deepdoctection 0.39.2__py3-none-any.whl → 0.39.4__py3-none-any.whl

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/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/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/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/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/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.dist-info → deepdoctection-0.39.4.dist-info}/METADATA

@@ -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.dist-info → deepdoctection-0.39.4.dist-info}/RECORD

@@ -1,4 +1,4 @@
-deepdoctection/__init__.py,sha256=KpFnC7nCOVQOgeXDWt0fIjKoikD3MRj-_iSAbARIWeQ,12754
+deepdoctection/__init__.py,sha256=4u3rGxdbvfFr1l6xzH6qO5MapLGspBv5XPhBGMeji_4,12754
 deepdoctection/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 deepdoctection/analyzer/__init__.py,sha256=icClxrd20XutD6LxLgEPIWceSs4j_QfI3szCE-9BL2w,729
 deepdoctection/analyzer/_config.py,sha256=1rfvVrp7cI2YLzpahD77aa1tZ_KFAIQ21DM1NWhxYiI,5058
@@ -19,14 +19,14 @@ deepdoctection/datapoint/__init__.py,sha256=3K406GbOPhoEp8koVaSbMocmSsmWifnSZ1SP
 deepdoctection/datapoint/annotation.py,sha256=FEgz4COxVDfjic0gG7kS6iHnWLBIgFnquQ63Cbj2a4Y,22531
 deepdoctection/datapoint/box.py,sha256=UAdSnLexvFyg4KK1u9kXdJxhaWTwRxTU-cnQcvl37Q8,23410
 deepdoctection/datapoint/convert.py,sha256=gJbHY2V8nlMht1N5VdxTmWSsOeydpFPTJsaJHp6XGgE,7516
-deepdoctection/datapoint/image.py,sha256=S6yfsIRQgMCl6HYAcHYJSBcbfdYKKtebtkEkkkrXsMQ,33619
+deepdoctection/datapoint/image.py,sha256=dImZfJr72jS8yanZx1n70p4lIv0Qa21-qlUcj0DZcls,34060
 deepdoctection/datapoint/view.py,sha256=XPyhbBr2cGIKdAiISBVZWxNxlSvN8kmGsD9P0mfpEEE,50772
 deepdoctection/datasets/__init__.py,sha256=-A3aR90aDsHPmVM35JavfnQ2itYSCn3ujl4krRni1QU,1076
 deepdoctection/datasets/adapter.py,sha256=Ly_vbOAgVI73V41FUccnSX1ECTOyesW_qsuvQuvOZbw,7796
 deepdoctection/datasets/base.py,sha256=DT4i-d74sIEiUNC6UspIHNJuHSK0t1dBv7qwadg4rLw,22341
 deepdoctection/datasets/dataflow_builder.py,sha256=cYU2zV3gZW2bFvMHimlO9VIl3BAUaCwML08cCIQ8Em4,4107
 deepdoctection/datasets/info.py,sha256=6y5TfiUhQppynbMFP5JmUPk95ggsVCtGIw4dYh2lVus,20501
-deepdoctection/datasets/registry.py,sha256=ZjzVzjsCgNXJuZZZtR98_yKocADmh4EBGV5JqJbGjWk,2543
+deepdoctection/datasets/registry.py,sha256=tvzMUk34ZD3AsedS1DFfYATYLxm-7bn2-8J1AJiXGKM,2616
 deepdoctection/datasets/save.py,sha256=Y9508Qqp8gIGN7pbGgVBBnkiC6NdCb9L2YR4wVvEUxM,3350
 deepdoctection/datasets/instances/__init__.py,sha256=XEc_4vT5lDn6bbZID9ujDEumWu8Ec2W-QS4pI_bfWWE,1388
 deepdoctection/datasets/instances/doclaynet.py,sha256=wRZT7wMTilZBLZ1gKY2cWReD1EGT735vOOTy0pD0N6M,12038
@@ -53,7 +53,7 @@ deepdoctection/extern/__init__.py,sha256=9Iks9b4Q_LynjcV167TVCoK8YsQRUcA2jjmAmDN
 deepdoctection/extern/base.py,sha256=ONPgappl_P5HSwQr42FatuRnwMTvUPecPsCztDTN0Hw,24108
 deepdoctection/extern/d2detect.py,sha256=zrKv1yurApnjD7QZIZk_8LYCahjmN82MQUjHjv8zvkQ,22127
 deepdoctection/extern/deskew.py,sha256=sPoixu8S9he-0wbs-jgxtPE2V9BiP4-3uZlb6F5Y1SA,3077
-deepdoctection/extern/doctrocr.py,sha256=T3_tvlih22_dVCBZypS1Y8tjQQB1fkAxIbGdUGHIapQ,24473
+deepdoctection/extern/doctrocr.py,sha256=WrFA0N54fr9C_ahGzZJb4H-fNzz5wXQFveFiERYAm74,24637
 deepdoctection/extern/fastlang.py,sha256=F4gK-SEwcCujjxH327ZDzMGWToJ49xS_dCKcePQ9IlY,4780
