-vaclav]

Fixed issues from second round of code review. Thank you for the feedback.

[Copilot]

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated no new comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 2 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[MartinPavella]

LGTM.
Once the tests start working (and pass), we can merge.

[novak-vaclav]

Latest tests passed 🥳

[MartinPavella]

Latest tests passed 🥳

Only 3 checks were run. Meta has currently disabled their CI due to some firewall changes. We will have to wait for them to fix it.

In the meantime, feel free to keep this branch up-to-date with main.

[MartinPavella]

Please update the branch to enable the tests.

[novak-vaclav]

@pytorchbot label "module: nxp"

[Copilot]

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 1 comment.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The variable name sliced_tensor_rank is misleading. It stores the size of the dimension being sliced (input_shape[dim]), not the rank of the tensor. Consider renaming to sliced_dim_size or dim_size for clarity.

Suggested change

	sliced_tensor_rank = input_shape[dim]
	end = int(np.clip(end, 0, sliced_tensor_rank))
	start = int(np.clip(start, 0, sliced_tensor_rank))
	sliced_dim_size = input_shape[dim]
	end = int(np.clip(end, 0, sliced_dim_size))
	start = int(np.clip(start, 0, sliced_dim_size))

[MartinPavella]

We will need to get back to this once Neutron 25_12 is released. But for now, we can merge.