deepdoctection 0.37.3__py3-none-any.whl → 0.39__py3-none-any.whl

deepdoctection/pipe/sub_layout.py

@@ -49,27 +49,27 @@ class DetectResultGenerator:
 
     def __init__(
         self,
-        categories: Mapping[int, ObjectTypes],
-        group_categories: Optional[list[list[int]]] = None,
-        exclude_category_ids: Optional[Sequence[int]] = None,
+        categories_name_as_key: Mapping[ObjectTypes, int],
+        group_categories: Optional[list[list[ObjectTypes]]] = None,
+        exclude_category_names: Optional[Sequence[ObjectTypes]] = None,
         absolute_coords: bool = True,
     ) -> None:
         """
-        :param categories: The dict of all possible detection categories
+        :param categories_name_as_key: The dict of all possible detection categories
         :param group_categories: If you only want to generate only one DetectResult for a group of categories, provided
                                  that the sum of the group is less than one, then you can pass a list of list for
                                  grouping category ids.
         :param absolute_coords: 'absolute_coords' value to be set in 'DetectionResult'
         """
-        self.categories = MappingProxyType(dict(categories.items()))
+        self.categories_name_as_key = MappingProxyType(dict(categories_name_as_key.items()))
         self.width: Optional[int] = None
         self.height: Optional[int] = None
         if group_categories is None:
-            group_categories = [[idx] for idx in self.categories]
+            group_categories = [[cat_name] for cat_name in self.categories_name_as_key]
         self.group_categories = group_categories
-        if exclude_category_ids is None:
-            exclude_category_ids = []
-        self.exclude_category_ids = exclude_category_ids
+        if exclude_category_names is None:
+            exclude_category_names = []
+        self.exclude_category_names = exclude_category_names
         self.dummy_for_group_generated = [False for _ in self.group_categories]
         self.absolute_coords = absolute_coords
 
@@ -83,17 +83,17 @@ class DetectResultGenerator:
 
         if self.width is None and self.height is None:
             raise ValueError("Initialize height and width first")
-
+        detect_result_list = self._detection_result_sanity_check(detect_result_list)
         count = self._create_condition(detect_result_list)
-        for category_id in self.categories:
-            if category_id not in self.exclude_category_ids:
-                if count[category_id] < 1:
-                    if not self._dummy_for_group_generated(category_id):
+        for category_name in self.categories_name_as_key:
+            if category_name not in self.exclude_category_names:
+                if count[category_name] < 1:
+                    if not self._dummy_for_group_generated(category_name):
                         detect_result_list.append(
                             DetectionResult(
                                 box=[0.0, 0.0, float(self.width), float(self.height)],  # type: ignore
-                                class_id=int(category_id),
-                                class_name=self.categories[category_id],
+                                class_id=self.categories_name_as_key[category_name],
+                                class_name=category_name,
                                 score=0.0,
                                 absolute_coords=self.absolute_coords,
                             )
@@ -102,8 +102,8 @@ class DetectResultGenerator:
         self.dummy_for_group_generated = self._initialize_dummy_for_group_generated()
         return detect_result_list
 
-    def _create_condition(self, detect_result_list: list[DetectionResult]) -> dict[int, int]:
-        count = Counter([ann.class_id for ann in detect_result_list])
+    def _create_condition(self, detect_result_list: list[DetectionResult]) -> dict[ObjectTypes, int]:
+        count = Counter([ann.class_name for ann in detect_result_list])
         cat_to_group_sum = {}
         for group in self.group_categories:
             group_sum = 0
@@ -113,9 +113,25 @@ class DetectResultGenerator:
                 cat_to_group_sum[el] = group_sum
         return cat_to_group_sum
 
