doc-page-extractor 0.1.1__py3-none-any.whl → 1.0.2__py3-none-any.whl

doc_page_extractor/table.py

@@ -1,71 +0,0 @@
-import os
-import torch
-
-from typing import Literal, Any
-from PIL.Image import Image
-from .types import TableLayoutParsedFormat
-from .utils import expand_image
-
-
-OutputFormat = Literal["latex", "markdown", "html"]
-
-class Table:
-  def __init__(
-      self,
-      device: Literal["cpu", "cuda"],
-      model_path: str,
-    ):
-    self._model: Any | None = None
-    self._model_path: str = model_path
-    self._ban: bool = False
-    if device == "cpu" or not torch.cuda.is_available():
-      self._ban = True
-
-  def predict(self, image: Image, format: TableLayoutParsedFormat) -> str | None:
-    if self._ban:
-      print("CUDA is not available. You cannot parse table from image.")
-      return None
-
-    output_format: str
-    if format == TableLayoutParsedFormat.LATEX:
-      output_format = "latex"
-    elif format == TableLayoutParsedFormat.MARKDOWN:
-      output_format = "markdown"
-    elif format == TableLayoutParsedFormat.HTML:
-      output_format = "html"
-    else:
-      raise ValueError(f"Table format {format} is not supported.")
-
-    image = expand_image(image, 0.1)
-    model = self._get_model()
-
-    with torch.no_grad():
-      results = model([image], output_format=output_format)
-
-    if len(results) == 0:
-      return None
-
-    return results[0]
-
-  def _get_model(self):
-    if self._model is None:
-      local_files_only: bool
-      if os.path.exists(self._model_path):
-        local_files_only = True
-      else:
-        local_files_only = False
-        os.makedirs(self._model_path)
-
-      from .struct_eqtable import build_model
-      model = build_model(
-        model_ckpt="U4R/StructTable-InternVL2-1B",
-        max_new_tokens=1024,
-        max_time=30,
-        lmdeploy=False,
-        flash_attn=True,
-        batch_size=1,
-        cache_dir=self._model_path,
-        local_files_only=local_files_only,
-      )
-      self._model = model.cuda()
-    return self._model

doc_page_extractor/types.py

@@ -1,67 +0,0 @@
-from dataclasses import dataclass
-from typing import Literal
-from enum import auto, Enum
-from PIL.Image import Image
-from .rectangle import Rectangle
-
-
-@dataclass
-class OCRFragment:
-  order: int
-  text: str
-  rank: float
-  rect: Rectangle
-
-class LayoutClass(Enum):
-  TITLE = 0
-  PLAIN_TEXT = 1
-  ABANDON = 2
-  FIGURE = 3
-  FIGURE_CAPTION = 4
-  TABLE = 5
-  TABLE_CAPTION = 6
-  TABLE_FOOTNOTE = 7
-  ISOLATE_FORMULA = 8
-  FORMULA_CAPTION = 9
-
-class TableLayoutParsedFormat(Enum):
-  LATEX = auto()
-  MARKDOWN = auto()
-  HTML = auto()
-
-@dataclass
-class BaseLayout:
-  rect: Rectangle
-  fragments: list[OCRFragment]
-
-@dataclass
-class PlainLayout(BaseLayout):
-  cls: Literal[
-    LayoutClass.TITLE,
-    LayoutClass.PLAIN_TEXT,
-    LayoutClass.ABANDON,
-    LayoutClass.FIGURE,
-    LayoutClass.FIGURE_CAPTION,
-    LayoutClass.TABLE_CAPTION,
-    LayoutClass.TABLE_FOOTNOTE,
-    LayoutClass.FORMULA_CAPTION,
-  ]
-
-@dataclass
-class TableLayout(BaseLayout):
-  parsed: tuple[str, TableLayoutParsedFormat] | None
-  cls: LayoutClass.TABLE
-
-@dataclass
-class FormulaLayout(BaseLayout):
-  latex: str | None
-  cls: LayoutClass.ISOLATE_FORMULA
-
-Layout = PlainLayout | TableLayout | FormulaLayout
-
-@dataclass
-class ExtractedResult:
-  rotation: float
-  layouts: list[Layout]
-  extracted_image: Image
-  adjusted_image: Image | None

