deepdoctection 0.34__py3-none-any.whl → 0.35__py3-none-any.whl

deepdoctection/extern/pdftext.py

@@ -24,21 +24,25 @@ from typing import Optional
 from lazy_imports import try_import
 
 from ..utils.context import save_tmp_file
-from ..utils.file_utils import get_pdfplumber_requirement
+from ..utils.file_utils import get_pdfplumber_requirement, get_pypdfium2_requirement
 from ..utils.settings import LayoutType, ObjectTypes
 from ..utils.types import Requirement
 from .base import DetectionResult, ModelCategories, PdfMiner
 
-with try_import() as import_guard:
+with try_import() as pdfplumber_import_guard:
     from pdfplumber.pdf import PDF, Page
 
+with try_import() as pypdfmium_import_guard:
+    import pypdfium2.raw as pypdfium_c
+    from pypdfium2 import PdfDocument
 
-def _to_detect_result(word: dict[str, str]) -> DetectionResult:
+
+def _to_detect_result(word: dict[str, str], class_name: ObjectTypes) -> DetectionResult:
     return DetectionResult(
         box=[float(word["x0"]), float(word["top"]), float(word["x1"]), float(word["bottom"])],
         class_id=1,
         text=word["text"],
-        class_name=LayoutType.WORD,
+        class_name=class_name,
     )
 
 
@@ -49,6 +53,7 @@ class PdfPlumberTextDetector(PdfMiner):
 
         pdf_plumber = PdfPlumberTextDetector()
         df = SerializerPdfDoc.load("path/to/document.pdf")
+        df.reset_state()
 
         for dp in df:
             detection_results = pdf_plumber.predict(dp["pdf_bytes"])
@@ -61,6 +66,8 @@ class PdfPlumberTextDetector(PdfMiner):
         pipe = DoctectionPipe([text_extract])
 
         df = pipe.analyze(path="path/to/document.pdf")
+        df.reset_state()
+
         for dp in df:
             ...
 
@@ -87,7 +94,7 @@ class PdfPlumberTextDetector(PdfMiner):
                 self._page = PDF(fin).pages[0]
                 self._pdf_bytes = pdf_bytes
                 words = self._page.extract_words(x_tolerance=self.x_tolerance, y_tolerance=self.y_tolerance)
-        detect_results = list(map(_to_detect_result, words))
+        detect_results = [_to_detect_result(word, self.get_category_names()[0]) for word in words]
         return detect_results
 
     @classmethod
@@ -113,3 +120,87 @@ class PdfPlumberTextDetector(PdfMiner):
 
