deepdoctection 0.43__py3-none-any.whl → 0.43.3__py3-none-any.whl

deepdoctection/__init__.py

@@ -25,11 +25,11 @@ from .utils.logger import LoggingRecord, logger
 
 # pylint: enable=wrong-import-position
 
-__version__ = "0.43"
+__version__ = "0.43.3"
 
 _IMPORT_STRUCTURE = {
     "analyzer": ["config_sanity_checks", "get_dd_analyzer", "ServiceFactory"],
-    "configs": [],
+    "configs": ["update_cfg_from_defaults"],
     "dataflow": [
         "DataFlowTerminated",
         "DataFlowResetStateNotCalled",
@@ -98,6 +98,7 @@ _IMPORT_STRUCTURE = {
         "List",
         "Cell",
         "Table",
+        "IMAGE_DEFAULTS",
         "Page",
     ],
     "datasets": [

deepdoctection/analyzer/config.py

@@ -902,3 +902,18 @@ cfg.LAYOUT_LINK.CHILD_CATEGORIES = [LayoutType.CAPTION]
 # Freezes the configuration to make it immutable.
 # This prevents accidental modification at runtime.
 cfg.freeze()
+
+def update_cfg_from_defaults() -> None:
+    """
+    Update the configuration with current values from IMAGE_DEFAULTS.
+    """
+    cfg.freeze(False)
+
+    # Update all dependent fields from IMAGE_DEFAULTS
+    cfg.TEXT_CONTAINER = IMAGE_DEFAULTS.TEXT_CONTAINER
+    cfg.WORD_MATCHING.PARENTAL_CATEGORIES = IMAGE_DEFAULTS.TEXT_BLOCK_CATEGORIES
+    cfg.TEXT_ORDERING.TEXT_BLOCK_CATEGORIES = IMAGE_DEFAULTS.TEXT_BLOCK_CATEGORIES
+    cfg.TEXT_ORDERING.FLOATING_TEXT_BLOCK_CATEGORIES = IMAGE_DEFAULTS.FLOATING_TEXT_BLOCK_CATEGORIES
+
+    # Re-freeze the configuration
+    cfg.freeze()

deepdoctection/datapoint/image.py

@@ -479,8 +479,8 @@ class Image:
 
     def remove(
         self,
-        annotation_ids: Optional[Union[str, list[str]]] = None,
-        service_ids: Optional[Union[str, list[str]]] = None,
+        annotation_ids: Optional[Union[str, Sequence[str]]] = None,
+        service_ids: Optional[Union[str, Sequence[str]]] = None,
     ) -> None:
         """
         Instead of removing consider deactivating annotations.

deepdoctection/datapoint/view.py

@@ -183,7 +183,7 @@ class Word(ImageAnnotationBaseView):
         attr_names = (
             set(WordType)
             .union(super().get_attribute_names())
-            .union({Relationships.READING_ORDER, Relationships.LAYOUT_LINK})
+            .union({Relationships.READING_ORDER, Relationships.LAYOUT_LINK, Relationships.LINK})
         )
         return {attr_name.value if isinstance(attr_name, ObjectTypes) else attr_name for attr_name in attr_names}
 
@@ -773,6 +773,7 @@ class Page(Image):
         "figures",
         "residual_layouts",
         "document_summary",
+        "document_mapping",
     }
     include_residual_text_container: bool = True
 
@@ -1389,19 +1390,32 @@ class Page(Image):
             include_residual_text_container=include_residual_text_container,
         )
 
-    def get_token(self) -> list[Mapping[str, str]]:
+    def get_entities(self) -> list[Mapping[str, str]]:
         """
         Returns:
-             A list of tuples with word and non default token tags
+             A list of dicts with the following structure:
+
+            ```python
+            {
+                "word": str,  # word characters
+                "entity": str, # token tag
+                "annotation_id": str, # annotation id of the word
+                "successor_annotation_id": Optional[str] # annotation_id of the successor word, if any
+            }
+            ```
+
         """
         block_with_order = self._order("layouts")
         all_words = []
         for block in block_with_order:
             all_words.extend(block.get_ordered_words())  # type: ignore
         return [
-            {"word": word.CHARACTERS, "entity": word.TOKEN_TAG}
+            {"word": word.characters,
+             "entity": word.token_tag.value,
+             "annotation_id": word.annotation_id,
+             "successor_annotation_id": word.successor[0].annotation_id if word.successor else None}
             for word in all_words
-            if word.TOKEN_TAG not in (TokenClasses.OTHER, None)
+            if word.token_tag not in (TokenClasses.OTHER, None)
         ]
 