doc_page_extractor/utils.py

@@ -1,32 +0,0 @@
-import os
-import re
-
-from math import ceil
-from PIL.Image import Image
-from PIL.ImageOps import expand
-
-
-def ensure_dir(path: str) -> str:
-  path = os.path.abspath(path)
-  os.makedirs(path, exist_ok=True)
-  return path
-
-def is_space_text(text: str) -> bool:
-  return re.match(r"^\s*$", text)
-
-def expand_image(image: Image, percent: float):
-  width, height = image.size
-  border_width = ceil(width * percent)
-  border_height = ceil(height * percent)
-  fill_color: tuple[int, ...]
-
-  if image.mode == "RGBA":
-    fill_color = (255, 255, 255, 255)
-  else:
-    fill_color = (255, 255, 255)
-
-  return expand(
-    image=image,
-    border=(border_width, border_height),
-    fill=fill_color,
-  )

doc_page_extractor-0.1.1.dist-info/METADATA

@@ -1,84 +0,0 @@
-Metadata-Version: 2.4
-Name: doc-page-extractor
-Version: 0.1.1
-Summary: doc page extractor can identify text and format in images and return structured data.
-Home-page: https://github.com/Moskize91/doc-page-extractor
-Author: Tao Zeyu
-Author-email: i@taozeyu.com
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: opencv-python<5.0,>=4.11.0
-Requires-Dist: pillow<11.0,>=10.3
-Requires-Dist: pyclipper<2.0,>=1.2.0
-Requires-Dist: numpy<2.0,>=1.24.0
-Requires-Dist: shapely<3.0,>=2.0.0
-Requires-Dist: transformers<=4.47,>=4.42.4
-Requires-Dist: doclayout_yolo>=0.0.3
-Requires-Dist: pix2tex<=0.2.0,>=0.1.4
-Requires-Dist: accelerate<2.0,>=1.6.0
-Dynamic: author
-Dynamic: author-email
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: license-file
-Dynamic: requires-dist
-Dynamic: summary
-
-# doc page extractor
-
-English | [中文](./README_zh-CN.md)
-
-## Introduction
-
-doc page extractor can identify text and format in images and return structured data.
-
-## Installation
-
-```shell
-pip install doc-page-extractor
-```
-
-```shell
-pip install onnxruntime==1.21.0
-```
-
-## Using CUDA
-
-Please refer to the introduction of [PyTorch](https://pytorch.org/get-started/locally/) and select the appropriate command to install according to your operating system.
-
-In addition, replace the command to install `onnxruntime` in the previous article with the following:
-
-```shell
-pip install onnxruntime-gpu==1.21.0
-```
-
-## Example
-
-```python
-from PIL import Image
-from doc_page_extractor import DocExtractor
-
-extractor = DocExtractor(
-  model_dir_path=model_path, # Folder address where AI model is downloaded and installed
-  device="cpu", # If you want to use CUDA, please change to device="cuda".
-)
-with Image.open("/path/to/your/image.png") as image:
-  result = extractor.extract(
-  image=image,
-  lang="ch", # Language of image text
-)
-for layout in result.layouts:
-  for fragment in layout.fragments:
-    print(fragment.rect, fragment.text)
-```
-
-## Acknowledgements
-
-The code of `doc_page_extractor/onnxocr` in this repo comes from [OnnxOCR](https://github.com/jingsongliujing/OnnxOCR).
-
-- [DocLayout-YOLO](https://github.com/opendatalab/DocLayout-YOLO)
-- [OnnxOCR](https://github.com/jingsongliujing/OnnxOCR)
-- [layoutreader](https://github.com/ppaanngggg/layoutreader)
-- [StructEqTable](https://github.com/Alpha-Innovator/StructEqTable-Deploy)
-- [LaTeX-OCR](https://github.com/lukas-blecher/LaTeX-OCR)