     def get_category_names(self) -> tuple[ObjectTypes, ...]:
         return self.categories.get_categories(as_dict=False)
+
+
+class Pdfmium2TextDetector(PdfMiner):
+    """
+    Text miner based on the pypdfium2 engine. It will return text on text line level and not on word level
+
+        pdfmium2 = Pdfmium2TextDetector()
+        df = SerializerPdfDoc.load("path/to/document.pdf")
+        df.reset_state()
+
+        for dp in df:
+            detection_results = pdfmium2.predict(dp["pdf_bytes"])
+
+    To use it in a more integrated way:
+
+        pdfmium2 = Pdfmium2TextDetector()
+        text_extract = TextExtractionService(pdfmium2)
+
+        pipe = DoctectionPipe([text_extract])
+
+        df = pipe.analyze(path="path/to/document.pdf")
+        df.reset_state()
+        for dp in df:
+            ...
+
+    """
+
+    def __init__(self) -> None:
+        self.name = "Pdfmium"
+        self.model_id = self.get_model_id()
+        self.categories = ModelCategories(init_categories={1: LayoutType.LINE})
+        self._page: Optional[Page] = None
+
+    def predict(self, pdf_bytes: bytes) -> list[DetectionResult]:
+        """
+        Call pypdfium2 and returns detected text as detection results
+
+        :param pdf_bytes: bytes of a single pdf page
+        :return: A list of DetectionResult
+        """
+
+        pdf = PdfDocument(pdf_bytes)
+        page = pdf.get_page(0)
+        text = page.get_textpage()
+        words = []
+        height = page.get_height()
+        for obj in page.get_objects((pypdfium_c.FPDF_PAGEOBJ_TEXT,)):
+            box = obj.get_pos()
+            if all(x > 0 for x in box):
+                words.append(
+                    {
+                        "text": text.get_text_bounded(*box),
+                        "x0": box[0],
+                        "x1": box[2],
+                        "top": height - box[3],
+                        "bottom": height - box[1],
+                    }
+                )
+        detect_results = [_to_detect_result(word, self.get_category_names()[0]) for word in words]
+        return detect_results
+
+    @classmethod
+    def get_requirements(cls) -> list[Requirement]:
+        return [get_pypdfium2_requirement()]
+
+    def get_width_height(self, pdf_bytes: bytes) -> tuple[float, float]:
+        """
+        Get the width and height of the full page
+        :param pdf_bytes: pdf_bytes generating the pdf
+        :return: width and height
+        """
+
+        if self._pdf_bytes == pdf_bytes and self._page is not None:
+            return self._page.bbox[2], self._page.bbox[3]  # pylint: disable=E1101
+        # if the pdf bytes is not equal to the cached pdf, will recalculate values
+        pdf = PdfDocument(pdf_bytes)
+        self._page = pdf.get_page(0)
+        self._pdf_bytes = pdf_bytes
+        if self._page is not None:
+            return self._page.get_width(), self._page.get_height()  # type: ignore
+        raise ValueError("Page not found")
+
+    def get_category_names(self) -> tuple[ObjectTypes, ...]:
+        return self.categories.get_categories(as_dict=False)

deepdoctection/extern/tessocr.py

@@ -421,6 +421,7 @@ class TesseractRotationTransformer(ImageTransformer):
     def __init__(self) -> None:
         self.name = fspath(_TESS_PATH) + "-rotation"
         self.categories = ModelCategories(init_categories={1: PageType.ANGLE})
+        self.model_id = self.get_model_id()
 
     def transform(self, np_img: PixelValues, specification: DetectionResult) -> PixelValues:
         """

deepdoctection/utils/env_info.py

@@ -20,6 +20,10 @@ Some useful function for collecting environment information.
 
 This is also the place where we give an overview of the important environment variables.
 
+For env variables with boolean character, use one of the following values:
+
+{"1", "True", "TRUE", "true", "yes"}
+
 `USE_TENSORFLOW
 USE_PYTORCH
 USE_CUDA
@@ -35,6 +39,12 @@ decide what image processing library the `viz_handler` should use. The default l
 to be installed separately. However, if both libraries have been detected `viz_handler` will opt for OpenCV.
 Use the variables to let choose `viz_handler` according to your preferences.
 
+`USE_DD_POPPLER
+USE_DD_PDFIUM`
+
+For PDF rendering we use PyPDFium2 as default but for legacy reasons, we also support Poppler. If you want to enforce
+Poppler set one to `USE_DD_POPPLER=True` and `USE_DD_PDFIUM=False` the other to False.
+
 `HF_CREDENTIALS`
 
 will be used by the `ModelDownloadManager` to pass your credentials if you have a model registered that resides in a
@@ -56,6 +66,7 @@ from typing import Optional
 
 import numpy as np
 from packaging import version
+from pypdf.errors import DependencyError
 from tabulate import tabulate
 
 from .file_utils import (
@@ -75,6 +86,7 @@ from .file_utils import (
     pdf_to_cairo_available,
     pdf_to_ppm_available,
     pdfplumber_available,
+    pypdfium2_available,
     pytorch_available,
     qpdf_available,
     scipy_available,
@@ -88,7 +100,7 @@ from .file_utils import (
 from .logger import LoggingRecord, logger
 from .types import KeyValEnvInfos, PathLikeOrStr
 
-__all__ = ["collect_env_info", "auto_select_viz_library", "ENV_VARS_TRUE"]
+__all__ = ["collect_env_info", "auto_select_viz_library", "auto_select_pdf_render_framework", "ENV_VARS_TRUE"]
 