     def __copy__(self) -> Page:

deepdoctection/extern/base.py

@@ -502,6 +502,7 @@ class TokenClassResult:
        semantic_name: semantic name
        bio_tag: bio tag
        score: prediction score
+       successor_uuid: uuid of the next token in the sequence
     """
 
     uuid: str
@@ -512,6 +513,7 @@ class TokenClassResult:
     bio_tag: ObjectTypes = DefaultType.DEFAULT_TYPE
     score: Optional[float] = None
     token_id: Optional[int] = None
+    successor_uuid: Optional[str] = None
 
 
 @dataclass

deepdoctection/extern/doctrocr.py

@@ -269,7 +269,10 @@ class DoctrTextlineDetector(DoctrTextlineDetectorMixin):
         if self.lib == "PT":
             self.device = get_torch_device(device)
 
-        self.doctr_predictor = self.get_wrapped_model(self.architecture, self.path_weights, self.device, self.lib)
+        self.doctr_predictor = self.get_wrapped_model(self.architecture,
+                                                      self.path_weights,
+                                                      self.device,
+                                                      self.lib)
 
     def predict(self, np_img: PixelValues) -> list[DetectionResult]:
         """
@@ -424,7 +427,7 @@ class DoctrTextRecognizer(TextRecognizer):
         return _get_doctr_requirements()
 
     def clone(self) -> DoctrTextRecognizer:
-        return self.__class__(self.architecture, self.path_weights, self.device, self.lib)
+        return self.__class__(self.architecture, self.path_weights, self.device, self.lib, self.path_config_json)
 
     @staticmethod
     def load_model(

deepdoctection/extern/texocr.py

@@ -60,7 +60,7 @@ def _textract_to_detectresult(response: JsonDict, width: int, height: int, text_
     return all_results
 
 
-def predict_text(np_img: PixelValues, client: boto3.client, text_lines: bool) -> list[DetectionResult]:  # type: ignore
+def predict_text(np_img: PixelValues, client: boto3.client, text_lines: bool) -> list[DetectionResult]:
     """
     Calls AWS Textract client (`detect_document_text`) and returns plain OCR results.
     AWS account required.

deepdoctection/mapper/xfundstruct.py

@@ -200,7 +200,7 @@ def xfund_to_image(
                 ann_ids.extend(entity_id_to_ann_id[linked_entity])
             for ann_id in ann_ids:
                 if ann_id != word.annotation_id:
-                    word.dump_relationship(Relationships.SEMANTIC_ENTITY_LINK, ann_id)
+                    word.dump_relationship(Relationships.LINK, ann_id)
 
     if mapping_context.context_error:
         return None

deepdoctection/pipe/anngen.py

@@ -19,7 +19,7 @@
 Datapoint manager
 """
 from dataclasses import asdict
-from typing import Optional, Union
+from typing import Optional, Union, Sequence
 
 import numpy as np
 
@@ -372,6 +372,19 @@ class DatapointManager:
             return None
         return ann.annotation_id
 
+    def remove_annotations(self, annotation_ids: Sequence[str]) -> None:
+        """
+        Removes the annotation by the given `annotation_id`.
+
+        Args:
+            annotation_ids: The `annotation_id` to remove.
+        """
+        self.assert_datapoint_passed()
+        self.datapoint.remove(annotation_ids)
+        for ann_id in annotation_ids:
+            if ann_id in self._cache_anns:
+                self._cache_anns.pop(ann_id)
+
     def deactivate_annotation(self, annotation_id: str) -> None:
         """
         Deactivates the annotation by the given `annotation_id`.

deepdoctection/pipe/order.py

@@ -738,10 +738,8 @@ class TextOrderService(TextLineServiceMixin):
                 text_block_anns.extend(residual_text_container_anns)
         for text_block_ann in text_block_anns:
             self.order_text_in_text_block(text_block_ann)
-        floating_text_block_anns_to_order = [
-            ann for ann in text_block_anns if ann.category_name in self.floating_text_block_categories
-        ]
-        self.order_blocks(floating_text_block_anns_to_order)
+        floating_text_block_anns = dp.get_annotation(category_names=self.floating_text_block_categories)
+        self.order_blocks(floating_text_block_anns)
         self._create_columns()
 
     def _create_columns(self) -> None:
@@ -803,9 +801,14 @@ class TextOrderService(TextLineServiceMixin):
         if self.include_residual_text_container:
             add_category.append(LayoutType.LINE)
 
