fix: skip problematic import

.pre-commit-config.yaml

@@ -29,7 +29,7 @@ repos:
     rev: 4.0.1
     hooks:
       - id: flake8
-        args: ["--max-line-length", "88", "--extend-ignore", "E203"]
+        args: ["--max-line-length", "88", "--extend-ignore", "E203,E402"]
   - repo: https://github.com/myint/autoflake
     rev: v1.4
     hooks:

README.md

@@ -1,13 +1,13 @@
 ---
-title: DoclingConverter
+title: PDFParsersPlayground
 emoji: 🐢
 colorFrom: blue
-colorTo: red
+colorTo: green
 sdk: gradio
 sdk_version: 5.7.1
 app_file: app.py
 pinned: false
-short_description: Convert documents to Markdown or JSON with metadata
+short_description: Convert PDF documents to Markdown with multiple open-source parsers
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -1,3 +1,8 @@
+from utils import fix_problematic_imports  # noqa
+
+fix_problematic_imports()  # noqa
+
+
 import time
 from pathlib import Path
 
@@ -46,9 +51,15 @@ def convert_document(path, method, enabled=True):
     elif method == "MinerU":
         text, debug_image_paths = convert_mineru(path, file_name)
 
-    end = time.time()
-    print(f"Conversion with {method} took {end - start} seconds")
-    return text, remove_images_from_markdown(text), debug_image_paths
+    duration = time.time() - start
+    duration_message = f"Conversion with {method} took *{duration:.2f} seconds*"
+    print(duration_message)
+    return (
+        duration_message,
+        text,
+        remove_images_from_markdown(text),
+        debug_image_paths,
+    )
 
 
 def show_tabs(selected_methods):
@@ -73,7 +84,8 @@ print("Warm-up sequence")
 for method in SUPPORTED_METHODS:
     for _ in range(1):
         convert_document(WARMUP_PDF_PATH, method)
-print("Start up time", time.time() - start_startup, "seconds")
+startup_duration = time.time() - start_startup
+print(f"Total start-up time: {startup_duration:.2f} seconds")
 
 with gr.Blocks(
     theme=gr.themes.Ocean(),
@@ -149,9 +161,19 @@ with gr.Blocks(
                                 markdown_text = gr.TextArea(
                                     lines=45, show_label=False, container=False
                                 )
+                            with gr.Tab("Reference"):
+                                output_description = gr.Markdown(
+                                    container=False,
+                                    show_label=False,
+                                )
 
                     output_components.extend(
-                        [markdown_render, markdown_text, debug_images]
+                        [
+                            output_description,
+                            markdown_render,
+                            markdown_text,
+                            debug_images,
+                        ]
                     )
                     output_tabs.append(output_tab)
                     visualization_sub_tabs.append(visual_sub_tab)
@@ -199,7 +221,7 @@ with gr.Blocks(
                 input_file, methods, method
             ),
             inputs=[input_file, methods],
-            outputs=output_components[idx * 3 : (idx + 1) * 3],
+            outputs=output_components[idx * 4 : (idx + 1) * 4],
         )
 
     click_event.then(

requirements.txt

@@ -1,5 +1,3 @@
-torch>=2.2.2,<=2.3.1
-torchvision>=0.17.2,<=0.18.1
 paddlepaddle-gpu @ https://paddle-whl.bj.bcebos.com/stable/cu118/paddlepaddle-gpu/paddlepaddle_gpu-3.0.0b1-cp310-cp310-linux_x86_64.whl
 detectron2 @ https://wheels-1251341229.cos.ap-shanghai.myqcloud.com/assets/whl/detectron2/detectron2-0.6-cp310-cp310-linux_x86_64.whl
 paddleocr==2.7.3
@@ -14,7 +12,7 @@ PyMuPDF>=1.24.9,<1.24.14
 pymupdf4llm
 unstructured[pdf]
 ultralytics>=8.3.48
-unimernet==0.2.3
 transformers<5.0.0,>=4.45.2
 struct-eqtable==0.3.2
 openai
+doclayout_yolo==0.0.2b1

utils.py

@@ -29,3 +29,21 @@ def trim_pages(pdf_path, output_path, trim_pages=5):
         copy2(pdf_path, str(output_file_path))
 
     return str(output_file_path)
+
+
+def fix_problematic_imports():
+    import sys
+    import types
+
+    # Create a fake 'UnimernetModel' class inside a fake 'Unimernet' module
+    fake_unimernet_module = types.ModuleType(
+        "magic_pdf.model.sub_modules.mfr.unimernet.Unimernet"
+    )
+    fake_unimernet_module.UnimernetModel = type(  # type: ignore
+        "UnimernetModel", (), {}
+    )
+
+    # Register fake module in sys.modules
+    sys.modules[
+        "magic_pdf.model.sub_modules.mfr.unimernet.Unimernet"
+    ] = fake_unimernet_module