 # pylint: disable=import-outside-toplevel
 
@@ -532,4 +544,21 @@ def auto_select_viz_library() -> None:
         os.environ["USE_DD_OPENCV"] = "False"
 
 
+def auto_select_pdf_render_framework() -> None:
+    """Setting pdf2image as default pdf rendering library if pdfium is not installed"""
+
+    # if env variables are already set, don't change them
+    if os.environ.get("USE_DD_POPPLER") or os.environ.get("USE_DD_PDFIUM"):
+        return
+    if pypdfium2_available():
+        os.environ["USE_DD_POPPLER"] = "False"
+        os.environ["USE_DD_PDFIUM"] = "True"
+        return
+    if pdf_to_cairo_available() or pdf_to_ppm_available():
+        os.environ["USE_DD_POPPLER"] = "True"
+        os.environ["USE_DD_PDFIUM"] = "False"
+        return
+    raise DependencyError("No pdf rendering library found. Please install Poppler or pdfium.")
+
+
 # pylint: enable=import-outside-toplevel

deepdoctection/utils/file_utils.py

@@ -616,6 +616,25 @@ def get_pillow_requirement() -> Requirement:
     return "pillow", pillow_available(), _PILLOW_ERR_MSG
 
 
+# Pypdfium2
+_PYPDFIUM2_AVAILABLE = importlib.util.find_spec("pypdfium2") is not None
+_PYPDFIUM2_ERR_MSG = f"pypdfium2 must be installed. {_GENERIC_ERR_MSG}"
+
+
+def pypdfium2_available() -> bool:
+    """
+    Returns True if pypdfium2 is installed
+    """
+    return bool(_PYPDFIUM2_AVAILABLE)
+
+
+def get_pypdfium2_requirement() -> Requirement:
+    """
+    Return pypdfium2 requirement
+    """
+    return "pypdfium2", pypdfium2_available(), _PYPDFIUM2_ERR_MSG
+
+
 # SpaCy
 _SPACY_AVAILABLE = importlib.util.find_spec("spacy") is not None
 _SPACY_ERR_MSG = f"SpaCy must be installed. {_GENERIC_ERR_MSG}"

deepdoctection/utils/metacfg.py

@@ -18,6 +18,7 @@
 """
 Class AttrDict for maintaining configs and some functions for generating and saving AttrDict instances to .yaml files
 """
+from __future__ import annotations
 
 import pprint
 from typing import Any
@@ -105,6 +106,17 @@ class AttrDict:
                 v = eval(v)  # pylint: disable=C0103, W0123
             setattr(dic, key, v)
 
+    def overwrite_config(self, other_config: AttrDict) -> None:
+        """
+        Overwrite the current config with values from another config.
+
+        :param other_config: The other AttrDict instance to copy values from.
+        :raises AttributeError: If a key from other_config is not an attribute of self.
+        """
+        if self._freezed:
+            raise AttributeError("Config was freezed! Cannot overwrite config.")
+        self.from_dict(other_config.to_dict())
+
     def freeze(self, freezed: bool = True) -> None:
         """
         :param freezed: freeze the instance, so that no attributes can be added or changed

deepdoctection/utils/pdf_utils.py

@@ -24,13 +24,16 @@ import subprocess
 import sys
 from errno import ENOENT
 from io import BytesIO
+from pathlib import Path
 from shutil import copyfile
-from typing import Generator, Optional
+from typing import Generator, Literal, Optional
 
+from lazy_imports import try_import
 from numpy import uint8
 from pypdf import PdfReader, PdfWriter, errors
 
 from .context import save_tmp_file, timeout_manager
+from .env_info import ENV_VARS_TRUE
 from .error import DependencyError, FileExtensionError
 from .file_utils import pdf_to_cairo_available, pdf_to_ppm_available, qpdf_available
 from .logger import LoggingRecord, logger
@@ -38,7 +41,17 @@ from .types import PathLikeOrStr, PixelValues
 from .utils import is_file_extension
 from .viz import viz_handler
 