-        assert set(self.floating_text_block_categories) <= set(
+        if set(self.floating_text_block_categories) <= set(
             self.text_block_categories + tuple(add_category)
-        ), "floating_text_block_categories must be a subset of text_block_categories"
+        ):
+            logger.warning("In most cases floating_text_block_categories must be a subset of text_block_categories. "
+                           "Adding categories to floating_text_block_categories, that do not belong to "
+                           "text_block_categories makes only sense for categories set have CHILD relationships with"
+                           " annotations that belong to text_block_categories.")
+
 
     def get_meta_annotation(self) -> MetaAnnotation:
         add_category = [self.text_container]

deepdoctection/utils/context.py

@@ -38,7 +38,7 @@ __all__ = ["timeout_manager", "save_tmp_file", "timed_operation"]
 
 
 @contextmanager
-def timeout_manager(proc: Any, seconds: Optional[int] = None) -> Iterator[str]:  # type: ignore
+def timeout_manager(proc: Any, seconds: Optional[int] = None) -> Iterator[str]:
     """
     Manager for time handling while some process is being called.
 

deepdoctection/utils/develop.py

@@ -71,6 +71,7 @@ def log_deprecated(name: str, text: str, eos: str = "", max_num_warnings: Option
     logger.info(LoggingRecord(f"[Deprecated] {info_msg}"))
 
 
+
 def deprecated(
     text: str = "", eos: str = "", max_num_warnings: Optional[int] = None
 ) -> Callable[[Callable[..., T]], Callable[..., T]]:

deepdoctection/utils/env_info.py

@@ -462,7 +462,7 @@ def pt_info(data: KeyValEnvInfos) -> KeyValEnvInfos:
                 data.append(("torchvision arch flags", msg))
             except (ImportError, AttributeError):
                 data.append(("torchvision._C", "Not found"))
-    except AttributeError:
+    except (AttributeError, ModuleNotFoundError):
         data.append(("torchvision", "unknown"))
 
     return data

deepdoctection/utils/file_utils.py

@@ -12,7 +12,6 @@ import importlib.util
 import multiprocessing as mp
 import string
 import subprocess
-import sys
 from os import environ, path
 from shutil import which
 from types import ModuleType
@@ -22,7 +21,6 @@ import importlib_metadata
 from packaging import version
 
 from .error import DependencyError
-from .logger import LoggingRecord, logger
 from .metacfg import AttrDict
 from .types import PathLikeOrStr, Requirement
 
@@ -662,13 +660,6 @@ def get_doctr_requirement() -> Requirement:
         On macOS, if `poppler` is not available, this function will recursively check the requirement.
         It is not yet known how to check whether `pango`, `gdk-pixbuf`, and `libffi` are installed.
     """
-    if sys.platform == "darwin":
-        if not get_poppler_version():
-            return get_doctr_requirement()
-        # don't know yet how to check whether pango gdk-pixbuf libffi are installed
-        logger.info(
-            LoggingRecord("package requires weasyprint. Check that poppler pango gdk-pixbuf libffi are installed")
-        )
     return "doctr", doctr_available(), _DOCTR_ERR_MSG
 
 

deepdoctection/utils/settings.py

@@ -80,6 +80,7 @@ class SummaryType(ObjectTypes):
 
     SUMMARY = "summary"
     DOCUMENT_SUMMARY = "document_summary"
+    DOCUMENT_MAPPING = "document_mapping"
 
 
 @object_types_registry.register("DocumentType")
@@ -228,8 +229,9 @@ class Relationships(ObjectTypes):
 
     CHILD = "child"
     READING_ORDER = "reading_order"
-    SEMANTIC_ENTITY_LINK = "semantic_entity_link"
+    LINK = "link"
     LAYOUT_LINK = "layout_link"
+    SUCCESSOR = "successor"
 
 
 @object_types_registry.register("Languages")

{deepdoctection-0.43.dist-info → deepdoctection-0.43.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepdoctection
-Version: 0.43
+Version: 0.43.3
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -192,7 +192,7 @@ Check the demo of a document layout analysis pipeline with OCR on 🤗
 - Fine-tuning and evaluation tools.
 - Lot's of [tutorials](https://github.com/deepdoctection/notebooks)
 
-Have a look at the [**introduction notebook**](https://github.com/deepdoctection/notebooks/blob/main/Get_Started.ipynb)
+Have a look at the [**introduction notebook**](https://github.com/deepdoctection/notebooks/blob/main/Analyzer_Get_Started.ipynb)
 for an easy start.
 