-    def _dummy_for_group_generated(self, category_id: int) -> bool:
+    @staticmethod
+    def _detection_result_sanity_check(detect_result_list: list[DetectionResult]) -> list[DetectionResult]:
+        """
+        Go through each detect_result in the list and check if the box argument has sensible coordinates:
+        ulx >= 0 and lrx - ulx >= 0 (same for y coordinate). Remove the detection result if this condition is not
+        satisfied. We need this check because if some detection results are not sane, we might end up with some
+        none existing categories.
+        """
+        sane_detect_results = []
+        for detect_result in detect_result_list:
+            if detect_result.box:
+                ulx, uly, lrx, lry = detect_result.box
+                if ulx >= 0 and lrx - ulx >= 0 and uly >= 0 and lry - uly >= 0:
+                    sane_detect_results.append(detect_result)
+        return sane_detect_results
+
+    def _dummy_for_group_generated(self, category_name: ObjectTypes) -> bool:
         for idx, group in enumerate(self.group_categories):
-            if category_id in group:
+            if category_name in group:
                 is_generated = self.dummy_for_group_generated[idx]
                 self.dummy_for_group_generated[idx] = True
                 return is_generated
@@ -176,10 +192,12 @@ class SubImageLayoutService(PipelineComponent):
         self.predictor = sub_image_detector
         super().__init__(self._get_name(sub_image_detector.name), self.predictor.model_id)
         if self.detect_result_generator is not None:
-            if self.detect_result_generator.categories != self.predictor.categories.get_categories():
+            if self.detect_result_generator.categories_name_as_key != self.predictor.categories.get_categories(
+                as_dict=True, name_as_key=True
+            ):
                 raise ValueError(
                     f"The categories of the 'detect_result_generator' must be the same as the categories of the "
-                    f"'sub_image_detector'. Got {self.detect_result_generator.categories} #"
+                    f"'sub_image_detector'. Got {self.detect_result_generator.categories_name_as_key} #"
                     f"and {self.predictor.categories.get_categories()}."
                 )
 

deepdoctection/train/hf_layoutlm_train.py

@@ -499,7 +499,9 @@ def train_hf_layoutlm(
         )
         pipeline_component_cls = pipeline_component_registry.get(pipeline_component_name)
         if dataset_type == DatasetType.SEQUENCE_CLASSIFICATION:
-            pipeline_component = pipeline_component_cls(tokenizer_fast, dd_model)
+            pipeline_component = pipeline_component_cls(tokenizer_fast,
+                                                        dd_model,
+                                                        use_other_as_default_category=True)
         else:
             pipeline_component = pipeline_component_cls(
                 tokenizer_fast,

deepdoctection/utils/pdf_utils.py

@@ -181,8 +181,6 @@ class PDFStreamer:
 
             streamer.close() # Do not forget to close the streamer, otherwise the file will never be closed and might
                              # cause memory leaks if you open many files.
-
-
     """
 
     def __init__(self, path_or_bytes: Union[PathLikeOrStr, bytes]) -> None:
@@ -223,7 +221,10 @@ class PDFStreamer:
 
 
 def _input_to_cli_str(
-    input_file_name: PathLikeOrStr, output_file_name: PathLikeOrStr, dpi: int, size: Optional[tuple[int, int]] = None
+    input_file_name: PathLikeOrStr,
+    output_file_name: PathLikeOrStr,
+    dpi: Optional[int] = None,
+    size: Optional[tuple[int, int]] = None,
 ) -> list[str]:
     cmd_args: list[str] = []
 
@@ -237,7 +238,10 @@ def _input_to_cli_str(
     if platform.system() == "Windows":
         command = command + ".exe"
     cmd_args.append(command)
-    cmd_args.extend(["-r", str(dpi), str(input_file_name)])
+
+    if dpi:
+        cmd_args.extend(["-r", str(dpi)])
+    cmd_args.append(str(input_file_name))
     cmd_args.append("-png")
     cmd_args.append(str(output_file_name))
 
@@ -275,7 +279,9 @@ def _run_poppler(poppler_args: list[str]) -> None:
             raise PopplerError(status=proc.returncode, message="Syntax Error: PDF cannot be read with Poppler")
 
