deepdoctection 0.43__py3-none-any.whl → 0.43.1__py3-none-any.whl

deepdoctection/__init__.py

@@ -25,11 +25,11 @@ from .utils.logger import LoggingRecord, logger
 
 # pylint: enable=wrong-import-position
 
-__version__ = "0.43"
+__version__ = "0.43.1"
 
 _IMPORT_STRUCTURE = {
     "analyzer": ["config_sanity_checks", "get_dd_analyzer", "ServiceFactory"],
-    "configs": [],
+    "configs": ["update_cfg_from_defaults"],
     "dataflow": [
         "DataFlowTerminated",
         "DataFlowResetStateNotCalled",
@@ -98,6 +98,7 @@ _IMPORT_STRUCTURE = {
         "List",
         "Cell",
         "Table",
+        "IMAGE_DEFAULTS",
         "Page",
     ],
     "datasets": [

deepdoctection/analyzer/config.py

@@ -902,3 +902,18 @@ cfg.LAYOUT_LINK.CHILD_CATEGORIES = [LayoutType.CAPTION]
 # Freezes the configuration to make it immutable.
 # This prevents accidental modification at runtime.
 cfg.freeze()
+
+def update_cfg_from_defaults() -> None:
+    """
+    Update the configuration with current values from IMAGE_DEFAULTS.
+    """
+    cfg.freeze(False)
+
+    # Update all dependent fields from IMAGE_DEFAULTS
+    cfg.TEXT_CONTAINER = IMAGE_DEFAULTS.TEXT_CONTAINER
+    cfg.WORD_MATCHING.PARENTAL_CATEGORIES = IMAGE_DEFAULTS.TEXT_BLOCK_CATEGORIES
+    cfg.TEXT_ORDERING.TEXT_BLOCK_CATEGORIES = IMAGE_DEFAULTS.TEXT_BLOCK_CATEGORIES
+    cfg.TEXT_ORDERING.FLOATING_TEXT_BLOCK_CATEGORIES = IMAGE_DEFAULTS.FLOATING_TEXT_BLOCK_CATEGORIES
+
+    # Re-freeze the configuration
+    cfg.freeze()

deepdoctection/extern/doctrocr.py

@@ -269,7 +269,10 @@ class DoctrTextlineDetector(DoctrTextlineDetectorMixin):
         if self.lib == "PT":
             self.device = get_torch_device(device)
 
-        self.doctr_predictor = self.get_wrapped_model(self.architecture, self.path_weights, self.device, self.lib)
+        self.doctr_predictor = self.get_wrapped_model(self.architecture,
+                                                      self.path_weights,
+                                                      self.device,
+                                                      self.lib)
 
     def predict(self, np_img: PixelValues) -> list[DetectionResult]:
         """
@@ -424,7 +427,7 @@ class DoctrTextRecognizer(TextRecognizer):
         return _get_doctr_requirements()
 
     def clone(self) -> DoctrTextRecognizer:
-        return self.__class__(self.architecture, self.path_weights, self.device, self.lib)
+        return self.__class__(self.architecture, self.path_weights, self.device, self.lib, self.path_config_json)
 
     @staticmethod
     def load_model(

deepdoctection/extern/texocr.py

@@ -60,7 +60,7 @@ def _textract_to_detectresult(response: JsonDict, width: int, height: int, text_
     return all_results
 
 
-def predict_text(np_img: PixelValues, client: boto3.client, text_lines: bool) -> list[DetectionResult]:  # type: ignore
+def predict_text(np_img: PixelValues, client: boto3.client, text_lines: bool) -> list[DetectionResult]:
     """
     Calls AWS Textract client (`detect_document_text`) and returns plain OCR results.
     AWS account required.