doc_page_extractor-0.1.1.dist-info/RECORD

@@ -1,44 +0,0 @@
-doc_page_extractor/__init__.py,sha256=9rWKSMTgzP7Xv15zA4upsyPaR8S8JeNpMyhWElRCW0M,311
-doc_page_extractor/clipper.py,sha256=5S1TI0aqMebwlPv_Ih4Nxpp6MchEjOih-CiZfMWUAhI,3201
-doc_page_extractor/downloader.py,sha256=NbGN9ARnER8-gd4T1uc3W98WMEClVxMrqnShq8HibTw,455
-doc_page_extractor/extractor.py,sha256=V0S9Nn65lKpkz8DnTUMcsAzUJGwSQvupIBTiVqzLpJ8,7303
-doc_page_extractor/latex.py,sha256=W_zAcksNRuru-WjCq4CSn07s_SWrDhikadJSy_Cg3Do,1954
-doc_page_extractor/layout_order.py,sha256=NwMzTPr4xsriz4slCwqwhw2-vrMu-qfwtcFsDu8d1yM,7426
-doc_page_extractor/layoutreader.py,sha256=BdC4oPbtpXoLmYhjuSFrKn6SNoT2zWw_gi95sGAUwrk,4031
-doc_page_extractor/ocr.py,sha256=hQhT9bdsJmWESqt1FODCoE19wfOroM8uHZiFoZZrkQU,5182
-doc_page_extractor/ocr_corrector.py,sha256=RfRA1jESEuqC8_a2kUEvHblT_B4xBjE0OApLMl1JiRg,3917
-doc_page_extractor/overlap.py,sha256=z1DF4_2OPvauDHwmz1SC1WosULkE84HKaRfNEgexPzc,5337
-doc_page_extractor/plot.py,sha256=4uibjS_x1SyEyjaJJd0YsBbzkgldDOCct4Ry2cOhdXU,2556
-doc_page_extractor/raw_optimizer.py,sha256=1KghECq_rJwuZZITTLQnGTKYivFKg_qDvMLN9g17sks,2844
-doc_page_extractor/rectangle.py,sha256=yeW6srdrsxaJg1eb3nn8oxtY0sfgeBk3hMiuJGaRXwY,1678
-doc_page_extractor/rotation.py,sha256=QCZ-HqfDxIhnQw8KRHki2myj6-UusvNY7Mpjsu-wI-4,4334
-doc_page_extractor/table.py,sha256=AWymTRbOet55uImW8QJqb90Qs_v2V2U1mZv0U6rSz3c,1891
-doc_page_extractor/types.py,sha256=7blT8YNKrOsc4qQdAhM7J7MEQjFcBwE0QV8-lipZBeQ,1305
-doc_page_extractor/utils.py,sha256=ZlQVOLPUg_v5J8u6SoD8XtMG_JkF-ERgjubc4LO5_Lg,688
-doc_page_extractor/onnxocr/__init__.py,sha256=BK4YpX4pU0nRxbcI5f5cbIVfdBEsx4W980QYmpNQaH0,38
-doc_page_extractor/onnxocr/cls_postprocess.py,sha256=o879Ned0RMUERYLviuToZ0xTvhn2UsYAb-yPC5gj8h4,822
-doc_page_extractor/onnxocr/db_postprocess.py,sha256=R3yXXfReiQgLaYIvvfnrFfshI202LjHMvcZwcLpjmTY,7913
-doc_page_extractor/onnxocr/imaug.py,sha256=Q192kIsRPI5zTm4RA_UUXlo6tvGJS8wrUaa-xrfnO_w,811
-doc_page_extractor/onnxocr/operators.py,sha256=0nLiV1dWej9vdPa_DO04F7SvqF-l9NOFgHUuHUPNvsw,5556
-doc_page_extractor/onnxocr/predict_base.py,sha256=LzRSPgxgFSRAreJOMpDTUKuBvvO_Qe5_5tK4lNuGl-w,1269
