Replace dm-tree with optree

[rostan-t]

Category:

Other (e.g. Documentation, Tests, Configuration)

Description:

dm-tree is not actively maintained and has known performance issues. optree is an alternative that provides a superset of the dm-tree features.

The dependency on dm-tree makes it harder to install DALI on Python 3.14 because it doesn't provide wheels and the source distribution doesn't compile. This is fixed by optree. Additionally, it supports free-threaded Python.

Additional information:

Affected modules and functionalities:

Conditionals, scalar promotion.

Key points relevant for the review:

Tests:

• Existing tests apply
• New tests added
  • Python tests
  • GTests
  • Benchmark
  • Other
• N/A

Checklist

Documentation

• Existing documentation applies
• Documentation updated
  • Docstring
  • Doxygen
  • RST
  • Jupyter
  • Other
• N/A

DALI team only

Requirements

• Implements new requirements
• Affects existing requirements
• N/A

REQ IDs: N/A

JIRA TASK: N/A

[greptile-apps]

Greptile Summary

This PR migrates from dm-tree to optree for tree structure manipulation. The migration is motivated by dm-tree being unmaintained and lacking Python 3.14 support, while optree provides a superset of features plus free-threaded Python support.

Key changes:

• Replaced tree.map_structure() with optree.tree_map() across conditionals and operator preprocessing
• Replaced tree.assert_same_structure() with manual structure comparison using optree.tree_structure() equality checks
• Updated all dependency files (setup.py, conda recipe, Dockerfile, test scripts)
• Added explicit numpy dependency in Dockerfile (was previously transitive through dm-tree)
• Updated test assertions to match optree's error message format

The changes are straightforward API replacements with equivalent functionality. All affected modules (conditionals, scalar promotion) maintain their original behavior.

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• The migration from dm-tree to optree is a clean, well-executed dependency replacement. All API mappings are correct (tree.map_structure → optree.tree_map, structure comparison logic properly reimplemented), dependency files comprehensively updated, and test expectations adjusted appropriately. The addition of explicit numpy dependency prevents transitive dependency issues. No logical changes to functionality.
• No files require special attention

Important Files Changed

Filename	Overview
dali/python/nvidia/dali/_conditionals.py	Replaced tree.map_structure with optree.tree_map and tree.assert_same_structure with manual structure comparison using optree.tree_structure
dali/python/nvidia/dali/ops/init.py	Updated import and replaced tree.map_structure with optree.tree_map for scalar constant promotion
dali/python/setup.py.in	Replaced dm-tree dependency with optree in Python package dependencies
docker/Dockerfile	Replaced dm-tree with optree and explicitly added numpy (previously a transitive dependency of dm-tree)

_{Last reviewed commit: 8aaeb81}

[greptile-apps]

_{7 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[rostan-t]

!build

[dali-automaton]

CI MESSAGE: [44703456]: BUILD FAILED

[dali-automaton]

CI MESSAGE: [44708749]: BUILD STARTED

[rostan-t]

!build

[greptile-apps]

_{7 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[dali-automaton]

CI MESSAGE: [44710400]: BUILD STARTED

[dali-automaton]

CI MESSAGE: [44708749]: BUILD FAILED

[dali-automaton]

CI MESSAGE: [44710400]: BUILD FAILED

[greptile-apps]

_{8 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[rostan-t]

!build

[dali-automaton]

CI MESSAGE: [44788092]: BUILD STARTED

[dali-automaton]

CI MESSAGE: [44788092]: BUILD PASSED