 
-def pdf_to_np_array_poppler(pdf_bytes: bytes, size: Optional[tuple[int, int]] = None, dpi: int = 200) -> PixelValues:
+def pdf_to_np_array_poppler(
+    pdf_bytes: bytes, size: Optional[tuple[int, int]] = None, dpi: Optional[int] = None
+) -> PixelValues:
     """
     Convert a single pdf page from its byte representation to a numpy array. This function will save the pdf as to a tmp
     file and then call poppler via `pdftoppm` resp. `pdftocairo` if the former is not available.
@@ -285,7 +291,8 @@ def pdf_to_np_array_poppler(pdf_bytes: bytes, size: Optional[tuple[int, int]] =
     :param dpi:  Image quality in DPI/dots-per-inch (default 200)
     :return: numpy array
     """
-
+    if dpi is None and size is None:
+        raise ValueError("Either dpi or size must be provided.")
     with save_tmp_file(pdf_bytes, "pdf_") as (tmp_name, input_file_name):
         _run_poppler(_input_to_cli_str(input_file_name, tmp_name, dpi, size))
         image = viz_handler.read_image(tmp_name + "-1.png")
@@ -293,7 +300,7 @@ def pdf_to_np_array_poppler(pdf_bytes: bytes, size: Optional[tuple[int, int]] =
     return image.astype(uint8)
 
 
-def pdf_to_np_array_pdfmium(pdf_bytes: bytes, dpi: int = 200) -> PixelValues:
+def pdf_to_np_array_pdfmium(pdf_bytes: bytes, dpi: Optional[int] = None) -> PixelValues:
     """
     Convert a single pdf page from its byte representation to a numpy array using pdfium.
 
@@ -301,12 +308,13 @@ def pdf_to_np_array_pdfmium(pdf_bytes: bytes, dpi: int = 200) -> PixelValues:
     :param dpi:  Image quality in DPI/dots-per-inch (default 200)
     :return: numpy array
     """
-
+    if dpi is None:
+        raise ValueError("dpi must be provided.")
     page = pypdfium2.PdfDocument(pdf_bytes)[0]
     return page.render(scale=dpi * 1 / 72).to_numpy().astype(uint8)
 
 
-def pdf_to_np_array(pdf_bytes: bytes, size: Optional[tuple[int, int]] = None, dpi: int = 200) -> PixelValues:
+def pdf_to_np_array(pdf_bytes: bytes, size: Optional[tuple[int, int]] = None, dpi: Optional[int] = None) -> PixelValues:
     """
     Convert a single pdf page from its byte representation to a numpy array. This function will either use Poppler or
     pdfium to render the pdf.

{deepdoctection-0.37.3.dist-info → deepdoctection-0.39.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: deepdoctection
-Version: 0.37.3
+Version: 0.39
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -17,7 +17,7 @@ Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: catalogue==2.0.10
-Requires-Dist: huggingface_hub<0.26,>=0.12.0
+Requires-Dist: huggingface_hub>=0.26.0
 Requires-Dist: importlib-metadata>=5.0.0
 Requires-Dist: jsonlines==3.1.0
 Requires-Dist: lazy-imports==0.3.1
@@ -36,7 +36,7 @@ Requires-Dist: tabulate>=0.7.7
 Requires-Dist: tqdm==4.64.0
 Provides-Extra: tf
 Requires-Dist: catalogue==2.0.10; extra == "tf"
-Requires-Dist: huggingface_hub<0.26,>=0.12.0; extra == "tf"
+Requires-Dist: huggingface_hub>=0.26.0; extra == "tf"
 Requires-Dist: importlib-metadata>=5.0.0; extra == "tf"
 Requires-Dist: jsonlines==3.1.0; extra == "tf"
 Requires-Dist: lazy-imports==0.3.1; extra == "tf"
@@ -68,7 +68,7 @@ Requires-Dist: distance==0.1.3; extra == "tf"
 Requires-Dist: lxml>=4.9.1; extra == "tf"
 Provides-Extra: pt
 Requires-Dist: catalogue==2.0.10; extra == "pt"
-Requires-Dist: huggingface_hub<0.26,>=0.12.0; extra == "pt"
+Requires-Dist: huggingface_hub>=0.26.0; extra == "pt"
 Requires-Dist: importlib-metadata>=5.0.0; extra == "pt"
 Requires-Dist: jsonlines==3.1.0; extra == "pt"
 Requires-Dist: lazy-imports==0.3.1; extra == "pt"
@@ -127,6 +127,16 @@ Requires-Dist: types-urllib3>=1.26.25.14; extra == "dev"
 Provides-Extra: test
 Requires-Dist: pytest==8.0.2; extra == "test"
 Requires-Dist: pytest-cov; extra == "test"
+Dynamic: author
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
 
