deepdoctection 1.2.6__tar.gz → 1.2.8__tar.gz

{deepdoctection-1.2.6 → deepdoctection-1.2.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepdoctection
-Version: 1.2.6
+Version: 1.2.8
 Summary: Repository for Document AI - server/inference core package
 Author: Dr. Janis Meyer
 License: Apache License 2.0

{deepdoctection-1.2.6 → deepdoctection-1.2.8}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "deepdoctection"
-version = "1.2.6"
+version = "1.2.8"
 authors = [
     {name = "Dr. Janis Meyer"}
 ]

{deepdoctection-1.2.6 → deepdoctection-1.2.8}/src/deepdoctection/__init__.py

@@ -12,7 +12,7 @@ from dd_core.utils.env_info import collect_env_info
 from dd_core.utils.file_utils import _LazyModule
 from dd_core.utils.logger import LoggingRecord, logger
 
-__version__ = "1.2.6"
+__version__ = "1.2.8"
 _IMPORT_STRUCTURE = {
     "analyzer": ["config_sanity_checks", "get_dd_analyzer", "ServiceFactory", "update_cfg_from_defaults"],
     "eval": [

{deepdoctection-1.2.6 → deepdoctection-1.2.8}/src/deepdoctection/pipe/anngen.py

@@ -95,9 +95,15 @@ def _set_image_keys_to_none(d: Any) -> None:
             _set_image_keys_to_none(item)
 
 
-def _image_to_cache_dict(image: Image) -> dict[str, Any]:
+def image_to_cache_dict(image: Image) -> dict[str, Any]:
+    """
+    Converting image to dict, by adding extras and removing higher hierarchies
+
+    Args:
+        image (Image): The image object to serialize and store.
+    """
     image.remove_image_from_lower_hierarchy()
-    export_dict = image.as_dict()
+    export_dict = image.as_dict(add_extras=True)
     _set_image_keys_to_none(export_dict)
     return export_dict
 
@@ -121,7 +127,7 @@ class LocalDataPointCacheStore(DataPointCacheStore):
         Args:
             max_pages (int): Maximum number of pages to keep per document.
         """
-        self._max_pages = max_pages
+        self.max_pages = max_pages
         self._pages: dict[str, dict[int, dict[str, Any]]] = {}
 
     def _get_cache_key(self, document_id: str, job_id: str | None) -> str:
@@ -158,9 +164,9 @@ class LocalDataPointCacheStore(DataPointCacheStore):
         if pages is None:
             pages = {}
             self._pages[cache_key] = pages
-        pages[page_number] = _image_to_cache_dict(image)
-        if self._max_pages > 0 and len(pages) > self._max_pages:
-            for k in sorted(pages.keys())[: -self._max_pages]:
+        pages[page_number] = image_to_cache_dict(image)
+        if self.max_pages > 0 and len(pages) > self.max_pages:
+            for k in sorted(pages.keys())[: -self.max_pages]:
                 pages.pop(k, None)
 
     def get_datapoints(self, document_id: str, last_d: int, job_id: str | None = None) -> tuple[Image, ...]:
@@ -224,10 +230,10 @@ class DatapointManager:
 
         if num_cached_datapoints < 0:
             raise ValueError("num_cached_datapoints must be >= 0")
-        self.num_cached_datapoints = num_cached_datapoints
         self.remove_pixel_values_from_cache = remove_pixel_values_from_cache
 
         self._cache_store = cache_store or LocalDataPointCacheStore(max_pages=num_cached_datapoints)
+        self.num_cached_datapoints = self._cache_store.max_pages
 
     def maybe_cache_datapoint(self, image: Optional[Image], job_id: str | None = None) -> None:
         """

{deepdoctection-1.2.6 → deepdoctection-1.2.8}/src/deepdoctection.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deepdoctection
-Version: 1.2.6
+Version: 1.2.8
 Summary: Repository for Document AI - server/inference core package
 Author: Dr. Janis Meyer
 License: Apache License 2.0