deepdoctection 0.34__py3-none-any.whl → 0.36__py3-none-any.whl

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/utils/utils.py

@@ -155,3 +155,42 @@ def is_file_extension(file_name: PathLikeOrStr, extension: Union[str, Sequence[s
     if isinstance(extension, str):
         return os.path.splitext(file_name)[-1].lower() == extension
     return os.path.splitext(file_name)[-1].lower() in extension
+
+
+def partition_list(base_list: list[str], stop_value: str) -> list[list[str]]:
+    """
+    Partitions a list of strings into sublists, where each sublist starts with the first occurrence of the stop value.
+    Consecutive stop values are grouped together in the same sublist.
+
+    :param base_list: The list of strings to be partitioned.
+    :param stop_value: The string value that indicates the start of a new partition.
+    :return: A list of lists, where each sublist is a partition of the original list.
+
+    ** Example:**
+
+        strings = ['a', 'a', 'c', 'c', 'b', 'd', 'c', 'c', 'a', 'b', 'a', 'b', 'a', 'a']
+        stop_string = 'a'
+        partition_list(strings, stop_string)
+
+       # Output [['a', 'a', 'c', 'c', 'b', 'd', 'c', 'c'], ['a', 'b'], ['a', 'b'], ['a', 'a']]
+    """
+
+    partitions = []
+    current_partition: list[str] = []
+    stop_found = False
+
+    for s in base_list:
+        if s == stop_value:
+            if not stop_found and current_partition:
+                partitions.append(current_partition)
+                current_partition = []
+            current_partition.append(s)
+            stop_found = True
+        else:
+            current_partition.append(s)
+            stop_found = False
+
+    if current_partition:
+        partitions.append(current_partition)
+
+    return partitions

deepdoctection/utils/viz.py

@@ -205,6 +205,7 @@ def draw_boxes(
     font_scale: float = 1.0,
     rectangle_thickness: int = 4,
     box_color_by_category: bool = True,
+    show_palette: bool = True,
 ) -> PixelValues:
     """
     Dray bounding boxes with category names into image.
@@ -216,6 +217,7 @@ def draw_boxes(
     :param font_scale: Font scale of text box
     :param rectangle_thickness: Thickness of bounding box
     :param box_color_by_category:
+    :param show_palette: Whether to show a color palette of the categories
     :return: A new image np.ndarray
     """
     if color is not None:
@@ -261,19 +263,20 @@ def draw_boxes(
         )
 
     # draw a (very ugly) color palette
-    y_0 = np_image.shape[0]
-    for category, col in category_to_color.items():
-        if category is not None:
-            np_image = viz_handler.draw_text(
-                np_image,
-                (np_image.shape[1], y_0),
-                category,
-                color=col,
-                font_scale=font_scale * 3,
-                rectangle_thickness=rectangle_thickness,
-            )
-            _, text_h = viz_handler.get_text_size(category, font_scale * 2)
-            y_0 = y_0 - int(10 * text_h)
+    if show_palette:
+        y_0 = np_image.shape[0]
+        for category, col in category_to_color.items():
+            if category is not None:
+                np_image = viz_handler.draw_text(
+                    np_image,
+                    (np_image.shape[1], y_0),
+                    category,
+                    color=col,
+                    font_scale=font_scale,
+                    rectangle_thickness=rectangle_thickness,
+                )
+                _, text_h = viz_handler.get_text_size(category, font_scale * 2)
+                y_0 = y_0 - int(1 * text_h)
 