 Check the [**release notes**](https://github.com/deepdoctection/deepdoctection/releases) for recent updates.
@@ -245,11 +245,11 @@ alt="text" width="40%">
 
 ## Requirements
 
-![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/requirements_deepdoctection_220525.png)
+![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/install_01.png)
 
 - Linux or macOS. Windows is not supported but there is a [Dockerfile](./docker/pytorch-cpu-jupyter/Dockerfile) available.
 - Python >= 3.9
-- 1.13 \<= PyTorch **or** 2.11 \<= Tensorflow < 2.16. (For lower Tensorflow versions the code will only run on a GPU).
+- 2.2 \<= PyTorch **or** 2.11 \<= Tensorflow < 2.16. (For lower Tensorflow versions the code will only run on a GPU).
   Tensorflow support will be stopped from Python 3.11 onwards.
 - To fine-tune models, a GPU is recommended.
 
@@ -344,8 +344,7 @@ pip install ".[tf]" # or "pip install -e .[tf]"
 
 ### Running a Docker container from Docker hub
 
-Pre-existing Docker images can be downloaded from the [Docker hub](https://hub.docker.
-com/r/deepdoctection/deepdoctection).
+Pre-existing Docker images can be downloaded from the [Docker hub](https://hub.docker.com/r/deepdoctection/deepdoctection).
 
 ```
 docker pull deepdoctection/deepdoctection:<release_tag> 

{deepdoctection-0.43.dist-info → deepdoctection-0.43.3.dist-info}/RECORD

@@ -1,7 +1,7 @@
-deepdoctection/__init__.py,sha256=T4BXZotL855uGwIHhore8lZAfSinpIcrpeIvrVsSCyc,12910
+deepdoctection/__init__.py,sha256=nJ6cTH9fWxNIGSDjMS7AoU9hbJwfbb-SvTwOw6eiPAg,12964
 deepdoctection/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 deepdoctection/analyzer/__init__.py,sha256=wg0BcFwdCeREwzZfa--Yx8HUJ9LPv5z5PmLwtkZdPH8,772
-deepdoctection/analyzer/config.py,sha256=CgH4etvZL0JdeIHAXMFdDro2VvVsF2itgQar_Ml94pw,41185
+deepdoctection/analyzer/config.py,sha256=Uwi9MYOym0QXzFcip7bRXoy732P4tZVMFBUnUZk9c1w,41761
 deepdoctection/analyzer/dd.py,sha256=2BGvZpl9o9khcaOV52-DPHMrs0DsqUO8cpdqFVHHzDQ,5176
 deepdoctection/analyzer/factory.py,sha256=DI0S38KAG2sIROrSximsWJsMbem91a9zXaeWsDNvkGg,37574
 deepdoctection/configs/__init__.py,sha256=TX_P6tqDOF1LK1mi9ruAl7x0mtv1Asm8cYWCz3Pe2dk,646
@@ -20,8 +20,8 @@ deepdoctection/datapoint/__init__.py,sha256=DOhcN81MhyPUuFw9F4pyEDyZseeD9qxb8NdB
 deepdoctection/datapoint/annotation.py,sha256=f32BNmzUGJoNMeGst2RGC2jmjJpzzjxyBRKFG8FCubY,23092
 deepdoctection/datapoint/box.py,sha256=QAS8sK2Ge4_ysW6zOYkLlzNwhSyw_mhYcYsxscClEno,31453
 deepdoctection/datapoint/convert.py,sha256=6ENXX3tBdY8ogb2NBPxsOsQMGnQux8ol5nrUfWS5tYE,7352
