Added Murty-based k-best partial assignment

[FlorianPfaff]

No description provided.

[github-actions]

✅MegaLinter analysis: Success

Descriptor	Linter	Files	Fixed	Errors	Warnings	Elapsed time
✅ COPYPASTE	jscpd	yes		no	no	12.64s
✅ JSON	prettier	2	0	0	0	0.49s
✅ JSON	v8r	2		0	0	2.56s
✅ MARKDOWN	markdownlint	1	0	0	0	0.66s
✅ MARKDOWN	markdown-table-formatter	1	0	0	0	0.27s
✅ PYTHON	bandit	334		0	0	4.96s
✅ PYTHON	black	334	10	0	0	8.46s
✅ PYTHON	flake8	334		0	0	2.77s
✅ PYTHON	isort	334	12	0	0	0.67s
✅ PYTHON	mypy	334		0	0	4.84s
✅ PYTHON	pylint	334		0	0	105.1s
✅ PYTHON	ruff	334	12	0	0	0.07s
✅ REPOSITORY	checkov	yes		no	no	22.99s
✅ REPOSITORY	gitleaks	yes		no	no	6.16s
✅ REPOSITORY	git_diff	yes		no	no	0.05s
✅ REPOSITORY	secretlint	yes		no	no	6.32s
✅ REPOSITORY	syft	yes		no	no	2.98s
✅ REPOSITORY	trivy-sbom	yes		no	no	2.85s
✅ REPOSITORY	trufflehog	yes		no	no	18.87s
✅ YAML	prettier	4	0	0	0	0.47s
✅ YAML	v8r	4		0	0	5.08s
✅ YAML	yamllint	4		0	0	0.4s

See detailed reports in MegaLinter artifacts

Your project could benefit from a custom flavor, which would allow you to run only the linters you need, and thus improve runtime performances. (Skip this info by defining FLAVOR_SUGGESTIONS: false)

• Documentation: Custom Flavors
• Command: npx mega-linter-runner@9.4.0 --custom-flavor-setup --custom-flavor-linters PYTHON_PYLINT,PYTHON_BLACK,PYTHON_FLAKE8,PYTHON_ISORT,PYTHON_BANDIT,PYTHON_MYPY,PYTHON_RUFF,COPYPASTE_JSCPD,JSON_V8R,JSON_PRETTIER,MARKDOWN_MARKDOWNLINT,MARKDOWN_MARKDOWN_TABLE_FORMATTER,REPOSITORY_CHECKOV,REPOSITORY_GIT_DIFF,REPOSITORY_GITLEAKS,REPOSITORY_SECRETLINT,REPOSITORY_SYFT,REPOSITORY_TRIVY_SBOM,REPOSITORY_TRUFFLEHOG,YAML_PRETTIER,YAML_YAMLLINT,YAML_V8R

Show us your support by starring ⭐ the repository

[github-actions]

Test Results

    12 files  ±  0     12 suites  ±0   4h 26m 33s ⏱️ - 21m 36s
   990 tests + 45    990 ✅ + 45      0 💤 ±  0  0 ❌ ±0 
12 164 runs  +540  9 416 ✅ +332  2 748 💤 +208  0 ❌ ±0 

Results for commit 7e934b6. ± Comparison against base commit b4815e8.

♻️ This comment has been updated with latest results.

[FlorianPfaff]