     return np_image
 

{deepdoctection-0.34.dist-info → deepdoctection-0.36.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: deepdoctection
-Version: 0.34
+Version: 0.36
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -16,114 +16,117 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 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: importlib-metadata >=5.0.0
-Requires-Dist: jsonlines ==3.1.0
-Requires-Dist: lazy-imports ==0.3.1
-Requires-Dist: mock ==4.0.3
-Requires-Dist: networkx >=2.7.1
-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: pyyaml >=6.0.1
-Requires-Dist: pyzmq >=16
-Requires-Dist: scipy >=1.13.1
-Requires-Dist: termcolor >=1.1
-Requires-Dist: tabulate >=0.7.7
-Requires-Dist: tqdm ==4.64.0
-Provides-Extra: dev
-Requires-Dist: python-dotenv ==1.0.0 ; extra == 'dev'
-Requires-Dist: click ; extra == 'dev'
-Requires-Dist: black ==23.7.0 ; extra == 'dev'
-Requires-Dist: isort ==5.13.2 ; extra == 'dev'
-Requires-Dist: pylint ==2.17.4 ; extra == 'dev'
-Requires-Dist: mypy ==1.4.1 ; extra == 'dev'
-Requires-Dist: wandb ; extra == 'dev'
-Requires-Dist: types-PyYAML >=6.0.12.12 ; extra == 'dev'
-Requires-Dist: types-termcolor >=1.1.3 ; extra == 'dev'
-Requires-Dist: types-tabulate >=0.9.0.3 ; extra == 'dev'
-Requires-Dist: types-tqdm >=4.66.0.5 ; extra == 'dev'
-Requires-Dist: lxml-stubs >=0.5.1 ; extra == 'dev'
-Requires-Dist: types-Pillow >=10.2.0.20240406 ; extra == 'dev'
-Requires-Dist: types-urllib3 >=1.26.25.14 ; extra == 'dev'
-Provides-Extra: docs
-Requires-Dist: tensorpack ==0.11 ; extra == 'docs'
-Requires-Dist: boto3 ==1.34.102 ; extra == 'docs'
-Requires-Dist: transformers >=4.36.0 ; extra == 'docs'
-Requires-Dist: accelerate >=0.29.1 ; extra == 'docs'
-Requires-Dist: pdfplumber >=0.11.0 ; extra == 'docs'
-Requires-Dist: lxml >=4.9.1 ; extra == 'docs'
-Requires-Dist: lxml-stubs >=0.5.1 ; extra == 'docs'
-Requires-Dist: jdeskew >=0.2.2 ; extra == 'docs'
-Requires-Dist: jinja2 ==3.0.3 ; extra == 'docs'
-Requires-Dist: mkdocs-material ; extra == 'docs'
-Requires-Dist: mkdocstrings-python ; extra == 'docs'
-Requires-Dist: griffe ==0.25.0 ; extra == 'docs'
+Requires-Dist: catalogue==2.0.10
+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
+Requires-Dist: mock==4.0.3
+Requires-Dist: networkx>=2.7.1
+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
+Requires-Dist: termcolor>=1.1
+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: importlib-metadata>=5.0.0; extra == "tf"
+Requires-Dist: jsonlines==3.1.0; extra == "tf"
+Requires-Dist: lazy-imports==0.3.1; extra == "tf"
+Requires-Dist: mock==4.0.3; extra == "tf"
+Requires-Dist: networkx>=2.7.1; extra == "tf"
+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"
+Requires-Dist: termcolor>=1.1; extra == "tf"
+Requires-Dist: tabulate>=0.7.7; extra == "tf"
+Requires-Dist: tqdm==4.64.0; extra == "tf"
+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: pycocotools>=2.0.2; extra == "tf"
+Requires-Dist: boto3==1.34.102; extra == "tf"
+Requires-Dist: pdfplumber>=0.11.0; extra == "tf"
+Requires-Dist: fasttext==0.9.2; extra == "tf"
+Requires-Dist: jdeskew>=0.2.2; extra == "tf"
+Requires-Dist: apted==1.0.3; extra == "tf"
+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 >=0.12.0 ; 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'
-Requires-Dist: mock ==4.0.3 ; extra == 'pt'
-Requires-Dist: networkx >=2.7.1 ; extra == 'pt'
-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: pyyaml >=6.0.1 ; extra == 'pt'
-Requires-Dist: pyzmq >=16 ; extra == 'pt'
-Requires-Dist: scipy >=1.13.1 ; extra == 'pt'
-Requires-Dist: termcolor >=1.1 ; extra == 'pt'
-Requires-Dist: tabulate >=0.7.7 ; extra == 'pt'
-Requires-Dist: tqdm ==4.64.0 ; extra == 'pt'
-Requires-Dist: timm >=0.9.16 ; extra == 'pt'
-Requires-Dist: transformers >=4.36.0 ; extra == 'pt'
-Requires-Dist: accelerate >=0.29.1 ; extra == 'pt'
-Requires-Dist: python-doctr ==0.8.1 ; extra == 'pt'
-Requires-Dist: boto3 ==1.34.102 ; extra == 'pt'
-Requires-Dist: pdfplumber >=0.11.0 ; extra == 'pt'
-Requires-Dist: fasttext ==0.9.2 ; extra == 'pt'
-Requires-Dist: jdeskew >=0.2.2 ; extra == 'pt'
-Requires-Dist: apted ==1.0.3 ; extra == 'pt'
-Requires-Dist: distance ==0.1.3 ; extra == 'pt'
-Requires-Dist: lxml >=4.9.1 ; extra == 'pt'
+Requires-Dist: catalogue==2.0.10; 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"
+Requires-Dist: mock==4.0.3; extra == "pt"
+Requires-Dist: networkx>=2.7.1; extra == "pt"
+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"
+Requires-Dist: termcolor>=1.1; extra == "pt"
