doc-page-extractor 0.2.1__tar.gz → 0.2.3__tar.gz

{doc_page_extractor-0.2.1 → doc_page_extractor-0.2.3}/PKG-INFO

@@ -1,30 +1,33 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.3
 Name: doc-page-extractor
-Version: 0.2.1
-Summary: doc page extractor can identify text and format in images and return structured data.
-Home-page: https://github.com/Moskize91/doc-page-extractor
+Version: 0.2.3
+Summary: 
+License: AGPL-3.0
 Author: Tao Zeyu
 Author-email: i@taozeyu.com
+Maintainer: Tao Zeyu
+Maintainer-email: i@taozeyu.com
+Requires-Python: >=3.10,<3.13
+Classifier: Development Status :: 2 - Pre-Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: GNU Affero General Public License v3
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Dist: accelerate (>=1.6.0,<2.0)
+Requires-Dist: doclayout_yolo (>=0.0.3)
+Requires-Dist: huggingface_hub (>=0.33.0,<1.0)
+Requires-Dist: numpy (>=1.24.0,<2.0)
+Requires-Dist: opencv-python (>=4.10.0,<5.0)
+Requires-Dist: pillow (>=10.3,<11.0)
+Requires-Dist: pix2tex (>=0.1.4,<=0.2.0)
+Requires-Dist: pyclipper (>=1.2.0,<2.0)
+Requires-Dist: shapely (>=2.0.0,<3.0)
+Requires-Dist: transformers (>=4.42.4,<=4.47)
+Project-URL: Repository, https://github.com/moskize91/doc-page-extractor
 Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: opencv-python<5.0,>=4.10.0
-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
-Requires-Dist: huggingface_hub<1.0,>=0.30.2
-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
 

{doc_page_extractor-0.2.1 → doc_page_extractor-0.2.3}/doc_page_extractor/__init__.py

@@ -2,6 +2,7 @@ from .extractor import DocExtractor
 from .clipper import clip, clip_from_image
 from .plot import plot
 from .rectangle import Point, Rectangle
+from .model import Model, HuggingfaceModel
 from .types import (
   ExtractedResult,
   OCRFragment,

{doc_page_extractor-0.2.1 → doc_page_extractor-0.2.3}/doc_page_extractor/extractor.py

@@ -52,6 +52,13 @@ class DocExtractor:
     self._latex: LaTeX = LaTeX(device, model)
     self._layout_order: LayoutOrder = LayoutOrder(device, model)
 
+  def prepare_models(self):
+    self._model.get_onnx_ocr_path()
+    self._model.get_yolo_path()
+    self._model.get_layoutreader_path()
+    self._model.get_struct_eqtable_path()
+    self._model.get_latex_path()
+
   def extract(
       self,
       image: Image,

{doc_page_extractor-0.2.1 → doc_page_extractor-0.2.3}/doc_page_extractor/model.py

@@ -1,10 +1,14 @@
 from os import PathLike
+from time import sleep
 from typing import runtime_checkable, Protocol
 from pathlib import Path
 from threading import Lock
 from huggingface_hub import hf_hub_download, snapshot_download, try_to_load_from_cache
 
 
+_RETRY_TIMES = 6
+_RETRY_SLEEP = 3.5
+
 @runtime_checkable
 class Model(Protocol):
   def get_onnx_ocr_path(self) -> Path:
@@ -31,9 +35,10 @@ class HuggingfaceModel(Model):
   def get_onnx_ocr_path(self) -> Path:
     return self._get_model_path(
       repo_id="moskize/OnnxOCR",
-      filename=None,
+      filename="README.md",
       repo_type=None,
-      is_snapshot=True
+      is_snapshot=True,
+      wanna_dir_path=True,
     )
 
   def get_yolo_path(self) -> Path:
@@ -42,14 +47,16 @@ class HuggingfaceModel(Model):
       filename="models/Layout/YOLO/doclayout_yolo_ft.pt",
       repo_type=None,
       is_snapshot=False,
+      wanna_dir_path=False,
     )
 
   def get_layoutreader_path(self) -> Path:
     return self._get_model_path(
       repo_id="hantian/layoutreader",
-      filename=None,
+      filename="model.safetensors",
       repo_type=None,
       is_snapshot=True,
+      wanna_dir_path=True,
     )
 
   def get_struct_eqtable_path(self) -> Path:
@@ -58,6 +65,7 @@ class HuggingfaceModel(Model):
       filename="model.safetensors",
       repo_type=None,
       is_snapshot=True,
+      wanna_dir_path=True,
     )
 
   def get_latex_path(self) -> Path:
@@ -66,38 +74,60 @@ class HuggingfaceModel(Model):
       filename="checkpoints/weights.pth",
       repo_type="space",
       is_snapshot=True,
+      wanna_dir_path=True,
     )
 