 
 <p align="center">

{deepdoctection-0.37.3.dist-info → deepdoctection-0.39.dist-info}/RECORD

@@ -1,11 +1,11 @@
-deepdoctection/__init__.py,sha256=SSW2Y9uos3Mnkihi-H-CmRF5V8nHUARbToRp83yWJB0,12655
+deepdoctection/__init__.py,sha256=0nxfBTu-aeg3DYu9g2kEAnt3Y-lCnHSgP31qvCnsLOs,12752
 deepdoctection/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 deepdoctection/analyzer/__init__.py,sha256=icClxrd20XutD6LxLgEPIWceSs4j_QfI3szCE-9BL2w,729
-deepdoctection/analyzer/_config.py,sha256=NZl_REM8Ge2xfxvHN-mZR5KURcHfZii3xfMlKQwckbA,4864
-deepdoctection/analyzer/dd.py,sha256=DUOhOtwipHw5nabYqn3WGR9aZcgP0ma_bi_tjf9xscw,5973
-deepdoctection/analyzer/factory.py,sha256=dEUOtdBS3yQGLqMqLR_kq5EYCR3IE30DjHNzE0spoQE,31519
+deepdoctection/analyzer/_config.py,sha256=OZMOPlyFv4gcyabPG6KO08EYx-0tUH82Ehs9YDv2B1Q,5027
+deepdoctection/analyzer/dd.py,sha256=bfR7e1JV7BwUNDRLu0jYZU7qQXnyA_vbRAJl2Ylrq5o,5905
+deepdoctection/analyzer/factory.py,sha256=Kf3Ztv5FEcF5yJf6i4I557aOIUHybuxIP0moHryguTQ,32344
 deepdoctection/configs/__init__.py,sha256=TX_P6tqDOF1LK1mi9ruAl7x0mtv1Asm8cYWCz3Pe2dk,646
-deepdoctection/configs/conf_dd_one.yaml,sha256=td7XsyVhdXkhh5Pie7sT_WNjGTaxBOWgpxhkobHd1H0,2325
+deepdoctection/configs/conf_dd_one.yaml,sha256=qnrDAST1PHBtdIKE_hdkZexW22FqVvNTI-PEo9wvinM,3025
 deepdoctection/configs/conf_tesseract.yaml,sha256=oF6szDyoi15FHvq7yFUNIEjfA_jNLhGxoowiRsz_zY4,35
 deepdoctection/dataflow/__init__.py,sha256=CWRHMpmJaPk4xY_oIIFubCt-z11SguWrMWxHZ7rdrvY,845
 deepdoctection/dataflow/base.py,sha256=z4DCComSj5wStEPjtk0093cNNGfUMiDqx8dqz36nS_o,6221
@@ -17,10 +17,10 @@ deepdoctection/dataflow/serialize.py,sha256=4pYC7m9h53JCu99waVeKpHDpsCDDdYCrSZpP
 deepdoctection/dataflow/stats.py,sha256=Bsr6v7lcesKXUYtO9wjqlzx_Yq_uyIF3Lel-tQ0i4wI,9619
 deepdoctection/datapoint/__init__.py,sha256=3K406GbOPhoEp8koVaSbMocmSsmWifnSZ1SPb7C1lOY,1643
 deepdoctection/datapoint/annotation.py,sha256=FEgz4COxVDfjic0gG7kS6iHnWLBIgFnquQ63Cbj2a4Y,22531
-deepdoctection/datapoint/box.py,sha256=tkFuVM6xfx2jL7W4UED4qHXV572LSRdIsVJbrEiyIxI,23524
-deepdoctection/datapoint/convert.py,sha256=O7920pIomyEkzXwxpFsrzfhn7Pl6UzVGhNzv90VcuKU,7099
-deepdoctection/datapoint/image.py,sha256=UDBKXJJpuKAUx0J-DGjvLGqrMV4N3kLpksJYxoVkong,33279
-deepdoctection/datapoint/view.py,sha256=1rVMuqucCrI5zlwyXMADJQBV38V_zSNFqFyBi3cMA1E,44914
+deepdoctection/datapoint/box.py,sha256=UAdSnLexvFyg4KK1u9kXdJxhaWTwRxTU-cnQcvl37Q8,23410