-deepdoctection/datapoint/image.py,sha256=nDaWUtdD5j6l_iXW9d2PoIyXBC8M3_idoEIXm7JWGyQ,35139
-deepdoctection/datapoint/view.py,sha256=5TYmKpNNZwJb-NrUXv08H3_zSfHKDHhg6LnEZjBqVns,56622
+deepdoctection/datapoint/image.py,sha256=kqwCz8DSc19hQpkl_4L1_Ek7_2KrH5KsV9e0S-R4n5w,35147
+deepdoctection/datapoint/view.py,sha256=lVUTi2vfh6w7k92kcwQNpKmfOGgZmM12bYJUFrO5Pno,57163
 deepdoctection/datasets/__init__.py,sha256=4ifjIwWCPYiS31GzUlVDScrkNOrb1eo5xHlRXNyg_58,994
 deepdoctection/datasets/adapter.py,sha256=VSLM_980aHi4TpgOxfxiBHiF_fUXyh348PXet6zTo-4,7779
 deepdoctection/datasets/base.py,sha256=HTIquJir2BZRTLl1HSQM0ICfvjIaWAjJeyz3BEHgdb0,23175
@@ -51,10 +51,10 @@ deepdoctection/eval/registry.py,sha256=us6EGN_tAia1Mk1mwWQwDeE-xqxcuopztdi8n-ieG
 deepdoctection/eval/tedsmetric.py,sha256=EcNeJynsmxyl5bOH3bjy2wE647ONf0SF5OZyGbVu35Q,9963
 deepdoctection/eval/tp_eval_callback.py,sha256=lqrOn2tdaRiF_Vr_9CwBr2ryatcWu3mQKya8YZ2pA9A,5261
 deepdoctection/extern/__init__.py,sha256=jG2qe5_X7fJFnJlx04Lf1KUTXKKKYtCkKR7WQ7looUk,991
-deepdoctection/extern/base.py,sha256=vs4EO9vkfyTW-nVM0dnlqsmnPPeHIXI2wW5b4Wpiz-Y,31547
+deepdoctection/extern/base.py,sha256=LomTR9HXcBU55MPDIA8D1rIamk7DUmToJmgcRXzCoeU,31650
 deepdoctection/extern/d2detect.py,sha256=I0oEkprr5iVpKpM3a3nknAU-sXwNoDQdp_B1gzzODsk,22374
 deepdoctection/extern/deskew.py,sha256=L_jU0rXh03qzwaT79EIqE_zYMUVeFwWDbsGbtahuL2k,3124
-deepdoctection/extern/doctrocr.py,sha256=d468jZpR4WBvZPt5Vze70dh6tZacBM1HO0pcy8ynOFM,25358
+deepdoctection/extern/doctrocr.py,sha256=vOI67NCeEznHzdY4zu1tggwg3_PfwV1eigsr2Z8EPzc,25543
 deepdoctection/extern/fastlang.py,sha256=4D9A-_hTXUcvXG6IJJknX34LrD71v08XtNdWgvXD7fE,4736
 deepdoctection/extern/hfdetr.py,sha256=N3eLNI5BsQS9_7YZyBeWndSgUydJij7ugZA9p4V1xaQ,14316
 deepdoctection/extern/hflayoutlm.py,sha256=3mZZ3byn00jSrLWO2vZFas9j4VrhbYQNmF1mwPG2ElQ,59642
@@ -62,7 +62,7 @@ deepdoctection/extern/hflm.py,sha256=y-9brzmT2NYtFoNcWHABNg2ZZQXSOP9CyqtT1OoeV9U
 deepdoctection/extern/model.py,sha256=-GbnuhLFq7jpBOvtpJe6IhGXxQdqwiM8epEd7IRELoU,18234
 deepdoctection/extern/pdftext.py,sha256=ljzPQn3yYAlS6MoZqzixD-fO2GlHwu1aMiOQ6qMIzbg,7513
 deepdoctection/extern/tessocr.py,sha256=SuPmngsJg38riL4b09z6_FIzJH6H3RIwoighG2GPMYM,17457
-deepdoctection/extern/texocr.py,sha256=93vGj0TX2gENMFV6_FDk3et1sDecrNeuozv5EfOR5nk,5931
+deepdoctection/extern/texocr.py,sha256=wVOuu6eUGao0mUbC8vrgdCsKfY1GqA1Am9560YgWyXU,5915
 deepdoctection/extern/tpdetect.py,sha256=Kr00n80V_OfE-EGfpjiVw1eAQ2n2tuT-hSco-dLSR9E,8516
 deepdoctection/extern/pt/__init__.py,sha256=3Cu0ZHjbYsJomru7-RQXEHihEQLegZrmLetlHiqS58I,742