   def _get_model_path(
         self,
         repo_id: str,
-        filename: str | None,
+        filename: str,
         repo_type: str | None,
         is_snapshot: bool,
+        wanna_dir_path: bool,
       ) -> Path:
+
     with self._lock:
-      cache_filename = "README.md"
-      if filename is not None:
-        cache_filename = filename
       model_path = try_to_load_from_cache(
         repo_id=repo_id,
-        filename=cache_filename,
+        filename=filename,
         repo_type=repo_type,
         cache_dir=self._model_cache_dir
       )
       if isinstance(model_path, str):
-        if filename is None:
-          model_path = Path(model_path).parent
-
-      elif is_snapshot:
-        model_path = snapshot_download(
-          cache_dir=self._model_cache_dir,
-          repo_id=repo_id,
-        )
+        model_path = Path(model_path)
+        if wanna_dir_path:
+          for _ in Path(filename).parts:
+            model_path = model_path.parent
+
       else:
-        model_path = hf_hub_download(
-          cache_dir=self._model_cache_dir,
-          repo_id=repo_id,
-          filename=filename,
-        )
-      return Path(model_path)
+        # https://github.com/huggingface/huggingface_hub/issues/1542#issuecomment-1630465844
+        latest_error: ConnectionError | None = None
+        for i in range(_RETRY_TIMES + 1):
+          if latest_error is not None:
+            print(f"Retrying to download {repo_id} model, attempt {i + 1}/{_RETRY_TIMES}...")
+            sleep(_RETRY_SLEEP)
+          try:
+            if is_snapshot:
+              model_path = snapshot_download(
+                cache_dir=self._model_cache_dir,
+                repo_id=repo_id,
+                repo_type=repo_type,
+                resume_download=True,
+              )
+            else:
+              model_path = hf_hub_download(
+                cache_dir=self._model_cache_dir,
+                repo_id=repo_id,
+                repo_type=repo_type,
+                filename=filename,
+                resume_download=True,
+              )
+            latest_error = None
+          except ConnectionError as err:
+            latest_error = err
+
+        if latest_error is not None:
+          raise latest_error
+        model_path = Path(model_path)
+
+      return model_path

doc_page_extractor-0.2.3/pyproject.toml

@@ -0,0 +1,48 @@
+[project]
+name = "doc-page-extractor"
+version = "0.2.3"
+description = ""
+authors = [
+    {name = "Tao Zeyu",email = "i@taozeyu.com"}
+]
+maintainers = [
+    {name = "Tao Zeyu", email = "i@taozeyu.com"}
+]
+license = {text = "AGPL-3.0"}
+readme = "README.md"
+requires-python = ">=3.10,<3.13"
+dependencies = [
+    "opencv-python>=4.10.0,<5.0",
+    "pillow>=10.3,<11.0",
+    "pyclipper>=1.2.0,<2.0",
+    "numpy>=1.24.0,<2.0",
+    "shapely>=2.0.0,<3.0",
+    "transformers>=4.42.4,<=4.47",
+    "doclayout_yolo>=0.0.3",
+    "pix2tex>=0.1.4,<=0.2.0",
+    "accelerate>=1.6.0,<2.0",
+    "huggingface_hub>=0.33.0,<1.0",
+]
+
+
+[build-system]
+requires = ["poetry-core>=2.0.0,<3.0.0"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.poetry]
+license = "AGPL-3.0"
+readme = "README.md"
+repository = "https://github.com/moskize91/doc-page-extractor"
+packages = [
+    {include = "doc_page_extractor" }
+]
+classifiers=[
+    "Development Status :: 2 - Pre-Alpha",
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: GNU Affero General Public License v3",
+    "Programming Language :: Python",
+    "Programming Language :: Python :: 3.10",
+]
+
+[tool.poetry.group.dev.dependencies]
+pylint = "^3.3.7"

doc_page_extractor-0.2.1/doc_page_extractor/struct_eqtable/__init__.py

@@ -1,49 +0,0 @@
-from .pix2s import Pix2Struct, Pix2StructTensorRT
-from .internvl import InternVL, InternVL_LMDeploy
-
-from transformers import AutoConfig
-
-
-__ALL_MODELS__ = {
-    'Pix2Struct': Pix2Struct,
-    'Pix2StructTensorRT': Pix2StructTensorRT,
-    'InternVL': InternVL,
-    'InternVL_LMDeploy': InternVL_LMDeploy,
-}
-
-
-def get_model_name(model_path):
-    model_config = AutoConfig.from_pretrained(
-        model_path,
-        trust_remote_code=True,
-    )
-
-    if 'Pix2Struct' in model_config.architectures[0]:
-        model_name = 'Pix2Struct'
-    elif 'InternVL' in model_config.architectures[0]:
-        model_name = 'InternVL'
-    else:
-        raise ValueError(f"Unsupported model type: {model_config.architectures[0]}")
-
-    return model_name
-
-
-def build_model(
-        model_ckpt='U4R/StructTable-InternVL2-1B',
-        cache_dir=None,
-        local_files_only=None,
-        **kwargs,
-    ):
-    model_name = get_model_name(model_ckpt)
-    if model_name == 'InternVL' and kwargs.get('lmdeploy', False):
-        model_name = 'InternVL_LMDeploy'
-    elif model_name == 'Pix2Struct' and kwargs.get('tensorrt_path', None):
-        model_name = 'Pix2StructTensorRT'
-
-    model = __ALL_MODELS__[model_name](
-        model_ckpt,
-        cache_dir=cache_dir,
-        local_files_only=local_files_only,
-        **kwargs
-    )
-    return model

doc_page_extractor-0.2.1/doc_page_extractor/struct_eqtable/internvl/__init__.py

@@ -1,2 +0,0 @@
-from .internvl import InternVL
-from .internvl_lmdeploy import InternVL_LMDeploy