@copilot Suppress the too-many-locals and fix the others

pyrecest/utils/assignment.py:207: error: Need type annotation for "solution_heap" (hint: "solution_heap: list[] = ...") [var-annotated]
pyrecest/utils/assignment.py:215: error: Need type annotation for "ranked_solutions" (hint: "ranked_solutions: list[] = ...") [var-annotated]
Found 2 errors in 1 file (checked 334 source files)
❌ PYTHON / pylint - 5 errors
************* Module pyrecest.filters.multi_hypothesis_tracker
pyrecest/filters/multi_hypothesis_tracker.py:474:4: R0914: Too many local variables (16/15) (too-many-locals)
************* Module pyrecest.tests.filters.test_multi_hypothesis_tracker
pyrecest/tests/filters/test_multi_hypothesis_tracker.py:64:17: W0212: Access to a protected member _enumerate_candidate_assignments of a client class (protected-access)
pyrecest/tests/filters/test_multi_hypothesis_tracker.py:79:17: W0212: Access to a protected member _enumerate_candidate_assignments of a client class (protected-access)
************* Module pyrecest.utils.assignment
pyrecest/utils/assignment.py:66:0: R0914: Too many local variables (19/15) (too-many-locals)
pyrecest/utils/assignment.py:124:0: R0914: Too many local variables (19/15) (too-many-locals)

---

[FlorianPfaff]

@copilot Do not use numpy (except numpy.testing), only use pyrecest.backend

[Copilot]

@copilot Suppress the too-many-locals and fix the others

pyrecest/utils/assignment.py:207: error: Need type annotation for "solution_heap" (hint: "solution_heap: list[] = ...") [var-annotate...

Fixed in 6c69d17:

• Added # pylint: disable=too-many-locals to murty_k_best_assignments
• Added type annotation solution_heap: list[tuple[float, int, _MurtySubproblem, dict]]
• Added type annotation ranked_solutions: list[dict]

[FlorianPfaff]

@copilot Ignore too-many-locals, fix the rest

************* Module pyrecest.filters.multi_hypothesis_tracker
pyrecest/filters/multi_hypothesis_tracker.py:474:4: R0914: Too many local variables (16/15) (too-many-locals)
************* Module pyrecest.tests.filters.test_multi_hypothesis_tracker
pyrecest/tests/filters/test_multi_hypothesis_tracker.py:64:17: W0212: Access to a protected member _enumerate_candidate_assignments of a client class (protected-access)
pyrecest/tests/filters/test_multi_hypothesis_tracker.py:79:17: W0212: Access to a protected member _enumerate_candidate_assignments of a client class (protected-access)
************* Module pyrecest.utils.assignment
pyrecest/utils/assignment.py:66:0: R0914: Too many local variables (19/15) (too-many-locals)

---

[Copilot]

@copilot Do not use numpy (except numpy.testing), only use pyrecest.backend

Done in bbc3b52. Replaced import numpy as _np with individual imports from pyrecest.backend (abs, any, asarray, concatenate, full, isfinite, sum, where, zeros) throughout assignment.py.

@copilot Ignore too-many-locals, fix the rest

************* Module pyrecest.filters.multi_hypothesis_tracker
...

Fixed in bbc3b52:

• Added # pylint: disable=too-many-locals to _solve_subproblem in assignment.py and _enumerate_candidate_assignments in multi_hypothesis_tracker.py
• Added inline # pylint: disable=protected-access on the two test lines accessing _enumerate_candidate_assignments

[FlorianPfaff]

@copilot Make the respectdive functions fail for jax backend and skip test cases with skipif (as in other files).

=========================== short test summary info ============================
FAILED pyrecest/tests/filters/test_multi_hypothesis_tracker.py::MultiHypothesisTrackerEnumerationTest::test_enumeration_matches_bruteforce - TypeError: JAX arrays are immutable and do not support in-place item assignment. Instead of x[idx] = y, use x = x.at[idx].set(y) or another .at[] method: https://docs.jax.dev/en/latest/_autosummary/jax.numpy.ndarray.at.html
FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_empty_column_case_returns_all_rows_unassigned - TypeError: JAX arrays are immutable and do not support in-place item assignment. Instead of x[idx] = y, use x = x.at[idx].set(y) or another .at[] method: https://docs.jax.dev/en/latest/_autosummary/jax.numpy.ndarray.at.html
FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_matches_bruteforce_for_small_random_problems - TypeError: JAX arrays are immutable and do not support in-place item assignment. Instead of x[idx] = y, use x = x.at[idx].set(y) or another .at[] method: https://docs.jax.dev/en/latest/_autosummary/jax.numpy.ndarray.at.html
= 3 failed, 475 passed, 472 skipped, 15 warnings, 9 subtests passed in 981.00s (0:16:20) =

[Copilot]

@copilot Make the respectdive functions fail for jax backend and skip test cases with skipif (as in other files).