deepdoctection/pipe/order.py

@@ -738,10 +738,8 @@ class TextOrderService(TextLineServiceMixin):
                 text_block_anns.extend(residual_text_container_anns)
         for text_block_ann in text_block_anns:
             self.order_text_in_text_block(text_block_ann)
-        floating_text_block_anns_to_order = [
-            ann for ann in text_block_anns if ann.category_name in self.floating_text_block_categories
-        ]
-        self.order_blocks(floating_text_block_anns_to_order)
+        floating_text_block_anns = dp.get_annotation(category_names=self.floating_text_block_categories)
+        self.order_blocks(floating_text_block_anns)
         self._create_columns()
 
     def _create_columns(self) -> None:
@@ -803,9 +801,14 @@ class TextOrderService(TextLineServiceMixin):
         if self.include_residual_text_container:
             add_category.append(LayoutType.LINE)
 
-        assert set(self.floating_text_block_categories) <= set(
+        if set(self.floating_text_block_categories) <= set(
             self.text_block_categories + tuple(add_category)
-        ), "floating_text_block_categories must be a subset of text_block_categories"
+        ):
+            logger.warning("In most cases floating_text_block_categories must be a subset of text_block_categories. "
+                           "Adding categories to floating_text_block_categories, that do not belong to "
+                           "text_block_categories makes only sense for categories set have CHILD relationships with"
+                           " annotations that belong to text_block_categories.")
+
 
     def get_meta_annotation(self) -> MetaAnnotation:
         add_category = [self.text_container]

deepdoctection/utils/context.py

@@ -38,7 +38,7 @@ __all__ = ["timeout_manager", "save_tmp_file", "timed_operation"]
 
 
 @contextmanager
-def timeout_manager(proc: Any, seconds: Optional[int] = None) -> Iterator[str]:  # type: ignore
+def timeout_manager(proc: Any, seconds: Optional[int] = None) -> Iterator[str]:
     """
     Manager for time handling while some process is being called.
 

deepdoctection/utils/develop.py

@@ -71,6 +71,7 @@ def log_deprecated(name: str, text: str, eos: str = "", max_num_warnings: Option
     logger.info(LoggingRecord(f"[Deprecated] {info_msg}"))
 
 
+
 def deprecated(
     text: str = "", eos: str = "", max_num_warnings: Optional[int] = None
 ) -> Callable[[Callable[..., T]], Callable[..., T]]:

deepdoctection/utils/env_info.py

@@ -462,7 +462,7 @@ def pt_info(data: KeyValEnvInfos) -> KeyValEnvInfos:
                 data.append(("torchvision arch flags", msg))
             except (ImportError, AttributeError):
                 data.append(("torchvision._C", "Not found"))
-    except AttributeError:
+    except (AttributeError, ModuleNotFoundError):
         data.append(("torchvision", "unknown"))
 
     return data

deepdoctection/utils/file_utils.py

@@ -12,7 +12,6 @@ import importlib.util
 import multiprocessing as mp
 import string
 import subprocess
-import sys
 from os import environ, path
 from shutil import which
 from types import ModuleType
@@ -22,7 +21,6 @@ import importlib_metadata
 from packaging import version
 
 from .error import DependencyError
-from .logger import LoggingRecord, logger
 from .metacfg import AttrDict
 from .types import PathLikeOrStr, Requirement
 
@@ -662,13 +660,6 @@ def get_doctr_requirement() -> Requirement:
         On macOS, if `poppler` is not available, this function will recursively check the requirement.
         It is not yet known how to check whether `pango`, `gdk-pixbuf`, and `libffi` are installed.
     """
-    if sys.platform == "darwin":
-        if not get_poppler_version():
-            return get_doctr_requirement()
-        # don't know yet how to check whether pango gdk-pixbuf libffi are installed
-        logger.info(
-            LoggingRecord("package requires weasyprint. Check that poppler pango gdk-pixbuf libffi are installed")
-        )
     return "doctr", doctr_available(), _DOCTR_ERR_MSG
 
 