+deepdoctection/datapoint/convert.py,sha256=gJbHY2V8nlMht1N5VdxTmWSsOeydpFPTJsaJHp6XGgE,7516
+deepdoctection/datapoint/image.py,sha256=S6yfsIRQgMCl6HYAcHYJSBcbfdYKKtebtkEkkkrXsMQ,33619
+deepdoctection/datapoint/view.py,sha256=srMyPQGsK4OSiorxkyG6UAIgpViM6Ks1CI3b5k97cjY,49452
 deepdoctection/datasets/__init__.py,sha256=-A3aR90aDsHPmVM35JavfnQ2itYSCn3ujl4krRni1QU,1076
 deepdoctection/datasets/adapter.py,sha256=Ly_vbOAgVI73V41FUccnSX1ECTOyesW_qsuvQuvOZbw,7796
 deepdoctection/datasets/base.py,sha256=DT4i-d74sIEiUNC6UspIHNJuHSK0t1dBv7qwadg4rLw,22341
@@ -55,10 +55,10 @@ deepdoctection/extern/d2detect.py,sha256=zrKv1yurApnjD7QZIZk_8LYCahjmN82MQUjHjv8
 deepdoctection/extern/deskew.py,sha256=sPoixu8S9he-0wbs-jgxtPE2V9BiP4-3uZlb6F5Y1SA,3077
 deepdoctection/extern/doctrocr.py,sha256=T3_tvlih22_dVCBZypS1Y8tjQQB1fkAxIbGdUGHIapQ,24473
 deepdoctection/extern/fastlang.py,sha256=F4gK-SEwcCujjxH327ZDzMGWToJ49xS_dCKcePQ9IlY,4780
-deepdoctection/extern/hfdetr.py,sha256=1NPW_u5eH2tP3ixZ91l4WR-O-wLVcrFsLWA7BqID0oM,12055
+deepdoctection/extern/hfdetr.py,sha256=JzHrrTyzS9qh6T2TsvKboAGZkIhno2txmSoLQ5Vd-lo,12077
 deepdoctection/extern/hflayoutlm.py,sha256=tFaf90FRbZzhSycdp8rGkeiPywQa6UcTEEwbayIXkr0,57023
 deepdoctection/extern/hflm.py,sha256=kwS6kcSlY_2m9u0RzBLTRq-UMM7c1PhyUaDTvSdejus,9217
-deepdoctection/extern/model.py,sha256=ViHHKPvbGmLCPw7ZESv_rmjlkA90UiBU6oZiHOMqNSw,59869
+deepdoctection/extern/model.py,sha256=lbVwDa3vD6VwCD_dsozcI8b4xDZs4KJ1628SxaDdtHQ,55378
 deepdoctection/extern/pdftext.py,sha256=KS_t27SUiYn_IOS_J2lF9lSSo22vLagxmxvYCY3CqXA,7228
 deepdoctection/extern/tessocr.py,sha256=tG7etMvZ-jHFdq-jJAHYMJii3ujDjMfAFYUsjBp3nKI,17444
 deepdoctection/extern/texocr.py,sha256=yMt5ZzKtsjd7ogrcNXba7zccGGGF9LXK194EtER6YNQ,5804
@@ -88,17 +88,17 @@ 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=EsYdUw8LAPsyqAfGfhNa6gAPVpUhP8GGOchSIKto_e0,15741
+deepdoctection/mapper/cats.py,sha256=s73JzONV2UQ71szfljurk7H1-UjDBWsW4oNLs5xePUk,16474
 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
 deepdoctection/mapper/laylmstruct.py,sha256=abMZkYU2W0e_VcCm_c0ZXNFuv-lfMFWcTedcZS5EYvE,42935
 deepdoctection/mapper/maputils.py,sha256=eI6ZcDg9W5uB6xQNBZpMIdEd86HlCxTtkJuyROdTqiw,8146
-deepdoctection/mapper/match.py,sha256=pCWZpz2R8JahiKXCw7dxKRTLiPgJXeVDgkddDPLy_c0,9643
-deepdoctection/mapper/misc.py,sha256=NLSSgk066Tkrrdi075HkqV7cP-iqT9fv_MtyAJ-8gOg,6743
+deepdoctection/mapper/match.py,sha256=Ed9FsuVPNp_faaW5PKnvUHZoEXcRcrO-muduTMzjp1s,8937