 deepdoctection/extern/pt/nms.py,sha256=2lSpEH8cI_QXdz5xL_OaitqsGoHhp5xvDssK5Yo8q4Q,2218
@@ -101,9 +101,9 @@ deepdoctection/mapper/pascalstruct.py,sha256=PviZjhTk4p5HDUTlF8qhWPyraKD0uh51f2h
 deepdoctection/mapper/prodigystruct.py,sha256=OWzPUbNDrqwFipH8YWI5eSxwMdA7qYczaFdsHNrE_4c,7001
 deepdoctection/mapper/pubstruct.py,sha256=UTyfUmzMSuf2BXtdYwHjK7ngsIwAxSZjwTxDtz6DySg,23416
 deepdoctection/mapper/tpstruct.py,sha256=dxtEVHYVnkH-zjjbHzkFrPgS9eheys6E-CMlsjaOnxo,5468
-deepdoctection/mapper/xfundstruct.py,sha256=XLUZ-yBMWtKFQ40vxHl6p8EZZvl68JdwJlV00A93Zy8,9108
+deepdoctection/mapper/xfundstruct.py,sha256=JVs92T5l3HedQvdc6ZMACj2mq5Co4WCS9B-CnQ01opA,9092
 deepdoctection/pipe/__init__.py,sha256=E3cYAVWOvMzIN7jbKFyqLjFXahcFGcAGkb1uChM_XCY,1034
-deepdoctection/pipe/anngen.py,sha256=Hfi7C6-iOv7t8tjFoz4FuIhcz6yMZx52f5SG9bsVnLg,16365
+deepdoctection/pipe/anngen.py,sha256=I7VFoTTa-sJCBwFJNaZeLD7pcrLjRPTX2I6-ZDCDHPI,16819
 deepdoctection/pipe/base.py,sha256=oszB_DepcFtORvDdGTZZPWMhk01C68RUWXHjeX7SF3M,18163
 deepdoctection/pipe/common.py,sha256=OcsqHr_c66Yqt98hFeKwaa0mciWMCauw0HZ3YnHx8MU,24586
 deepdoctection/pipe/concurrency.py,sha256=_EKZi4eCeF3mVHytZL_fMwyqa25C2aR9g8vrIFB8iR4,9780
@@ -111,7 +111,7 @@ deepdoctection/pipe/doctectionpipe.py,sha256=ik5F92F3klI5Nve_AnyIRj-ApMoKHSR2Sjc
 deepdoctection/pipe/language.py,sha256=T5g5_2GIsbTltAmn_PFymMUMoik8_b0uJNx8f5dT9MM,5898
 deepdoctection/pipe/layout.py,sha256=oAldMtwyZee1IqpuflKKvmeL2Z_nXFiqwFMS4VYv5eI,6391
 deepdoctection/pipe/lm.py,sha256=nYI2bm0sc9d3JMlIPyNyd4XxXFRBIHRUYfMImuek6b4,19793
-deepdoctection/pipe/order.py,sha256=9OarsHKwVqT1bTDIn7XGeGLgpetEJW3uLjuJQDdhjG4,40684
+deepdoctection/pipe/order.py,sha256=8CqEWUA6U9HxThKKGP9yJMbRaML2by7do0Gdhl_7AdI,40964
 deepdoctection/pipe/refine.py,sha256=SrMcAWXRO5tJpqaZCEz9RzvjPyiQiE8fZ9TXBcaBKck,23310
 deepdoctection/pipe/registry.py,sha256=uT5fnHjffoNGk2JPuD2-pMYtO3Iko7-wrwVZVCWLtok,906
 deepdoctection/pipe/segment.py,sha256=rHhEWr5zZ1ppj-gMa-q-UCr1AYTWpUW7oA1umwebqBI,61302
@@ -125,25 +125,25 @@ deepdoctection/train/hf_layoutlm_train.py,sha256=bNL5OCLKytshG6kaTJDLTQOcvWKwEsY
 deepdoctection/train/tp_frcnn_train.py,sha256=Tltb-v2JD5oPuHCZGA9B5DM4ZaidoBITlH93QX-KPKI,13570
 deepdoctection/utils/__init__.py,sha256=brBceRWeov9WXMiJTjyJOF2rHMP8trGGRRjhMdZ61nI,2371
 deepdoctection/utils/concurrency.py,sha256=9ly81D5i2ZFzKfXMIUSmhT42eMs4QttsRhWXdkIk6Aw,5832
-deepdoctection/utils/context.py,sha256=GXgIGQ10JwosE6FQSPdlJoCXyeM_cahEcbQ2mxGNofI,4538
-deepdoctection/utils/develop.py,sha256=4myrqBDypM6tQ2a2Jo3Q20RuE_W2czykpXBwgXPrxNw,3568