{deepdoctection-0.43.dist-info → deepdoctection-0.43.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepdoctection
-Version: 0.43
+Version: 0.43.1
 Summary: Repository for Document AI
 Home-page: https://github.com/deepdoctection/deepdoctection
 Author: Dr. Janis Meyer
@@ -245,11 +245,11 @@ alt="text" width="40%">
 
 ## Requirements
 
-![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/requirements_deepdoctection_220525.png)
+![requirements](https://github.com/deepdoctection/deepdoctection/raw/master/docs/tutorials/_imgs/install_01.png)
 
 - Linux or macOS. Windows is not supported but there is a [Dockerfile](./docker/pytorch-cpu-jupyter/Dockerfile) available.
 - Python >= 3.9
-- 1.13 \<= PyTorch **or** 2.11 \<= Tensorflow < 2.16. (For lower Tensorflow versions the code will only run on a GPU).
+- 2.2 \<= PyTorch **or** 2.11 \<= Tensorflow < 2.16. (For lower Tensorflow versions the code will only run on a GPU).
   Tensorflow support will be stopped from Python 3.11 onwards.
 - To fine-tune models, a GPU is recommended.
 

{deepdoctection-0.43.dist-info → deepdoctection-0.43.1.dist-info}/RECORD

@@ -1,7 +1,7 @@
-deepdoctection/__init__.py,sha256=T4BXZotL855uGwIHhore8lZAfSinpIcrpeIvrVsSCyc,12910
+deepdoctection/__init__.py,sha256=QYGjP3fSt1deLMEKIb7LUDVlzZgi1Q7phZQADkOjlGk,12964
 deepdoctection/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 deepdoctection/analyzer/__init__.py,sha256=wg0BcFwdCeREwzZfa--Yx8HUJ9LPv5z5PmLwtkZdPH8,772
-deepdoctection/analyzer/config.py,sha256=CgH4etvZL0JdeIHAXMFdDro2VvVsF2itgQar_Ml94pw,41185
+deepdoctection/analyzer/config.py,sha256=Uwi9MYOym0QXzFcip7bRXoy732P4tZVMFBUnUZk9c1w,41761
 deepdoctection/analyzer/dd.py,sha256=2BGvZpl9o9khcaOV52-DPHMrs0DsqUO8cpdqFVHHzDQ,5176
 deepdoctection/analyzer/factory.py,sha256=DI0S38KAG2sIROrSximsWJsMbem91a9zXaeWsDNvkGg,37574
 deepdoctection/configs/__init__.py,sha256=TX_P6tqDOF1LK1mi9ruAl7x0mtv1Asm8cYWCz3Pe2dk,646
@@ -54,7 +54,7 @@ deepdoctection/extern/__init__.py,sha256=jG2qe5_X7fJFnJlx04Lf1KUTXKKKYtCkKR7WQ7l
 deepdoctection/extern/base.py,sha256=vs4EO9vkfyTW-nVM0dnlqsmnPPeHIXI2wW5b4Wpiz-Y,31547
 deepdoctection/extern/d2detect.py,sha256=I0oEkprr5iVpKpM3a3nknAU-sXwNoDQdp_B1gzzODsk,22374
 deepdoctection/extern/deskew.py,sha256=L_jU0rXh03qzwaT79EIqE_zYMUVeFwWDbsGbtahuL2k,3124