-__all__ = ["decrypt_pdf_document", "get_pdf_file_reader", "get_pdf_file_writer", "PDFStreamer", "pdf_to_np_array"]
+with try_import() as pt_import_guard:
+    import pypdfium2
+
+__all__ = [
+    "decrypt_pdf_document",
+    "get_pdf_file_reader",
+    "get_pdf_file_writer",
+    "PDFStreamer",
+    "pdf_to_np_array",
+    "split_pdf",
+]
 
 
 def decrypt_pdf_document(path: PathLikeOrStr) -> bool:
@@ -234,7 +247,7 @@ 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(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: int = 200) -> 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.
@@ -250,3 +263,73 @@ def pdf_to_np_array(pdf_bytes: bytes, size: Optional[tuple[int, int]] = None, dp
         image = viz_handler.read_image(tmp_name + "-1.png")
 
     return image.astype(uint8)
+
+
+def pdf_to_np_array_pdfmium(pdf_bytes: bytes, dpi: int = 200) -> PixelValues:
+    """
+    Convert a single pdf page from its byte representation to a numpy array using pdfium.
+
+    :param pdf_bytes: Bytes representing the PDF file
+    :param dpi:  Image quality in DPI/dots-per-inch (default 200)
+    :return: numpy array
+    """
+
+    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:
+    """
+    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.
+
+    :param pdf_bytes: Bytes representing the PDF file
+    :param size: Size of the resulting image(s), uses (width, height) standard
+    :param dpi:  Image quality in DPI/dots-per-inch (default 200)
+    :return: numpy array
+    """
+    if os.environ.get("USE_DD_PDFIUM", "False") in ENV_VARS_TRUE:
+        if size is not None:
+            logger.warning(
+                LoggingRecord(
+                    f"pdf_to_np_array_pdfmium does not support the size parameter. Will use dpi = {dpi} instead."
+                )
+            )
+        return pdf_to_np_array_pdfmium(pdf_bytes, dpi)
+    return pdf_to_np_array_poppler(pdf_bytes, size, dpi)
+
+
+def split_pdf(
+    pdf_path: PathLikeOrStr, output_dir: PathLikeOrStr, file_type: Literal["image", "pdf"], dpi: int = 200
+) -> None:
+    """
+    Split a pdf into single pages. The pages are saved as single pdf/png files in a subfolder of the output directory.
+
+    :param pdf_path: Path to the pdf file
+    :param output_dir: Path to the output directory
+    :param file_type: Type of the output file. Either "image" or "pdf"
+    :param dpi: Image quality in DPI/dots-per-inch (default
+    """
+    pdf_path = Path(pdf_path)
+    filename = pdf_path.stem
+    output_dir = Path(output_dir)
+    file_dir = output_dir / filename
+    if not file_dir.exists():
+        os.makedirs(file_dir)
+
+    with open(pdf_path, "rb") as file:
+        pdf = PdfReader(file)
+        for i, page in enumerate(pdf.pages):
+            writer = PdfWriter()
+            writer.add_page(page)
+            if file_type == ".pdf":
+                with open(file_dir / f"{filename}_{i}.pdf", "wb") as out:
+                    writer.write(out)
+                    writer.close()
+            else:
+                with BytesIO() as buffer:
+                    writer.write(buffer)
+                    buffer.seek(0)
+                    np_image = pdf_to_np_array(buffer.getvalue(), dpi=dpi)
+                    viz_handler.write_image(file_dir / f"{filename}_{i}.png", np_image)
+                    writer.close()