-doc_page_extractor/onnxocr/predict_cls.py,sha256=28MliSQIyHc82EUbdkQb31KaB90rSzSmps1v6WsskQk,3065
-doc_page_extractor/onnxocr/predict_det.py,sha256=VYsvNbCJQi1UuetwvR_hx-U7JScHyFjmJmo8YwHyQt4,4092
-doc_page_extractor/onnxocr/predict_rec.py,sha256=qQrCs5jzCf5PYp-iEKJ53pcx_xRoJdJyavPvsvuh5Ic,10999
-doc_page_extractor/onnxocr/predict_system.py,sha256=yoqXunAsoboPsWe7qQjvQf2_SMW1T1QMriEoiGdX3BM,2721
-doc_page_extractor/onnxocr/rec_postprocess.py,sha256=qZt5Ripal7z9hniKq5e7azOkD9e6NR1ylWpRpznhweg,29556
-doc_page_extractor/onnxocr/utils.py,sha256=AQoHgQyv-jpPo4BsVzq3r7_ze698EZ-a7LJobm2fwUI,1864
-doc_page_extractor/struct_eqtable/__init__.py,sha256=QoTsNuJfpNSrMIMd6Cot1jJqWk88_lDqFP_C2rcVJO4,1329
-doc_page_extractor/struct_eqtable/internvl/__init__.py,sha256=2aOsU-aHkFv_gjdP8LeUXjj_9-0d4x79iyxh4cCzaEw,79
-doc_page_extractor/struct_eqtable/internvl/conversation.py,sha256=s7DceRlM6JtHmxgyuE6vqu5XVT1fHzhzCL_I6r8MI1c,15129
-doc_page_extractor/struct_eqtable/internvl/internvl.py,sha256=ovVZG-PuBrsj_9lEoNPOygJ-2en3v6gPzRfWjDpWNOM,7678
-doc_page_extractor/struct_eqtable/internvl/internvl_lmdeploy.py,sha256=ACHxFntxS38G43PzE955Nv4fjKk_-Oz4y_o9JEjQwlg,2608
-doc_page_extractor/struct_eqtable/pix2s/__init__.py,sha256=cXRo4eg6u1-TXktZ8rQf0HIzLmmScIwYQhbxMKl-MyA,76
-doc_page_extractor/struct_eqtable/pix2s/pix2s.py,sha256=fCNve8PNeJ3-AWJIhSeGtp-mYKoMXfW0CIpszkQnAaA,2535
-doc_page_extractor/struct_eqtable/pix2s/pix2s_trt.py,sha256=zSGw45JhWdZ3iuJel5Chsy-NzsOHx9QyPQIUAzzCjFE,43880
-doc_page_extractor-0.1.1.dist-info/licenses/LICENSE,sha256=TfPDBt3ar0uv_f9cqCDMZ5rIzW3CY8anRRd4PkL6ejs,34522
-tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-tests/test_history_bus.py,sha256=WaCUW3U75SESMcLq0f5FKnpVUVRDvmfxLFE7Zo83e48,2517
-doc_page_extractor-0.1.1.dist-info/METADATA,sha256=5bQtvYgjNghsYER1zmc19_6BH4JrgrLj9_KxUmnLnHc,2436
-doc_page_extractor-0.1.1.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-doc_page_extractor-0.1.1.dist-info/top_level.txt,sha256=ErNybD_lBzAmw8mVBAK4htsAH_hp14jioZVex-tUqvM,25
-doc_page_extractor-0.1.1.dist-info/RECORD,,