-deepdoctection/extern/doctrocr.py,sha256=d468jZpR4WBvZPt5Vze70dh6tZacBM1HO0pcy8ynOFM,25358
+deepdoctection/extern/doctrocr.py,sha256=vOI67NCeEznHzdY4zu1tggwg3_PfwV1eigsr2Z8EPzc,25543
 deepdoctection/extern/fastlang.py,sha256=4D9A-_hTXUcvXG6IJJknX34LrD71v08XtNdWgvXD7fE,4736
 deepdoctection/extern/hfdetr.py,sha256=N3eLNI5BsQS9_7YZyBeWndSgUydJij7ugZA9p4V1xaQ,14316
 deepdoctection/extern/hflayoutlm.py,sha256=3mZZ3byn00jSrLWO2vZFas9j4VrhbYQNmF1mwPG2ElQ,59642
@@ -62,7 +62,7 @@ deepdoctection/extern/hflm.py,sha256=y-9brzmT2NYtFoNcWHABNg2ZZQXSOP9CyqtT1OoeV9U
 deepdoctection/extern/model.py,sha256=-GbnuhLFq7jpBOvtpJe6IhGXxQdqwiM8epEd7IRELoU,18234
 deepdoctection/extern/pdftext.py,sha256=ljzPQn3yYAlS6MoZqzixD-fO2GlHwu1aMiOQ6qMIzbg,7513
 deepdoctection/extern/tessocr.py,sha256=SuPmngsJg38riL4b09z6_FIzJH6H3RIwoighG2GPMYM,17457
-deepdoctection/extern/texocr.py,sha256=93vGj0TX2gENMFV6_FDk3et1sDecrNeuozv5EfOR5nk,5931
+deepdoctection/extern/texocr.py,sha256=wVOuu6eUGao0mUbC8vrgdCsKfY1GqA1Am9560YgWyXU,5915
 deepdoctection/extern/tpdetect.py,sha256=Kr00n80V_OfE-EGfpjiVw1eAQ2n2tuT-hSco-dLSR9E,8516
 deepdoctection/extern/pt/__init__.py,sha256=3Cu0ZHjbYsJomru7-RQXEHihEQLegZrmLetlHiqS58I,742
 deepdoctection/extern/pt/nms.py,sha256=2lSpEH8cI_QXdz5xL_OaitqsGoHhp5xvDssK5Yo8q4Q,2218
@@ -111,7 +111,7 @@ deepdoctection/pipe/doctectionpipe.py,sha256=ik5F92F3klI5Nve_AnyIRj-ApMoKHSR2Sjc
 deepdoctection/pipe/language.py,sha256=T5g5_2GIsbTltAmn_PFymMUMoik8_b0uJNx8f5dT9MM,5898
 deepdoctection/pipe/layout.py,sha256=oAldMtwyZee1IqpuflKKvmeL2Z_nXFiqwFMS4VYv5eI,6391
 deepdoctection/pipe/lm.py,sha256=nYI2bm0sc9d3JMlIPyNyd4XxXFRBIHRUYfMImuek6b4,19793
-deepdoctection/pipe/order.py,sha256=9OarsHKwVqT1bTDIn7XGeGLgpetEJW3uLjuJQDdhjG4,40684
+deepdoctection/pipe/order.py,sha256=8CqEWUA6U9HxThKKGP9yJMbRaML2by7do0Gdhl_7AdI,40964
 deepdoctection/pipe/refine.py,sha256=SrMcAWXRO5tJpqaZCEz9RzvjPyiQiE8fZ9TXBcaBKck,23310
 deepdoctection/pipe/registry.py,sha256=uT5fnHjffoNGk2JPuD2-pMYtO3Iko7-wrwVZVCWLtok,906
 deepdoctection/pipe/segment.py,sha256=rHhEWr5zZ1ppj-gMa-q-UCr1AYTWpUW7oA1umwebqBI,61302
@@ -125,11 +125,11 @@ deepdoctection/train/hf_layoutlm_train.py,sha256=bNL5OCLKytshG6kaTJDLTQOcvWKwEsY
 deepdoctection/train/tp_frcnn_train.py,sha256=Tltb-v2JD5oPuHCZGA9B5DM4ZaidoBITlH93QX-KPKI,13570
 deepdoctection/utils/__init__.py,sha256=brBceRWeov9WXMiJTjyJOF2rHMP8trGGRRjhMdZ61nI,2371