{deepdoctection-0.34.dist-info → deepdoctection-0.35.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: deepdoctection
-Version: 0.34
+Version: 0.35
 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 >=0.12.0
+Requires-Dist: huggingface-hub <0.26,>=0.12.0
 Requires-Dist: importlib-metadata >=5.0.0
 Requires-Dist: jsonlines ==3.1.0
 Requires-Dist: lazy-imports ==0.3.1
@@ -27,6 +27,7 @@ Requires-Dist: numpy <2.0,>=1.21
 Requires-Dist: packaging >=20.0
 Requires-Dist: Pillow >=10.0.0
 Requires-Dist: pypdf >=3.16.0
+Requires-Dist: pypdfium2 >=4.30.0
 Requires-Dist: pyyaml >=6.0.1
 Requires-Dist: pyzmq >=16
 Requires-Dist: scipy >=1.13.1
@@ -63,7 +64,7 @@ Requires-Dist: mkdocstrings-python ; extra == 'docs'
 Requires-Dist: griffe ==0.25.0 ; extra == 'docs'
 Provides-Extra: pt
 Requires-Dist: catalogue ==2.0.10 ; extra == 'pt'
-Requires-Dist: huggingface-hub >=0.12.0 ; extra == 'pt'
+Requires-Dist: huggingface-hub <0.26,>=0.12.0 ; 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'
@@ -73,6 +74,7 @@ Requires-Dist: numpy <2.0,>=1.21 ; extra == 'pt'
 Requires-Dist: packaging >=20.0 ; extra == 'pt'
 Requires-Dist: Pillow >=10.0.0 ; extra == 'pt'
 Requires-Dist: pypdf >=3.16.0 ; extra == 'pt'
+Requires-Dist: pypdfium2 >=4.30.0 ; extra == 'pt'
 Requires-Dist: pyyaml >=6.0.1 ; extra == 'pt'
 Requires-Dist: pyzmq >=16 ; extra == 'pt'
 Requires-Dist: scipy >=1.13.1 ; extra == 'pt'
@@ -95,7 +97,7 @@ Requires-Dist: pytest ==8.0.2 ; extra == 'test'
 Requires-Dist: pytest-cov ; extra == 'test'
 Provides-Extra: tf
 Requires-Dist: catalogue ==2.0.10 ; extra == 'tf'
-Requires-Dist: huggingface-hub >=0.12.0 ; extra == 'tf'
+Requires-Dist: huggingface-hub <0.26,>=0.12.0 ; 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'
@@ -105,6 +107,7 @@ Requires-Dist: numpy <2.0,>=1.21 ; extra == 'tf'
 Requires-Dist: packaging >=20.0 ; extra == 'tf'
 Requires-Dist: Pillow >=10.0.0 ; extra == 'tf'
 Requires-Dist: pypdf >=3.16.0 ; extra == 'tf'
+Requires-Dist: pypdfium2 >=4.30.0 ; extra == 'tf'
 Requires-Dist: pyyaml >=6.0.1 ; extra == 'tf'
 Requires-Dist: pyzmq >=16 ; extra == 'tf'
 Requires-Dist: scipy >=1.13.1 ; extra == 'tf'
@@ -172,9 +175,9 @@ pipelines. Its core function does not depend on any specific deep learning libra
  - Document layout analysis and table recognition now runs with 
    [**Torchscript**](https://pytorch.org/docs/stable/jit.html) (CPU) as well and [**Detectron2**](https://github.com/facebookresearch/detectron2/tree/main/detectron2) is not required 
    anymore for basic inference.
- - [**new**] More angle predictors for determining the rotation of a document based on [**Tesseract**](https://github.com/tesseract-ocr/tesseract) and [**DocTr**](https://github.com/mindee/doctr)
+ - More angle predictors for determining the rotation of a document based on [**Tesseract**](https://github.com/tesseract-ocr/tesseract) and [**DocTr**](https://github.com/mindee/doctr)
    (not contained in the built-in Analyzer).
- - [**new**] Token classification with [**LiLT**](https://github.com/jpWang/LiLT) via 
+ - Token classification with [**LiLT**](https://github.com/jpWang/LiLT) via 
    [**transformers**](https://github.com/huggingface/transformers). 
    We have added a model wrapper for token classification with LiLT and added a some LiLT models to the model catalog 
    that seem to look promising, especially if you want to train a model on non-english data. The training script for 
@@ -263,7 +266,7 @@ documentation.
 