-deepdoctection/utils/env_info.py,sha256=yyRyjQT3xz73yEcqauPM6GdCpvWwYGAH-_KUHfXBrtM,19855
+deepdoctection/utils/context.py,sha256=5QfdzxsiSPnNs1qtJdgjguIoD8srLQ2W8oeDzwp9F78,4522
+deepdoctection/utils/develop.py,sha256=x2MhbmoKZyRluesmc01is7ldrUN9c0TX4OAuc1yt6dI,3569
+deepdoctection/utils/env_info.py,sha256=b1WohrfQuoL-BPN0_s8Rjtwzx-WKvCyaX2I4qYl1Emc,19878
 deepdoctection/utils/error.py,sha256=sIry8F5MZ0yLvKfAwVz90IorKWVvjoRqcC0L8qq8mLk,2480
-deepdoctection/utils/file_utils.py,sha256=D4cua4i3Q-4ZeSRXXWEL4z7hp0M1qh9rrVSfS5t-Hzo,25643
+deepdoctection/utils/file_utils.py,sha256=EepfAZVADaqpBdVq2LOJXLFLsMd_oZF_FAKUHOAhiZ0,25246
 deepdoctection/utils/fs.py,sha256=KTS9FJzZk9le_vmIPr9IisJw0AyTfjkyX1KoWQy4DNs,12729
 deepdoctection/utils/identifier.py,sha256=Jt12MeZf7eC1qciY5Fp_AYUGxYVcjsy7xNBUvJil7dU,2270
 deepdoctection/utils/logger.py,sha256=ddQ0xBStluf8OvoRlEB8YkqyRR-ZYgyJYLClTmJJMAU,10290
 deepdoctection/utils/metacfg.py,sha256=5M390--ZMoyJEt5oZOwFMGt2i8OF_ayeb0NVmUO_3OQ,7235
 deepdoctection/utils/mocks.py,sha256=IkN3-IzAl4eX0ibgKIHg8IY7ykVw6BnpF6XnxKnKaZI,2389
 deepdoctection/utils/pdf_utils.py,sha256=BrxTuY9j0COyIRkJchJ0tt2h6ZsA2an6z-H8E8QwgUQ,13490
-deepdoctection/utils/settings.py,sha256=OrFEe9Mll3UuDhjyS-cTCv_q1ZSr30Jpl9nQxk__t2I,12824
+deepdoctection/utils/settings.py,sha256=0P6nh9-84wWyMm9J7w9I7gI1xo8mN4M4xZ0IXzcqDbE,12862
 deepdoctection/utils/tqdm.py,sha256=kx3Ivf0x85S0ZmEaN5mImu0V6isOgygOU8iyr2U99XU,1850
 deepdoctection/utils/transform.py,sha256=jgeCyQWLN9q79jCGW7jysyKUKcJ1AVMk8OslF-3fbag,16095
 deepdoctection/utils/types.py,sha256=ti4WdtIJSg3TGK_YPkkoY9PYGMnR2tTX6Xfik8U1pNk,2986
 deepdoctection/utils/utils.py,sha256=NBUb1qbx8Jm-AvYN1Sdbk0huXhbAKxZ-ZtOcMespsMM,7064
 deepdoctection/utils/viz.py,sha256=bujRIujvX317rPz4jBrj0yd3WP8wPjDUiI5GUrw9MzQ,27339
-deepdoctection-0.43.dist-info/licenses/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
-deepdoctection-0.43.dist-info/METADATA,sha256=F08hSp-kyzwW1tsTWyRMtWG3TaJsA8_LgueZ7irkBqA,13404
-deepdoctection-0.43.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-deepdoctection-0.43.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
-deepdoctection-0.43.dist-info/RECORD,,
+deepdoctection-0.43.3.dist-info/licenses/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
+deepdoctection-0.43.3.dist-info/METADATA,sha256=tFoZH2VB3ZkiTwzVmd1OBlg1f6eFLrKwwcLyYo6BX1g,13389
+deepdoctection-0.43.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+deepdoctection-0.43.3.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
+deepdoctection-0.43.3.dist-info/RECORD,,