deepdoctection 0.39.2__tar.gz → 0.39.4__tar.gz

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: deepdoctection
-Version: 0.39.2
+Version: 0.39.4
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -57,7 +57,7 @@ Requires-Dist: tensorpack==0.11; extra == "tf"
 Requires-Dist: protobuf==3.20.1; extra == "tf"
 Requires-Dist: tensorflow-addons>=0.17.1; extra == "tf"
 Requires-Dist: tf2onnx>=1.9.2; extra == "tf"
-Requires-Dist: python-doctr==0.8.1; extra == "tf"
+Requires-Dist: python-doctr==0.9.0; extra == "tf"
 Requires-Dist: pycocotools>=2.0.2; extra == "tf"
 Requires-Dist: boto3==1.34.102; extra == "tf"
 Requires-Dist: pdfplumber>=0.11.0; extra == "tf"
@@ -88,7 +88,7 @@ Requires-Dist: tqdm==4.64.0; extra == "pt"
 Requires-Dist: timm>=0.9.16; extra == "pt"
 Requires-Dist: transformers>=4.48.0; extra == "pt"
 Requires-Dist: accelerate>=0.29.1; extra == "pt"
-Requires-Dist: python-doctr==0.8.1; extra == "pt"
+Requires-Dist: python-doctr==0.9.0; extra == "pt"
 Requires-Dist: boto3==1.34.102; extra == "pt"
 Requires-Dist: pdfplumber>=0.11.0; extra == "pt"
 Requires-Dist: fasttext-wheel; extra == "pt"

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/__init__.py

@@ -25,7 +25,7 @@ from .utils.logger import LoggingRecord, logger
 
 # pylint: enable=wrong-import-position
 
-__version__ = "0.39.2"
+__version__ = "0.39.4"
 
 _IMPORT_STRUCTURE = {
     "analyzer": ["config_sanity_checks", "get_dd_analyzer", "ServiceFactory"],

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/datapoint/image.py

@@ -702,11 +702,11 @@ class Image:
         return get_uuid(self.image_id, *container_ids)
 
     def save(
-        self,
-        image_to_json: bool = True,
-        highest_hierarchy_only: bool = False,
-        path: Optional[PathLikeOrStr] = None,
-        dry: bool = False,
+            self,
+            image_to_json: bool = True,
+            highest_hierarchy_only: bool = False,
+            path: Optional[PathLikeOrStr] = None,
+            dry: bool = False,
     ) -> Optional[Union[ImageDict, str]]:
         """
         Export image as dictionary. As numpy array cannot be serialized `image` values will be converted into
@@ -719,6 +719,18 @@ class Image:
 
         :return: optional dict
         """
+
+        def set_image_keys_to_none(d):  # type: ignore
+            if isinstance(d, dict):
+                for key, value in d.items():
+                    if key == '_image':
+                        d[key] = None
+                    else:
+                        set_image_keys_to_none(value)
+            elif isinstance(d, list):
+                for item in d:
+                    set_image_keys_to_none(item)
+
         if path is None:
             path = Path(self.location)
         path = Path(path)
@@ -734,7 +746,7 @@ class Image:
         export_dict = self.as_dict()
         export_dict["location"] = fspath(export_dict["location"])
         if not image_to_json:
-            export_dict["_image"] = None
+            set_image_keys_to_none(export_dict)
         if dry:
             return export_dict
         with open(path_json, "w", encoding="UTF-8") as file:

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/datasets/registry.py

@@ -18,6 +18,7 @@
 """
 Module for DatasetRegistry
 """
+import inspect
 
 import catalogue  # type: ignore
 from tabulate import tabulate
@@ -47,7 +48,10 @@ def get_dataset(name: str) -> DatasetBase:
     :param name: A dataset name
     :return: An instance of a dataset
     """
-    return dataset_registry.get(name)()
+    ds = dataset_registry.get(name)
+    if inspect.isclass(ds):
+        return ds()
+    return ds
 
 
 def print_dataset_infos(add_license: bool = True, add_info: bool = True) -> None:

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/extern/doctrocr.py

@@ -79,8 +79,8 @@ def _load_model(
         state_dict = torch.load(os.fspath(path_weights), map_location=device)
         for key in list(state_dict.keys()):
             state_dict["model." + key] = state_dict.pop(key)
-        doctr_predictor.load_state_dict(state_dict)
-        doctr_predictor.to(device)
+        doctr_predictor.load_state_dict(state_dict)  # type: ignore
+        doctr_predictor.to(device)  # type: ignore
     elif lib == "TF":
         # Unzip the archive
         params_path = Path(path_weights).parent
@@ -88,9 +88,9 @@ def _load_model(
         if is_zip_path:
             with ZipFile(path_weights, "r") as file:
                 file.extractall(path=params_path)
-                doctr_predictor.model.load_weights(params_path / "weights")
+                doctr_predictor.model.load_weights(params_path / "weights")  # type: ignore
         else:
-            doctr_predictor.model.load_weights(os.fspath(path_weights))
+            doctr_predictor.model.load_weights(os.fspath(path_weights))  # type: ignore
 
 
 def auto_select_lib_for_doctr() -> Literal["PT", "TF"]:
@@ -125,7 +125,7 @@ def doctr_predict_text_lines(
         DetectionResult(
             box=box[:4].tolist(), class_id=1, score=box[4], absolute_coords=False, class_name=LayoutType.WORD
         )
-        for box in raw_output[0]["words"]
+        for box in raw_output[0]["words"]  # type: ignore
     ]
     return detection_results
 
@@ -480,7 +480,7 @@ class DoctrTextRecognizer(TextRecognizer):
         return auto_select_lib_for_doctr()
 
     def clear_model(self) -> None:
-        self.doctr_predictor = None
+        self.doctr_predictor = None  # type: ignore
 
 
 class DocTrRotationTransformer(ImageTransformer):
@@ -527,7 +527,8 @@ class DocTrRotationTransformer(ImageTransformer):
         return viz_handler.rotate_image(np_img, specification.angle)  # type: ignore
 
     def predict(self, np_img: PixelValues) -> DetectionResult:
-        angle = estimate_orientation(np_img, self.number_contours, self.ratio_threshold_for_lines)
+        angle = estimate_orientation(np_img, n_ct=self.number_contours,
+                                     ratio_threshold_for_lines=self.ratio_threshold_for_lines)
         if angle < 0:
             angle += 360
         return DetectionResult(angle=round(angle, 2))

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/mapper/cats.py

@@ -32,7 +32,7 @@ from .maputils import LabelSummarizer, curry
 @curry
 def cat_to_sub_cat(
     dp: Image,
-    categories_dict_names_as_key: dict[TypeOrStr, int],
+    categories_dict_names_as_key: Optional[dict[TypeOrStr, int]] = None,
     cat_to_sub_cat_dict: Optional[dict[TypeOrStr, TypeOrStr]] = None,
 ) -> Image:
     """
@@ -45,7 +45,8 @@ def cat_to_sub_cat(
     :param cat_to_sub_cat_dict: e.g. {'foo': 'sub_cat_1', 'bak': 'sub_cat_2'}
     :return: Image with updated Annotations
     """
-
+    if categories_dict_names_as_key is None:
+        categories_dict_names_as_key = {}
     if cat_to_sub_cat_dict is None:
         return dp
     cat_to_sub_cat_dict_obj_type = {get_type(key): get_type(value) for key, value in cat_to_sub_cat_dict.items()}
@@ -54,7 +55,7 @@ def cat_to_sub_cat(
         sub_cat = ann.get_sub_category(sub_cat_type)
         if sub_cat:
             ann.category_name = sub_cat.category_name
-            ann.category_id = categories_dict_names_as_key[ann.category_name]
+            ann.category_id = categories_dict_names_as_key.get(ann.category_name,DEFAULT_CATEGORY_ID)
 