 ## Requirements
 
-![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/requirements_deepdoctection.png)
+![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/requirements_deepdoctection_081124.png)
 
 Everything in the overview listed below the **deep**doctection layer are necessary requirements and have to be installed 
 separately. 
@@ -272,13 +275,16 @@ separately.
 - Python >= 3.9
 - 1.13 <= PyTorch  **or** 2.11 <= Tensorflow < 2.16. (For lower Tensorflow versions the code will only run on a GPU).
 In general, if you want to train or fine-tune models, a GPU is required.
-- **deep**doctection uses Python wrappers for [Poppler](https://poppler.freedesktop.org/) to convert PDF documents into 
-images. 
+
 - With respect to the Deep Learning framework, you must decide between [Tensorflow](https://www.tensorflow.org/install?hl=en)
   and [PyTorch](https://pytorch.org/get-started/locally/).
 - [Tesseract](https://github.com/tesseract-ocr/tesseract) OCR engine will be used through a Python wrapper. The core 
   engine has to be installed separately.
 
+
+- For release `v.0.34.0` and below **deep**doctection uses Python wrappers for [Poppler](https://poppler.freedesktop.org/) to convert PDF
+  documents into images. For release `v.0.35.0` this dependency will be optional.
+
 The following overview shows the availability of the models in conjunction with the DL framework.
 
 | Task                                          | PyTorch | Torchscript    |  Tensorflow  |
@@ -396,8 +402,8 @@ to develop this framework.
 ## Problems
 
 We try hard to eliminate bugs. We also know that the code is not free of issues. We welcome all issues relevant to this
-repo and try to address them as quickly as possible. Bug fixes or enhancements will be deployed in a new release every 4 
-to 6 weeks.
+repo and try to address them as quickly as possible. Bug fixes or enhancements will be deployed in a new release every 10 
+to 12 weeks.
 