+deepdoctection/mapper/misc.py,sha256=vX-fV420Te00eD-cqTiWBV2twHqdBcBV2_7rAFRgPRg,7164
 deepdoctection/mapper/pascalstruct.py,sha256=TzVU1p0oiw0nOuxTFFbEB9vXJxH1v6VUvTJ7MD0manU,3828
 deepdoctection/mapper/prodigystruct.py,sha256=Re4Sd_zAp6qOvbXZLmMJeG0IGEfMQxebuyDeZgMcTa8,6827
-deepdoctection/mapper/pubstruct.py,sha256=YxsrZ-E0pD45Mm_VCPQB9yEgHsTPkw4htt-3DwCRX1k,23361
+deepdoctection/mapper/pubstruct.py,sha256=PAJ2N1HSPNS6F2ZrIwlD7PiBhIM-rJscK_Ti8OR_IGs,23370
 deepdoctection/mapper/tpstruct.py,sha256=YNABRibvcISD5Lavg3jouoE4FMdqXEJoM-hNoB_rnww,4481
 deepdoctection/mapper/xfundstruct.py,sha256=_3r3c0K82fnF2h1HxA85h-9ETYrHwcERa6MNc6Ko6Z8,8807
 deepdoctection/pipe/__init__.py,sha256=ywTVoetftdL6plXg2YlBzMfmqBZupq7yXblSVyvvkcQ,1127
@@ -106,21 +106,21 @@ deepdoctection/pipe/anngen.py,sha256=3319l4aaXzcY4w6ItVBNPX8LGS5fHFDVtyVY9KMefac
 deepdoctection/pipe/base.py,sha256=ynNg5SSRuUVxN69VWOO3Oi7WSeGrYwn3A56NQMBJDvw,14222
 deepdoctection/pipe/common.py,sha256=haOb4v0jLX3r41BSC8cVseX2E320_HkSrGlZsQiKE2g,17728
 deepdoctection/pipe/concurrency.py,sha256=AAKRsVgaBEYNluntbDa46SBF1JZ_XqnWLDSWrNvAzEo,9657
-deepdoctection/pipe/doctectionpipe.py,sha256=wCg96P9Pb54i5AVgG02b4FljobM64_qEML_GxiULy-4,11765
+deepdoctection/pipe/doctectionpipe.py,sha256=bGW3ugky-fb-nEe-3bvO6Oc_4_6w82cQboGM_6p2eIo,12530
 deepdoctection/pipe/language.py,sha256=5zI0UQC6Fh12_r2pfVL42HoCGz2hpHrOhpXAn5m-rYw,5451
 deepdoctection/pipe/layout.py,sha256=xIhnJpyUSbvLbhTXyAKXY1hmG9352jihGYFSclTH_1g,5567
-deepdoctection/pipe/lm.py,sha256=tLuCtML-S_kTEYcDAEtM3NBYmR7Aovv9p5TcXYL_AAg,16693
+deepdoctection/pipe/lm.py,sha256=Ygj6MmBvBZ7l4RGCwBuhmMsOM0Ep3LWteNg7bzh-UmI,17703
 deepdoctection/pipe/order.py,sha256=PnJZiCnxFluJiECXLTZT0c1Rr66vIRBFraa_G41UA2k,40121
-deepdoctection/pipe/refine.py,sha256=RjaOv5dvVxVrMm_ze-yeAqc46Be2I--7UARLezYbNxA,22250
+deepdoctection/pipe/refine.py,sha256=dTfI396xydPdbzpfo4yqFcuxl3UAB1y-WbSQn1o76ec,22367
 deepdoctection/pipe/registry.py,sha256=aFx-Tn0xhVA5l5H18duNW5QoTNKQltybsEUEzsMgUfg,902
-deepdoctection/pipe/segment.py,sha256=R-7TfQcBYSPljnQwZ2Sq0zr2CevF-_kxAfgKMA-kRH8,50144
-deepdoctection/pipe/sub_layout.py,sha256=XoERGwrJIQ8WcNb1Mba4ulmUoWMSRF1OiyGEDG4PbVM,12654
+deepdoctection/pipe/segment.py,sha256=CR83HQMW0hrRG8W6pFuB0YibxQMWpqI7_LaUIcJcQwo,59116
+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
 deepdoctection/train/__init__.py,sha256=YFTRAZF1F7cEAKTdAIi1BLyYb6rSRcwq09Ui5Lu8d6E,1071