 deepdoctection/extern/hfdetr.py,sha256=JzHrrTyzS9qh6T2TsvKboAGZkIhno2txmSoLQ5Vd-lo,12077
 deepdoctection/extern/hflayoutlm.py,sha256=tFaf90FRbZzhSycdp8rGkeiPywQa6UcTEEwbayIXkr0,57023
@@ -88,7 +88,7 @@ deepdoctection/extern/tp/tpfrcnn/utils/__init__.py,sha256=kiPlXxHlTGN9eI7YE9Bgwt
 deepdoctection/extern/tp/tpfrcnn/utils/box_ops.py,sha256=aBLqPg_ApaiimtBRaOsLKTZZFIBh87vVtqjLPMaX9fQ,2379
 deepdoctection/extern/tp/tpfrcnn/utils/np_box_ops.py,sha256=O-q1GQiOEd1lN1MQDsJvHwD2OmBO-qHNeqJ1Qnec93g,3539
 deepdoctection/mapper/__init__.py,sha256=Xqb34aCjslZDQnqQgCSvnloL5DbdT9eHhn-StpVPbzE,1130
-deepdoctection/mapper/cats.py,sha256=s73JzONV2UQ71szfljurk7H1-UjDBWsW4oNLs5xePUk,16474
+deepdoctection/mapper/cats.py,sha256=Go9k9wiSid1aSPSteTCE0AgQ1tZmOA8pfOPkhKSQhhg,16601
 deepdoctection/mapper/cocostruct.py,sha256=GcbUpPFUg67pcOHQluWBFOFcGaYnlZcTmwBDERBVgCA,5978
 deepdoctection/mapper/d2struct.py,sha256=Dx-YnycsIQH4a5-9Gn_yMhiQ-gOFgMueNeH3rhXjuCU,8555
 deepdoctection/mapper/hfstruct.py,sha256=2PjGKsYturVJBimLT1CahYh09KSRAFEHz_QNtC162kQ,5551
@@ -113,7 +113,7 @@ deepdoctection/pipe/lm.py,sha256=x9NoYpivdjQF1r76a7PPrUuBEmuHP7ZukuXFDkXhXBc,175
 deepdoctection/pipe/order.py,sha256=PnJZiCnxFluJiECXLTZT0c1Rr66vIRBFraa_G41UA2k,40121
 deepdoctection/pipe/refine.py,sha256=dTfI396xydPdbzpfo4yqFcuxl3UAB1y-WbSQn1o76ec,22367
 deepdoctection/pipe/registry.py,sha256=aFx-Tn0xhVA5l5H18duNW5QoTNKQltybsEUEzsMgUfg,902
-deepdoctection/pipe/segment.py,sha256=CR83HQMW0hrRG8W6pFuB0YibxQMWpqI7_LaUIcJcQwo,59116
+deepdoctection/pipe/segment.py,sha256=mWYRg7UR80PtIj1SIg_hiujDcCtLlvKJUP9vx4ZpW0Y,59318
 deepdoctection/pipe/sub_layout.py,sha256=N1RcID-boORcwsW_j0l64HpUu3rff0ge5qEanudLYgk,13838
 deepdoctection/pipe/text.py,sha256=h9q6d3HFOs7LOg-iwdLUPiQxrPqgunBVNmtYMBrfRQE,11180
 deepdoctection/pipe/transform.py,sha256=9Om7X7hJeL4jgUwHM1CHa4sb5v7Qo1PtVG0ls_3nI7w,3798
@@ -141,8 +141,8 @@ deepdoctection/utils/transform.py,sha256=3kCgsEeRkG1efCdkfvj7tUFMs-e2jbjbflq826F
 deepdoctection/utils/types.py,sha256=_3dmPdCIZNLbgU5QP5k_c5phDf18xLe1kYL6t2nM45s,2953
 deepdoctection/utils/utils.py,sha256=csVs_VvCq4QBETPoE2JdTTL4MFYnD4xh-Js5vRb612g,6492
 deepdoctection/utils/viz.py,sha256=Jf8ePNYWlpuyaS6SeTYQ4OyA3eNhtgjvAQZnGNdgHC0,27051
-deepdoctection-0.39.2.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
-deepdoctection-0.39.2.dist-info/METADATA,sha256=Z6b0DeMKoXZOR6O7Yxu7Qt5_MLgCxxvyNuWsPAhFo0M,19741
-deepdoctection-0.39.2.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-deepdoctection-0.39.2.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
-deepdoctection-0.39.2.dist-info/RECORD,,
+deepdoctection-0.39.4.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
+deepdoctection-0.39.4.dist-info/METADATA,sha256=9DvyyvO9Ofd9YFuqQ2SVO2Bi69WayBC2VZyrpMCimZQ,19741
+deepdoctection-0.39.4.dist-info/WHEEL,sha256=52BFRY2Up02UkjOa29eZOS2VxUrpPORXg1pkohGGUS8,91
+deepdoctection-0.39.4.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
+deepdoctection-0.39.4.dist-info/RECORD,,

{deepdoctection-0.39.2.dist-info → deepdoctection-0.39.4.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.8.0)
+Generator: setuptools (76.0.0)
 Root-Is-Purelib: true
 Tag: py3-none-any