+Requires-Dist: tabulate>=0.7.7; extra == "pt"
+Requires-Dist: tqdm==4.64.0; extra == "pt"
+Requires-Dist: timm>=0.9.16; extra == "pt"
+Requires-Dist: transformers>=4.36.0; extra == "pt"
+Requires-Dist: accelerate>=0.29.1; extra == "pt"
+Requires-Dist: python-doctr==0.8.1; extra == "pt"
+Requires-Dist: boto3==1.34.102; extra == "pt"
+Requires-Dist: pdfplumber>=0.11.0; extra == "pt"
+Requires-Dist: fasttext==0.9.2; extra == "pt"
+Requires-Dist: jdeskew>=0.2.2; extra == "pt"
+Requires-Dist: apted==1.0.3; extra == "pt"
+Requires-Dist: distance==0.1.3; extra == "pt"
+Requires-Dist: lxml>=4.9.1; extra == "pt"
+Provides-Extra: docs
+Requires-Dist: tensorpack==0.11; extra == "docs"
+Requires-Dist: boto3==1.34.102; extra == "docs"
+Requires-Dist: transformers>=4.36.0; extra == "docs"
+Requires-Dist: accelerate>=0.29.1; extra == "docs"
+Requires-Dist: pdfplumber>=0.11.0; extra == "docs"
+Requires-Dist: lxml>=4.9.1; extra == "docs"
+Requires-Dist: lxml-stubs>=0.5.1; extra == "docs"
+Requires-Dist: jdeskew>=0.2.2; extra == "docs"
+Requires-Dist: jinja2==3.0.3; extra == "docs"
+Requires-Dist: mkdocs-material; extra == "docs"
+Requires-Dist: mkdocstrings-python; extra == "docs"
+Requires-Dist: griffe==0.25.0; extra == "docs"
+Provides-Extra: dev
+Requires-Dist: python-dotenv==1.0.0; extra == "dev"
+Requires-Dist: click; extra == "dev"
+Requires-Dist: black==23.7.0; extra == "dev"
+Requires-Dist: isort==5.13.2; extra == "dev"
+Requires-Dist: pylint==2.17.4; extra == "dev"
+Requires-Dist: mypy==1.4.1; extra == "dev"
+Requires-Dist: wandb; extra == "dev"
+Requires-Dist: types-PyYAML>=6.0.12.12; extra == "dev"
+Requires-Dist: types-termcolor>=1.1.3; extra == "dev"
+Requires-Dist: types-tabulate>=0.9.0.3; extra == "dev"
+Requires-Dist: types-tqdm>=4.66.0.5; extra == "dev"
+Requires-Dist: lxml-stubs>=0.5.1; extra == "dev"
+Requires-Dist: types-Pillow>=10.2.0.20240406; extra == "dev"
+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'
-Provides-Extra: tf
-Requires-Dist: catalogue ==2.0.10 ; extra == 'tf'
-Requires-Dist: huggingface-hub >=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'
-Requires-Dist: mock ==4.0.3 ; extra == 'tf'
-Requires-Dist: networkx >=2.7.1 ; extra == 'tf'
-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: pyyaml >=6.0.1 ; extra == 'tf'
-Requires-Dist: pyzmq >=16 ; extra == 'tf'
-Requires-Dist: scipy >=1.13.1 ; extra == 'tf'
-Requires-Dist: termcolor >=1.1 ; extra == 'tf'
-Requires-Dist: tabulate >=0.7.7 ; extra == 'tf'
-Requires-Dist: tqdm ==4.64.0 ; extra == 'tf'
-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: pycocotools >=2.0.2 ; extra == 'tf'
-Requires-Dist: boto3 ==1.34.102 ; extra == 'tf'
-Requires-Dist: pdfplumber >=0.11.0 ; extra == 'tf'
-Requires-Dist: fasttext ==0.9.2 ; extra == 'tf'
-Requires-Dist: jdeskew >=0.2.2 ; extra == 'tf'
-Requires-Dist: apted ==1.0.3 ; extra == 'tf'
-Requires-Dist: distance ==0.1.3 ; extra == 'tf'
-Requires-Dist: lxml >=4.9.1 ; extra == 'tf'
+Requires-Dist: pytest==8.0.2; extra == "test"
+Requires-Dist: pytest-cov; extra == "test"
 
 
 <p align="center">
@@ -172,13 +175,17 @@ 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)
-   (not contained in the built-in Analyzer).
- - [**new**] Token classification with [**LiLT**](https://github.com/jpWang/LiLT) via 
+ - 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)
+ - 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 
-   LayoutLM can be used for LiLT as well and we will be providing a notebook on how to train a model on a custom dataset soon.
+   LayoutLM can be used for LiLT as well.
+ - [**new**] There are two notebooks available that show, how to write a 
+   [custom predictor](https://github.com/deepdoctection/notebooks/blob/main/Doclaynet_Analyzer_Config.ipynb) based on 
+   a third party library that has not been supported yet and how to use 
+   [advanced configuration](https://github.com/deepdoctection/notebooks/blob/main/Doclaynet_Analyzer_Config.ipynb) to 
+   get links between layout segments e.g. captions and tables or figures.
 
 **deep**doctection provides on top of that methods for pre-processing inputs to models like cropping or resizing and to 
 post-process results, like validating duplicate outputs, relating words to detected layout segments or ordering words 
@@ -263,7 +270,7 @@ 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 +279,16 @@ 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 +406,8 @@ 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 ...