 deepdoctection/train/d2_frcnn_train.py,sha256=sFc_G-mEpaM8d1CCE0_6Gl4nBh11X2RYRBA3p_ylFJQ,16000
 deepdoctection/train/hf_detr_train.py,sha256=8ydysxzOPE_IPoNFGaHb7PbKr9Nbl41rcY4lbylQavU,10783
-deepdoctection/train/hf_layoutlm_train.py,sha256=BNjPgPAvxm4beHULqzo58u-gW7GcTGiZAk2rF6TootM,22532
+deepdoctection/train/hf_layoutlm_train.py,sha256=irSg-IpbVoSlaw1-vZCej2mCZcctONtXr5Z2NQAc_a4,22680
 deepdoctection/train/tp_frcnn_train.py,sha256=pEpXokSVGveqo82pRnhnAmHPmjQ_8wQWpqM4ZyNHJgs,13049
 deepdoctection/utils/__init__.py,sha256=brBceRWeov9WXMiJTjyJOF2rHMP8trGGRRjhMdZ61nI,2371
 deepdoctection/utils/concurrency.py,sha256=nIhpkSncmv0LBB8PtcOLY-BsRGlfcDpz7foVdgzZd20,4598
@@ -134,15 +134,15 @@ deepdoctection/utils/identifier.py,sha256=QkNaGGqPynHwDPnd3_m8iur4Cv64rcQa7qolCE
 deepdoctection/utils/logger.py,sha256=J0OVKiXP_2A82MWbbJoOeMEJ-75aZu5npgaS_yI6mVA,10003
 deepdoctection/utils/metacfg.py,sha256=hD76KQ_RnD_5B02qLI2Zxf3WfnsnXhEI_KUTKpw91RI,5711
 deepdoctection/utils/mocks.py,sha256=IkN3-IzAl4eX0ibgKIHg8IY7ykVw6BnpF6XnxKnKaZI,2389
-deepdoctection/utils/pdf_utils.py,sha256=G0m8kUn2HwwyZWH_BcrDkm-m3MP9GN9SWHj5VhB7swY,12845
+deepdoctection/utils/pdf_utils.py,sha256=Fi0eZ2GbnO7N61Rd8b8YRKRff4dalHAzkcn3zpGPoic,13119
 deepdoctection/utils/settings.py,sha256=k6OyuWbj-IPeaO9zT9RZ-5Yad1wNhWGYqGLZdtgXAZY,12464
 deepdoctection/utils/tqdm.py,sha256=cBUtR0L1x0KMeYrLP2rrzyzCamCjpQAKroHXLv81_pk,1820
 deepdoctection/utils/transform.py,sha256=3kCgsEeRkG1efCdkfvj7tUFMs-e2jbjbflq826F2GPU,8502
 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-0.37.3.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
-deepdoctection-0.37.3.dist-info/METADATA,sha256=aD8ZORAl_RBxZXaL0uojCx-bs8Xe3x9ocLSpvXn5nRc,19545
-deepdoctection-0.37.3.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
-deepdoctection-0.37.3.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
-deepdoctection-0.37.3.dist-info/RECORD,,
+deepdoctection-0.39.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
+deepdoctection-0.39.dist-info/METADATA,sha256=0OcWmWb8bssiwC2_Xnb6hUKyY0ISv6Bc5qVGosbrn3c,19741
+deepdoctection-0.39.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+deepdoctection-0.39.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
+deepdoctection-0.39.dist-info/RECORD,,

{deepdoctection-0.37.3.dist-info → deepdoctection-0.39.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.6.0)
+Generator: setuptools (75.8.0)
 Root-Is-Purelib: true
 Tag: py3-none-any