Upload 4 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/tapes[[:space:]]5782_EN.pdf filter=lfs diff=lfs merge=lfs -text
+assets/Tesa.png filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,65 @@
+import gradio as gr
+import os
+from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
+from magic_pdf.pipe.UNIPipe import UNIPipe
+from magic_pdf.tools.common import parse_pdf_by_unipipe
+
+def convert_pdf(pdf_file, enable_formula, enable_table, max_pages=100):
+    try:
+        # 获取上传文件路径
+        file_path = pdf_file.name
+        parent_path = os.path.dirname(file_path)
+        
+        # 初始化 MinerU 管道
+        image_writer = DiskReaderWriter(parent_path)
+        jso_useful_key = {
+            "model_list": [],
+            "enable_formula_recognition": enable_formula,
+            "enable_table_recognition": enable_table
+        }
+        pipe = UNIPipe(file_path, jso_useful_key, image_writer=image_writer)
+        
+        # 执行 PDF 解析
+        pipe.pipe_classify()
+        pipe.pipe_parse()
+        
+        # 转换为 Markdown
+        md_content = parse_pdf_by_unipipe(file_path, pipe.pipe_mkユニ, max_pages=max_pages)
+        
+        # 保存 Markdown 到临时文件
+        output_dir = "output"
+        os.makedirs(output_dir, exist_ok=True)
+        output_file = os.path.join(output_dir, f"{os.path.basename(file_path)}.md")
+        with open(output_file, "w", encoding="utf-8") as f:
+            f.write(md_content)
+        
+        return md_content, output_file
+    except Exception as e:
+        return f"Error: {str(e)}", None
+
+# 定义 Gradio 界面
+with gr.Blocks(title="MinerU PDF to Markdown Converter") as iface:
+    gr.Markdown("# MinerU PDF to Markdown Converter")
+    gr.Markdown("Upload a PDF file, configure options, and convert to Markdown.")
+    
+    with gr.Row():
+        with gr.Column():
+            pdf_input = gr.File(label="Upload PDF File", file_types=[".pdf"])
+            formula_checkbox = gr.Checkbox(label="Enable formula recognition", value=False)
+            table_checkbox = gr.Checkbox(label="Enable table recognition", value=False)
+            max_pages = gr.Number(label="Max Pages to Process", value=100, precision=0)
+            convert_button = gr.Button("Convert to Markdown")
+        
+        with gr.Column():
+            markdown_output = gr.Markdown(label="Markdown Output")
+            download_button = gr.File(label="Download Markdown File")
+    
+    # 绑定按钮事件
+    convert_button.click(
+        fn=convert_pdf,
+        inputs=[pdf_input, formula_checkbox, table_checkbox, max_pages],
+        outputs=[markdown_output, download_button]
+    )
+
+# 启动 Gradio 应用
+iface.launch()

assets/tapes 5782_EN.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31b46992626dd2500a7ffdc1157ed6906b29ccad669fb3c59c568d27cdc4bc54
+size 321979

requirements.txt

@@ -0,0 +1,5 @@
+mineru[core]
+gradio==4.44.0
+torch==2.4.1
+numpy==1.26.4
+PyMuPDF==1.24.10