     return dp
 

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection/pipe/segment.py

@@ -1190,14 +1190,16 @@ class PubtablesSegmentationService(PipelineComponent):
                             for key, value in cell_rn_cn_to_ann_id.items():
                                 if key[idx] == item_number:
                                     cell_ann = dp.get_annotation(annotation_ids=value)[0]
-                                    self.dp_manager.set_category_annotation(
-                                        item_header_cell_name, None, item_header_cell_name, cell_ann.annotation_id
-                                    )
+                                    if item_header_cell_name not in cell_ann.sub_categories:
+                                        self.dp_manager.set_category_annotation(
+                                            item_header_cell_name, None, item_header_cell_name, cell_ann.annotation_id
+                                        )
                                 else:
                                     cell_ann = dp.get_annotation(annotation_ids=value)[0]
-                                    self.dp_manager.set_category_annotation(
-                                        item_header_cell_name, None, CellType.BODY, cell_ann.annotation_id
-                                    )
+                                    if CellType.BODY not in cell_ann.sub_categories:
+                                        self.dp_manager.set_category_annotation(
+                                            item_header_cell_name, None, CellType.BODY, cell_ann.annotation_id
+                                        )
 
             # TODO: the summaries should be sub categories of the underlying ann
             self.dp_manager.set_summary_annotation(

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: deepdoctection
-Version: 0.39.2
+Version: 0.39.4
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -57,7 +57,7 @@ Requires-Dist: tensorpack==0.11; extra == "tf"
 Requires-Dist: protobuf==3.20.1; extra == "tf"
 Requires-Dist: tensorflow-addons>=0.17.1; extra == "tf"
 Requires-Dist: tf2onnx>=1.9.2; extra == "tf"
-Requires-Dist: python-doctr==0.8.1; extra == "tf"
+Requires-Dist: python-doctr==0.9.0; extra == "tf"
 Requires-Dist: pycocotools>=2.0.2; extra == "tf"
 Requires-Dist: boto3==1.34.102; extra == "tf"
 Requires-Dist: pdfplumber>=0.11.0; extra == "tf"
@@ -88,7 +88,7 @@ Requires-Dist: tqdm==4.64.0; extra == "pt"
 Requires-Dist: timm>=0.9.16; extra == "pt"
 Requires-Dist: transformers>=4.48.0; extra == "pt"
 Requires-Dist: accelerate>=0.29.1; extra == "pt"
-Requires-Dist: python-doctr==0.8.1; extra == "pt"
+Requires-Dist: python-doctr==0.9.0; extra == "pt"
 Requires-Dist: boto3==1.34.102; extra == "pt"
 Requires-Dist: pdfplumber>=0.11.0; extra == "pt"
 Requires-Dist: fasttext-wheel; extra == "pt"

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/deepdoctection.egg-info/requires.txt

@@ -69,7 +69,7 @@ tqdm==4.64.0
 timm>=0.9.16
 transformers>=4.48.0
 accelerate>=0.29.1
-python-doctr==0.8.1
+python-doctr==0.9.0
 boto3==1.34.102
 pdfplumber>=0.11.0
 fasttext-wheel
@@ -105,7 +105,7 @@ tensorpack==0.11
 protobuf==3.20.1
 tensorflow-addons>=0.17.1
 tf2onnx>=1.9.2
-python-doctr==0.8.1
+python-doctr==0.9.0
 pycocotools>=2.0.2
 boto3==1.34.102
 pdfplumber>=0.11.0

{deepdoctection-0.39.2 → deepdoctection-0.39.4}/setup.py

@@ -89,7 +89,7 @@ _DEPS = [
     "pdfplumber>=0.11.0",
     "tensorflow-addons>=0.17.1",
     "tf2onnx>=1.9.2",
-    "python-doctr==0.8.1",
+    "python-doctr==0.9.0",
     "fasttext-wheel",
     # dev dependencies
     "python-dotenv==1.0.0",