=========================== short test summary info ============================
...

Done in 1419db9. murty_k_best_assignments now raises NotImplementedError when the JAX backend is active, and the three affected tests in test_assignment.py and test_multi_hypothesis_tracker.py are decorated with @unittest.skipIf(pyrecest.backend.__backend_name__ == "jax", ...) following the pattern used in other test files.

[FlorianPfaff]

@copilot Use from copy import copy

=========================== short test summary info ============================
FAILED pyrecest/tests/filters/test_multi_hypothesis_tracker.py::MultiHypothesisTrackerEnumerationTest::test_enumeration_matches_bruteforce - AttributeError: 'Tensor' object has no attribute 'copy'
FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_empty_column_case_returns_all_rows_unassigned - AttributeError: 'Tensor' object has no attribute 'copy'
FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_matches_bruteforce_for_small_random_problems - AttributeError: 'Tensor' object has no attribute 'copy'
= 3 failed, 804 passed, 177 skipped, 5585436 warnings, 18 subtests

[Copilot]

@copilot Use from copy import copy

=========================== short test summary info ============================
...

Fixed in c054500. Added from copy import copy and replaced both x.copy() calls (cost_matrix.copy() and augmented_cost_matrix.copy()) with copy(x) in assignment.py.

[FlorianPfaff]

@copilot The error below only happens for the pytorch interface. Can you fix it?

=========================== short test summary info ============================
FAILED pyrecest/tests/filters/test_multi_hypothesis_tracker.py::MultiHypothesisTrackerEnumerationTest::test_enumeration_matches_bruteforce - TypeError: full() received an invalid combination of arguments - got (int, int, dtype=type), but expected one of:

• (tuple of ints size, Number fill_value, *, tuple of names names, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
• (tuple of ints size, Number fill_value, *, Tensor out = None, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
  FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_empty_column_case_returns_all_rows_unassigned - TypeError: full() received an invalid combination of arguments - got (int, int, dtype=type), but expected one of:
• (tuple of ints size, Number fill_value, *, tuple of names names, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
• (tuple of ints size, Number fill_value, *, Tensor out = None, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
  FAILED pyrecest/tests/utils/test_assignment.py::MurtyAssignmentTest::test_matches_bruteforce_for_small_random_problems - TypeError: full() received an invalid combination of arguments - got (int, int, dtype=type), but expected one of:
• (tuple of ints size, Number fill_value, *, tuple of names names, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
• (tuple of ints size, Number fill_value, *, Tensor out = None, torch.dtype dtype = None, torch.layout layout = None, torch.device device = None, bool pin_memory = False, bool requires_grad = False)
  = 3 failed, 808 passed, 179 skipped, 5585436 warnings, 18 subtests passed in 2027.54s (0:33:47) =

[Copilot]

@copilot The error below only happens for the pytorch interface. Can you fix it?

=========================== short test summary info ============================
...

Fixed in 7e934b6. Three PyTorch-specific issues were addressed in assignment.py:

1. _full(n_rows, -1, dtype=int) → _full((n_rows,), -1, dtype=_int64) — PyTorch requires a tuple for the shape argument and a proper dtype (imported int64 from pyrecest.backend).
2. .astype(int) → _asarray(..., dtype=_int64) — cross-backend int casting.
3. copy(matrix) → _array(matrix) — copy.copy() creates a shallow alias for PyTorch tensors (shared storage), causing the original augmented_cost_matrix to be corrupted when modified_cost_matrix was written to. Using array() from pyrecest.backend creates a true independent copy on both numpy and PyTorch.