 deepdoctection/utils/concurrency.py,sha256=9ly81D5i2ZFzKfXMIUSmhT42eMs4QttsRhWXdkIk6Aw,5832
-deepdoctection/utils/context.py,sha256=GXgIGQ10JwosE6FQSPdlJoCXyeM_cahEcbQ2mxGNofI,4538
-deepdoctection/utils/develop.py,sha256=4myrqBDypM6tQ2a2Jo3Q20RuE_W2czykpXBwgXPrxNw,3568
-deepdoctection/utils/env_info.py,sha256=yyRyjQT3xz73yEcqauPM6GdCpvWwYGAH-_KUHfXBrtM,19855
+deepdoctection/utils/context.py,sha256=5QfdzxsiSPnNs1qtJdgjguIoD8srLQ2W8oeDzwp9F78,4522
+deepdoctection/utils/develop.py,sha256=x2MhbmoKZyRluesmc01is7ldrUN9c0TX4OAuc1yt6dI,3569
+deepdoctection/utils/env_info.py,sha256=b1WohrfQuoL-BPN0_s8Rjtwzx-WKvCyaX2I4qYl1Emc,19878
 deepdoctection/utils/error.py,sha256=sIry8F5MZ0yLvKfAwVz90IorKWVvjoRqcC0L8qq8mLk,2480
-deepdoctection/utils/file_utils.py,sha256=D4cua4i3Q-4ZeSRXXWEL4z7hp0M1qh9rrVSfS5t-Hzo,25643
+deepdoctection/utils/file_utils.py,sha256=EepfAZVADaqpBdVq2LOJXLFLsMd_oZF_FAKUHOAhiZ0,25246
 deepdoctection/utils/fs.py,sha256=KTS9FJzZk9le_vmIPr9IisJw0AyTfjkyX1KoWQy4DNs,12729
 deepdoctection/utils/identifier.py,sha256=Jt12MeZf7eC1qciY5Fp_AYUGxYVcjsy7xNBUvJil7dU,2270
 deepdoctection/utils/logger.py,sha256=ddQ0xBStluf8OvoRlEB8YkqyRR-ZYgyJYLClTmJJMAU,10290
@@ -142,8 +142,8 @@ deepdoctection/utils/transform.py,sha256=jgeCyQWLN9q79jCGW7jysyKUKcJ1AVMk8OslF-3
 deepdoctection/utils/types.py,sha256=ti4WdtIJSg3TGK_YPkkoY9PYGMnR2tTX6Xfik8U1pNk,2986
 deepdoctection/utils/utils.py,sha256=NBUb1qbx8Jm-AvYN1Sdbk0huXhbAKxZ-ZtOcMespsMM,7064
 deepdoctection/utils/viz.py,sha256=bujRIujvX317rPz4jBrj0yd3WP8wPjDUiI5GUrw9MzQ,27339
-deepdoctection-0.43.dist-info/licenses/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
-deepdoctection-0.43.dist-info/METADATA,sha256=F08hSp-kyzwW1tsTWyRMtWG3TaJsA8_LgueZ7irkBqA,13404
-deepdoctection-0.43.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-deepdoctection-0.43.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
-deepdoctection-0.43.dist-info/RECORD,,
+deepdoctection-0.43.1.dist-info/licenses/LICENSE,sha256=GQ0rUvuGdrMNEI3iHK5UQx6dIMU1QwAuyXsxUHn5MEQ,11351
+deepdoctection-0.43.1.dist-info/METADATA,sha256=jD_6fJFeK-4XpVsjUL0BgFooiuhFPNE9rK0RR6-_2gY,13381
+deepdoctection-0.43.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+deepdoctection-0.43.1.dist-info/top_level.txt,sha256=hs2DdoOL9h4mnHhmO82BT4pz4QATIoOZ20PZmlnxFI8,15
+deepdoctection-0.43.1.dist-info/RECORD,,