 ## If you like **deep**doctection ...
  

{deepdoctection-0.34.dist-info → deepdoctection-0.35.dist-info}/RECORD

@@ -1,9 +1,11 @@
-deepdoctection/__init__.py,sha256=lgfD5PlxwSqTwMnKBtcpzKH9emJ4UtyWaWrpM9Pn0Ng,12596
+deepdoctection/__init__.py,sha256=RZpawNRTJPKNPFuONawVOsYWdr-rI8PPNXZhlPtOKtc,12580
 deepdoctection/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-deepdoctection/analyzer/__init__.py,sha256=g86MeZz_BIQ2-b8kDIss7osPUNrFhT-Z3Eu7Wm02pFI,706
-deepdoctection/analyzer/dd.py,sha256=j3G6PFmXe9XBTwtu8-g9D3yAx7obaNzfZ2yl7rEOUqg,20234
+deepdoctection/analyzer/__init__.py,sha256=icClxrd20XutD6LxLgEPIWceSs4j_QfI3szCE-9BL2w,729
+deepdoctection/analyzer/_config.py,sha256=0cWtaI2e3jHNhufHZAqMje0YTTDAogKAHVl4VpYojAo,4874
+deepdoctection/analyzer/dd.py,sha256=DUOhOtwipHw5nabYqn3WGR9aZcgP0ma_bi_tjf9xscw,5973
+deepdoctection/analyzer/factory.py,sha256=T9jxtVLNFhocbsfWIGLPfFrEv21zQJzM6VdFt0yxMyg,23849
 deepdoctection/configs/__init__.py,sha256=TX_P6tqDOF1LK1mi9ruAl7x0mtv1Asm8cYWCz3Pe2dk,646
-deepdoctection/configs/conf_dd_one.yaml,sha256=d4ZTMQ1oTIYMFctQAaQBKK6iQP4LsViUDrPvsnaLumo,2220
+deepdoctection/configs/conf_dd_one.yaml,sha256=orP-oeqtWbz5S9FJZJKxy1UqMwOYjL9g0DOX-wbamqU,2239
 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
@@ -14,11 +16,11 @@ deepdoctection/dataflow/parallel_map.py,sha256=8FhxJBWV-kjJrJ27jQtP3yYF6Ev6rz98w
 deepdoctection/dataflow/serialize.py,sha256=4pYC7m9h53JCu99waVeKpHDpsCDDdYCrSZpP2QYSsgs,4555
 deepdoctection/dataflow/stats.py,sha256=Bsr6v7lcesKXUYtO9wjqlzx_Yq_uyIF3Lel-tQ0i4wI,9619
 deepdoctection/datapoint/__init__.py,sha256=3K406GbOPhoEp8koVaSbMocmSsmWifnSZ1SPb7C1lOY,1643
-deepdoctection/datapoint/annotation.py,sha256=3hDwNf3bm7qi0xnvfKn459hxZe4BdiLPiFt03hJBbUQ,22517
+deepdoctection/datapoint/annotation.py,sha256=FEgz4COxVDfjic0gG7kS6iHnWLBIgFnquQ63Cbj2a4Y,22531
 deepdoctection/datapoint/box.py,sha256=tkFuVM6xfx2jL7W4UED4qHXV572LSRdIsVJbrEiyIxI,23524
-deepdoctection/datapoint/convert.py,sha256=9L3YS89nGPAV8dqPZ-KOLVxvatj_zax2yP5RD-fuZCU,6718
-deepdoctection/datapoint/image.py,sha256=WyGcVYNrC-sv7bxODunEttxcQCFhplpWkWLLVQ266C0,32766
-deepdoctection/datapoint/view.py,sha256=Hdz67F8UtIkQjFW6U2mKeQ1WIdaXq4dOx95ymyQFLiU,42137
+deepdoctection/datapoint/convert.py,sha256=Gw2IjNiEotPu1yuMZqrIYB0mCAwafKt-VgMnrHj6S7U,6808
+deepdoctection/datapoint/image.py,sha256=EvZlVwJjMAcL1z8RNPBvZ8fwdJvkGuGpcFxCP1y26Go,33045
+deepdoctection/datapoint/view.py,sha256=7qSX4DQw9OPQQSKfSjV8e5i6jLyu6hOMceSKJAob2N8,42154
 deepdoctection/datasets/__init__.py,sha256=-A3aR90aDsHPmVM35JavfnQ2itYSCn3ujl4krRni1QU,1076
 deepdoctection/datasets/adapter.py,sha256=Ly_vbOAgVI73V41FUccnSX1ECTOyesW_qsuvQuvOZbw,7796
 deepdoctection/datasets/base.py,sha256=DT4i-d74sIEiUNC6UspIHNJuHSK0t1dBv7qwadg4rLw,22341
@@ -57,8 +59,8 @@ deepdoctection/extern/hfdetr.py,sha256=1NPW_u5eH2tP3ixZ91l4WR-O-wLVcrFsLWA7BqID0
 deepdoctection/extern/hflayoutlm.py,sha256=KfoWx9_Rpa1Y2L51HLrYvenfWaTB4SVTmVJH00Cqb-s,56510
 deepdoctection/extern/hflm.py,sha256=kwS6kcSlY_2m9u0RzBLTRq-UMM7c1PhyUaDTvSdejus,9217
 deepdoctection/extern/model.py,sha256=ViHHKPvbGmLCPw7ZESv_rmjlkA90UiBU6oZiHOMqNSw,59869
