Refactor: Extract duplicated code from test and infer verbs into shared utilities

[Copilot]

Change Description

Two verb modules (test.py and infer.py) contained significant code duplication for model weight loading and batch result saving.

Solution Description

Extracted shared logic into reusable utilities:

• load_model_weights() → models/model_utils.py
  Consolidates ~96 lines of duplicated code from test.py and infer.py. Handles explicit path or auto-detection from recent training run. Parameterized by verb name.

• create_save_batch_callback() → pytorch_ignite.py
  Consolidates ~84 lines of duplicated code from test.py and infer.py. Factory function returns closure that saves model outputs batch-by-batch during inference/testing. Includes validation and error handling to ensure batches contain required object_id field, with clear error messages to guide dataset implementation.

Scope: Only test.py and infer.py were refactored. The train.py verb remains unchanged and retains its own _log_params method. Similarly, test.py has its own _log_params method (not shared with train).

Impact: Reduced code duplication in test and infer verbs while maintaining clear separation of concerns between verbs. Removed unused variables (batch_len, write_index) from the callback implementation for cleaner code.

Breaking changes: None. All verb interfaces unchanged. The callback now requires datasets to provide object_id in batch dictionaries, which was already the expected behavior in the original implementation.

Code Quality

• I have read the Contribution Guide and agree to the Code of Conduct
• My code follows the code style of this project
• My code builds (or compiles) cleanly without any errors or warnings
• My code contains relevant comments and necessary documentation

Original prompt

@copilot, will you take a moment to address this refactor?

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[codecov]

Codecov Report

❌ Patch coverage is 82.45614% with 10 lines in your changes missing coverage. Please review.
✅ Project coverage is 62.96%. Comparing base (808dbfa) to head (d94afb1).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/hyrax/models/model_utils.py	76.00%	6 Missing ⚠️
src/hyrax/pytorch_ignite.py	73.33%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #646      +/-   ##
==========================================
+ Coverage   62.95%   62.96%   +0.01%     
==========================================
  Files          58       59       +1     
  Lines        5782     5730      -52     
==========================================
- Hits         3640     3608      -32     
+ Misses       2142     2122      -20     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Before [808dbfa]	After [3b5ebea]	Ratio	Benchmark (Parameter)
3.74G	3.88G	1.04	vector_db_benchmarks.VectorDBInsertBenchmarks.peakmem_load_vector_db(16384, 'qdrant')
1.91±0.01s	1.96±0.03s	1.03	benchmarks.time_rebuild_manifest_help
16.8962717254681	17.46608516124805	1.03	data_cache_benchmarks.DataCacheBenchmarks.track_cache_hsc1k_hyrax_size_undercount
113±1μs	116±2μs	1.03	data_request_benchmarks.DatasetRequestBenchmarks.time_request_all_data
9.17±0.1ms	9.34±0.1ms	1.02	vector_db_benchmarks.VectorDBSearchBenchmarks.time_search_by_vector_many_shards(128, 'chromadb')
1.92±0.02s	1.94±0.02s	1.01	benchmarks.time_lookup_help
36.7±0.4ms	37.0±0.1ms	1.01	benchmarks.time_nb_obj_construct
1.91±0.01s	1.94±0.03s	1.01	benchmarks.time_prepare_help
1.91±0.01s	1.93±0.01s	1.01	benchmarks.time_train_help
1.27G	1.29G	1.01	vector_db_benchmarks.VectorDBInsertBenchmarks.peakmem_load_vector_db(2048, 'qdrant')

Click here to view all benchmarks.

[drewoldag]

@copilot, please address the two comments where you are explicitly called out.

[Copilot]

Pull request overview

This PR refactors duplicated code from the test.py and infer.py verbs by extracting shared functionality into reusable utilities. The refactoring consolidates model weight loading and batch result saving logic, reducing code duplication and improving maintainability.

Changes:

• Extracted load_model_weights() function to models/model_utils.py for shared model weight loading logic
• Extracted create_save_batch_callback() factory function to pytorch_ignite.py for creating batch result saving callbacks
• Updated test.py and infer.py to use the new shared utilities, removing ~180 lines of duplicated code
• Updated test imports and function calls to reference the new utility functions

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
src/hyrax/models/model_utils.py	New file containing load_model_weights() utility function that consolidates model weight loading logic from test and infer verbs
src/hyrax/pytorch_ignite.py	Added create_save_batch_callback() factory function that creates closures for saving batch results during inference/testing
src/hyrax/verbs/test.py	Removed load_model_weights() static method and _save_batch nested function, replaced with calls to new shared utilities
src/hyrax/verbs/infer.py	Removed load_model_weights() static method and _save_batch nested function, replaced with calls to new shared utilities
tests/hyrax/test_infer.py	Updated test imports to reference new load_model_weights() location in models/model_utils.py

[Copilot]

The data_loader_indexes parameter is documented and accepted but never used in the function body. The old implementations in test.py and infer.py used this parameter as a fallback when batch didn't contain object_id. Consider either removing this parameter if it's truly unnecessary, or documenting why it's passed but unused.

[mtauraso]

This looks good to me

[drewoldag]

@copilot please address PR comments.

[drewoldag]

Looking good now