ws-bom-robot-app 0.0.99__tar.gz → 0.0.100__tar.gz

{ws_bom_robot_app-0.0.99/ws_bom_robot_app.egg-info → ws_bom_robot_app-0.0.100}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ws_bom_robot_app
-Version: 0.0.99
+Version: 0.0.100
 Summary: A FastAPI application serving ws bom/robot/llm platform ai.
 Home-page: https://github.com/websolutespa/bom
 Author: Websolute Spa
@@ -147,12 +147,12 @@ dockerize base image
 
 ```pwsh
 <# cpu #>
-#docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
-#docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
-docker build -f Dockerfile-robot-base-cpu -t ghcr.io/websolutespa/ws-bom-robot-base:cpu .
+docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
+docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:cpu
 <# gpu #>
-docker build -f Dockerfile-robot-base-gpu -t ghcr.io/websolutespa/ws-bom-robot-base:gpu .
+docker build -f Dockerfile-robot-base-gpu -t ws-bom-robot-base:gpu .
+docker tag ws-bom-robot-base:gpu ghcr.io/websolutespa/ws-bom-robot-base:gpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:gpu
 ```
 

{ws_bom_robot_app-0.0.99 → ws_bom_robot_app-0.0.100}/README.md

@@ -80,12 +80,12 @@ dockerize base image
 
 ```pwsh
 <# cpu #>
-#docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
-#docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
-docker build -f Dockerfile-robot-base-cpu -t ghcr.io/websolutespa/ws-bom-robot-base:cpu .
+docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
+docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:cpu
 <# gpu #>
-docker build -f Dockerfile-robot-base-gpu -t ghcr.io/websolutespa/ws-bom-robot-base:gpu .
+docker build -f Dockerfile-robot-base-gpu -t ws-bom-robot-base:gpu .
+docker tag ws-bom-robot-base:gpu ghcr.io/websolutespa/ws-bom-robot-base:gpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:gpu
 ```
 

{ws_bom_robot_app-0.0.99 → ws_bom_robot_app-0.0.100}/setup.py

@@ -4,7 +4,7 @@ _requirements = [line.split('#')[0].strip() for line in open("requirements.txt")
 
 setup(
     name="ws_bom_robot_app",
-    version="0.0.99",
+    version="0.0.100",
     description="A FastAPI application serving ws bom/robot/llm platform ai.",
     long_description=open("README.md", encoding='utf-8').read(),
     long_description_content_type="text/markdown",

{ws_bom_robot_app-0.0.99 → ws_bom_robot_app-0.0.100}/ws_bom_robot_app/llm/vector_store/loader/base.py

@@ -117,11 +117,14 @@ class Loader():
     #region log
     import warnings
     warnings.filterwarnings("ignore", message=".*pin_memory.*no accelerator is found.*")
+    warnings.filterwarnings("ignore", category=UserWarning)
     log_msg_to_ignore = [
         "Going to convert document batch...",
         "Initializing pipeline for",
         "Accelerator device:",
         "detected formats:",
+        "The text detection result is empty",
+        "RapidOCR returned empty result!",       
     ]
     class MessageFilter(logging.Filter):
         def __init__(self, patterns):
@@ -139,8 +142,11 @@ class Loader():
         'docling.document_converter',
         'docling.datamodel',
         'docling.datamodel.document',
+        'docling.models',
+        'docling.models.rapidocr_model',
         'docling.utils.accelerator_utils',
-        'unstructured'
+        'unstructured',
+        'RapidOCR'
     ]
     for logger_name in loggers_to_filter:
         logging.getLogger(logger_name).addFilter(message_filter)

{ws_bom_robot_app-0.0.99 → ws_bom_robot_app-0.0.100}/ws_bom_robot_app/llm/vector_store/loader/docling.py

@@ -8,31 +8,39 @@ from docling.datamodel.accelerator_options import AcceleratorDevice, Accelerator
 from docling.datamodel.pipeline_options import PdfPipelineOptions, TableStructureOptions, TableFormerMode
 from langchain_community.document_loaders import UnstructuredFileLoader
 from ws_bom_robot_app.llm.vector_store.db.base import VectorDBStrategy
-from docling.datamodel.pipeline_options import TableStructureOptions, TableFormerMode, EasyOcrOptions
+from docling.datamodel.pipeline_options import TableStructureOptions, TableFormerMode, RapidOcrOptions
 from docling.document_converter import DocumentConverter, PdfFormatOption, ImageFormatOption
 
 def _doclingConverter() -> DocumentConverter:
-  pipeline_options = PdfPipelineOptions(
-      accelerator_options=AcceleratorOptions(
+  _pipeline_config = {
+      "accelerator_options": AcceleratorOptions(
           device=AcceleratorDevice.AUTO,
           cuda_use_flash_attention2=False,
       ),
-      table_structure_options=TableStructureOptions(mode=TableFormerMode.FAST),
-      #do_ocr=False,
-      ocr_options=EasyOcrOptions()
-  )
+      "table_structure_options": TableStructureOptions(mode=TableFormerMode.ACCURATE),
+  }
+  _base_pipeline_options = PdfPipelineOptions(
+      **_pipeline_config,
+      do_ocr=False)  
+  _ocr_pipeline_options = PdfPipelineOptions(
+      **_pipeline_config,
+      ocr_options=RapidOcrOptions(
+         print_verbose=False,         
+         text_score=0.5,
+         #rapidocr_params={"det_use_cuda": True}
+         ))  
   doc_converter = DocumentConverter(
-      format_options={
+      format_options={          
           InputFormat.PDF: PdfFormatOption(
-              pipeline_options=pipeline_options,
+              pipeline_options=_base_pipeline_options,
           ),
           InputFormat.IMAGE: ImageFormatOption(
-              pipeline_options=pipeline_options,
+              pipeline_options=_ocr_pipeline_options,
           ),
       }
   )
-  doc_converter.initialize_pipeline(InputFormat.PDF)
-  doc_converter.initialize_pipeline(InputFormat.IMAGE)
+  for frm in [InputFormat.PDF, InputFormat.IMAGE]:
+     doc_converter.initialize_pipeline(frm)
   return doc_converter
 
 class DoclingLoader(BaseLoader):

{ws_bom_robot_app-0.0.99 → ws_bom_robot_app-0.0.100/ws_bom_robot_app.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ws_bom_robot_app
-Version: 0.0.99
+Version: 0.0.100
 Summary: A FastAPI application serving ws bom/robot/llm platform ai.
 Home-page: https://github.com/websolutespa/bom
 Author: Websolute Spa
@@ -147,12 +147,12 @@ dockerize base image
 
 ```pwsh
 <# cpu #>
-#docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
-#docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
-docker build -f Dockerfile-robot-base-cpu -t ghcr.io/websolutespa/ws-bom-robot-base:cpu .
+docker build -f Dockerfile-robot-base-cpu -t ws-bom-robot-base:cpu .
+docker tag ws-bom-robot-base:cpu ghcr.io/websolutespa/ws-bom-robot-base:cpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:cpu
 <# gpu #>
-docker build -f Dockerfile-robot-base-gpu -t ghcr.io/websolutespa/ws-bom-robot-base:gpu .
+docker build -f Dockerfile-robot-base-gpu -t ws-bom-robot-base:gpu .
+docker tag ws-bom-robot-base:gpu ghcr.io/websolutespa/ws-bom-robot-base:gpu
 docker push ghcr.io/websolutespa/ws-bom-robot-base:gpu
 ```