-deepdoctection/extern/pdftext.py,sha256=9EvDstMBeOeCFXM21wKaj5iTOUJSt8_50RfGdMcMjIA,4048
-deepdoctection/extern/tessocr.py,sha256=GCTcVHm6oOXS2Xq76j-xY9etRDDJA5qfqWJ5AJ-Kn8k,17400
+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
 deepdoctection/extern/tpdetect.py,sha256=yAk1duQdoX-_pHLHgvhU7OOSiDy863q6XUMpjpYR734,8477
 deepdoctection/extern/pt/__init__.py,sha256=3Cu0ZHjbYsJomru7-RQXEHihEQLegZrmLetlHiqS58I,742
@@ -124,23 +126,23 @@ deepdoctection/utils/__init__.py,sha256=brBceRWeov9WXMiJTjyJOF2rHMP8trGGRRjhMdZ6
 deepdoctection/utils/concurrency.py,sha256=nIhpkSncmv0LBB8PtcOLY-BsRGlfcDpz7foVdgzZd20,4598
 deepdoctection/utils/context.py,sha256=VSnJnTtRGuq3w-0-syTf9DXOhR7WsPvWLLWTxKIBYec,4186
 deepdoctection/utils/develop.py,sha256=4HyTarkFbJwctL-Hgu1TU_LSJppHvaroDbcyHsxhIA8,3444
-deepdoctection/utils/env_info.py,sha256=Fm6A4XfJsYQmW5TzPmwn7_jh9qx5jqYlt00k9NK0yR8,18007
+deepdoctection/utils/env_info.py,sha256=TnCA-LOTj4WIHd9yvn1AaoPWsLmPgc42l-BJmGV6zmM,19147
 deepdoctection/utils/error.py,sha256=_3q9VepKfEhsM3H033_Fu0hwBzMSjsWALsjyJbGAZr8,2367
-deepdoctection/utils/file_utils.py,sha256=koYsfHtl0-nh8T9nUb215Rc1X-WDvk2gEjyw-YJVZ34,19019
+deepdoctection/utils/file_utils.py,sha256=IRElrcND0YEiU1QELw5hfXeNA39uE2_nyzh9-X7YcxI,19477
 deepdoctection/utils/fs.py,sha256=C4ktrzjoVtX9kgycv5YrEigDI9byi65b6_D0aKsGM4Y,10161
 deepdoctection/utils/identifier.py,sha256=QkNaGGqPynHwDPnd3_m8iur4Cv64rcQa7qolCE7Qphk,2159
 deepdoctection/utils/logger.py,sha256=J0OVKiXP_2A82MWbbJoOeMEJ-75aZu5npgaS_yI6mVA,10003
-deepdoctection/utils/metacfg.py,sha256=AGAE-KOymOLsarpUBBYawpVSXImvJyUeOD4LD2W_7Yo,5196
+deepdoctection/utils/metacfg.py,sha256=hD76KQ_RnD_5B02qLI2Zxf3WfnsnXhEI_KUTKpw91RI,5711
 deepdoctection/utils/mocks.py,sha256=IkN3-IzAl4eX0ibgKIHg8IY7ykVw6BnpF6XnxKnKaZI,2389
-deepdoctection/utils/pdf_utils.py,sha256=H5BdLXvDlvTEfb-3zcRjy207PeqEnaymkG122R7UA4o,8635
+deepdoctection/utils/pdf_utils.py,sha256=OAQjE9xHVNcDsFqAvX47Lu-mgmoMpVXqIf5pOK8AwxY,11595
 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=ANzyIX6AY1yc-4gcn6yxksV84sPrJDaUurUNVatAFu8,5168
 deepdoctection/utils/viz.py,sha256=Xm6pKlhM29UWBBGZHlWFl9XYFDAqaYDdwHXwe26Hvqo,25728
-deepdoctection-0.34.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
-deepdoctection-0.34.dist-info/METADATA,sha256=YJ5XJnf7zMlDmr6f7vqvFNL11hy-ZEz8VbdYgii0AQo,19169
-deepdoctection-0.34.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-deepdoctection-0.34.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
-deepdoctection-0.34.dist-info/RECORD,,
+deepdoctection-0.35.dist-info/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
+deepdoctection-0.35.dist-info/METADATA,sha256=B6pPQjRYWcqd1p-3ul3PhflYOcKq2ZpP5D-i8kr7qgk,19403
+deepdoctection-0.35.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
+deepdoctection-0.35.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
+deepdoctection-0.35.dist-info/RECORD,,

{deepdoctection-0.34.dist-info → deepdoctection-0.35.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.1.0)
+Generator: setuptools (75.3.0)
 Root-Is-Purelib: true
 Tag: py3-none-any