⚡️ Speed up function merge_out_layout_with_ocr_layout by 30%

CHANGELOG.md

@@ -1,3 +1,8 @@
+## 0.18.32-dev0
+
+### Enhancements
+- Optimized `merge_out_layout_with_ocr_layout` (codeflash)
+
 ## 0.18.31
 
 ### Enhancements

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.18.31"  # pragma: no cover
+__version__ = "0.18.32-dev0"  # pragma: no cover

unstructured/partition/pdf_image/ocr.py

@@ -471,13 +471,9 @@ def supplement_layout_with_ocr_elements(
         else:
             ocr_regions_to_add = ocr_layout
     else:
-        mask = (
-            ~bboxes1_is_almost_subregion_of_bboxes2(
-                ocr_layout.element_coords, layout.element_coords, subregion_threshold
-            )
-            .sum(axis=1)
-            .astype(bool)
-        )
+        mask = ~bboxes1_is_almost_subregion_of_bboxes2(
+            ocr_layout.element_coords, layout.element_coords, subregion_threshold
+        ).any(axis=1)
 
         # add ocr regions that are not covered by layout
         ocr_regions_to_add = ocr_layout.slice(mask)

unstructured/partition/pdf_image/pdfminer_processing.py

@@ -812,31 +812,28 @@ def aggregate_embedded_text_by_block(
     if len(source_regions) == 0 or len(target_region) == 0:
         return "", None
 
-    mask = (
-        bboxes1_is_almost_subregion_of_bboxes2(
-            source_regions.element_coords,
-            target_region.element_coords,
-            subregion_threshold,
-        )
-        .sum(axis=1)
-        .astype(bool)
-    )
-
-    text = " ".join([text for text in source_regions.slice(mask).texts if text])
-
-    if sum(mask):
-        source_bboxes = source_regions.slice(mask).element_coords
+    mask = bboxes1_is_almost_subregion_of_bboxes2(
+        source_regions.element_coords,
+        target_region.element_coords,
+        subregion_threshold,
+    ).any(axis=1)
+
+    if mask.any():
+        sliced = source_regions.slice(mask)
+        text = " ".join([text for text in sliced.texts if text])
+        source_bboxes = sliced.element_coords
         target_bboxes = target_region.element_coords
 
         iou = _aggregated_iou(source_bboxes, target_bboxes[0, :])
 
         fully_filled = (
-            all(flag == IsExtracted.TRUE for flag in source_regions.slice(mask).is_extracted_array)
+            all(flag == IsExtracted.TRUE for flag in sliced.is_extracted_array)
             and iou > text_coverage_threshold
         )
         is_extracted = IsExtracted.TRUE if fully_filled else IsExtracted.PARTIAL
     else:
         # if nothing is sliced then it is not extracted
+        text = ""
         is_extracted = IsExtracted.FALSE
     return text, is_extracted