⚡️ Speed up method LabelStudioAnnotation.to_dict by 3,362%

[codeflash-ai]

📄 3,362% (33.62x) speedup for LabelStudioAnnotation.to_dict in unstructured/staging/label_studio.py

⏱️ Runtime : 13.3 milliseconds → 385 microseconds (best of 135 runs)

📝 Explanation and details

The optimization achieves a dramatic 3361% speedup by eliminating expensive deep copy operations and using more efficient data structures.

Key optimizations:

1. Replaced deepcopy(self.__dict__) with dict(self.__dict__): The original code used deepcopy twice - once to create the initial dictionary and again to create _annotation_dict. The profiler shows these deepcopy calls consumed 97.8% of the total runtime (65.1% + 32.7%). The optimization uses a shallow copy instead, which is orders of magnitude faster since we only need the top-level dictionary structure.

2. Dictionary comprehension instead of pop operations: The final filtering step was replaced from a loop with .pop() calls to a single dictionary comprehension {k: v for k, v in annotation_dict.items() if v is not None}. This eliminates the need for the second deepcopy entirely and is more efficient than iterating and mutating the dictionary.

Why this works safely: The shallow copy is sufficient because the code immediately replaces the nested objects (result and reviews) with new lists created by calling .to_dict() on their elements. This means the original nested structure isn't shared, maintaining the same isolation behavior as the deep copy.

Performance impact: The profiler shows the time dropped from 87.9ms to 1.9ms for the core function logic, with most time now spent on the actual work (converting results/reviews) rather than unnecessary copying. This optimization is particularly beneficial for objects with many fields or nested structures, making it ideal for data serialization workflows where to_dict() may be called frequently.

✅ Correctness verification report:

Test	Status
⚙️ Existing Unit Tests	✅ 77 Passed
🌀 Generated Regression Tests	🔘 None Found
⏪ Replay Tests	🔘 None Found
🔎 Concolic Coverage Tests	🔘 None Found
📊 Tests Coverage	88.9%

⚙️ Existing Unit Tests and Runtime

Test File::Test Function	Original ⏱️	Optimized ⏱️	Speedup
staging/test_label_studio.py::test_created_annotation	25.4μs	1.62μs	1462%✅

To edit these changes git checkout codeflash/optimize-LabelStudioAnnotation.to_dict-mjcbd4c4 and push.