{"id":23174662,"url":"https://github.com/mansurpro/docuparse","last_synced_at":"2026-03-20T00:54:50.780Z","repository":{"id":266921653,"uuid":"899771407","full_name":"MansurPro/DocuParse","owner":"MansurPro","description":"DocuParse is a high-performance tool for converting PDF documents into clean, structured Markdown files. Designed for speed and accuracy, it extracts and formats content while minimizing errors like hallucinations and repetitions.","archived":false,"fork":false,"pushed_at":"2025-08-10T16:12:07.000Z","size":124,"stargazers_count":1,"open_issues_count":1,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-08-10T17:27:23.019Z","etag":null,"topics":["digital-archive","document-layout-analysis","google-colab","huggingface-transformers","markdown-conversion","pdf-parsing","pdf-to-markdown","tesseract-ocr","text-extraction"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/MansurPro.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-12-07T01:30:12.000Z","updated_at":"2025-08-10T16:12:14.000Z","dependencies_parsed_at":"2024-12-07T02:24:28.951Z","dependency_job_id":"c1224893-a685-43ac-ab60-ef64c1e65254","html_url":"https://github.com/MansurPro/DocuParse","commit_stats":null,"previous_names":["mansurpro/docuparse"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/MansurPro/DocuParse","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MansurPro%2FDocuParse","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MansurPro%2FDocuParse/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MansurPro%2FDocuParse/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MansurPro%2FDocuParse/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MansurPro","download_url":"https://codeload.github.com/MansurPro/DocuParse/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MansurPro%2FDocuParse/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28572570,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-19T14:39:55.009Z","status":"ssl_error","status_checked_at":"2026-01-19T14:39:01.217Z","response_time":67,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["digital-archive","document-layout-analysis","google-colab","huggingface-transformers","markdown-conversion","pdf-parsing","pdf-to-markdown","tesseract-ocr","text-extraction"],"created_at":"2024-12-18T05:30:53.058Z","updated_at":"2026-01-19T15:01:39.947Z","avatar_url":"https://github.com/MansurPro.png","language":"Python","readme":"\n# 📄 **DocuParse**\n\nDocuParse is a high-performance tool for converting PDF documents into clean, structured Markdown files. Designed for speed and accuracy, it extracts and formats content while minimizing errors like hallucinations and repetitions.\n\n---\n\n## 🚀 **Key Features**\n\n- **Multi-Format Support**: Converts PDFs, EPUBs, and MOBIs into Markdown.\n- **Accurate Layout Detection**: Utilizes AI models to detect page layouts, columns, and format equations in LaTeX.\n- **Enhanced Formatting**: Cleans headers, footers, and artifacts, while preserving code blocks and tables.\n- **Multi-Language Support**: Processes documents in various languages, optimized for English, French, Spanish, and more.\n- **Cloud-Based Processing**: Leverages Google Colab for GPU-accelerated operations.\n\n---\n\n## 🛠️ **How It Works**\n\nDocuParse is powered by a robust AI pipeline:\n1. **Text Extraction**: Extracts text with or without OCR as needed.\n2. **Layout Analysis**: Identifies and segments content using advanced AI models.\n3. **Content Cleaning**: Applies heuristics to clean and format content blocks.\n4. **Post-Processing**: Combines content blocks into a structured Markdown document.\n\n---\n\n## 📂 **Project Structure**\n\n```plaintext\nDocuParse/\n│\n├── scripts/             # Utility scripts for setup and processing\n├── data/                # Sample input and output files\n├── examples/            # Example documents and Markdown outputs\n├── models/              # AI models used for text extraction and layout analysis\n└── README.md            # Project documentation\n```\n\n---\n\n## 🖥️ **Usage**\n\n### **1. Clone the Repository**\n\n```bash\ngit clone https://github.com/MansurPro/DocuParse.git\ncd DocuParse\n```\n\n### **2. Set Up the Environment**\n\nRun DocuParse in Google Colab for GPU-accelerated processing. Install the required Python packages:\n\n```bash\npip install -r requirements.txt\n```\n\n### **3. Convert a Single File**\n\nUse the ``docuparse`` CLI to convert a PDF into Markdown:\n\n```bash\npython -m docuparse.cli convert /path/to/file.pdf /path/to/output.md --max-pages 10\n```\n\n### **4. Run the API Server**\n\nStart a FastAPI server that exposes a ``/convert`` endpoint:\n\n```bash\nuvicorn docuparse.fastapi_app:app --reload\n```\n\n### **5. Launch the Gradio UI**\n\n```bash\npython -m docuparse.gradio_app\n```\n\n---\n\n## 🎨 **Examples**\n\n| **Input (PDF)**                        | **Output (Markdown)**                |\n|----------------------------------------|---------------------------------------|\n| Textbook: Think Python                 | [View](examples/thinkpython.md)       |\n| Scientific Paper: Switch Transformers  | [View](examples/switch_transformers.md) |\n| Simple Example | [View](examples/sample.md) |\n\n---\n\n## 📊 **Performance**\n\n- **Speed**: DocuParse processes documents up to **10x faster** than similar tools.\n- **Accuracy**: Minimizes hallucinations and ensures well-structured Markdown outputs.\n- **GPU Utilization**: Efficiently leverages GPU resources for parallel processing.\n\n---\n\n## 📜 **License**\n\nThis project is licensed under the MIT License. See the `LICENSE` file for details.\n\n---\n\n## 🙌 **Acknowledgments**\n\nThis project was made possible by incredible open-source models and datasets, including:\n- **Tesseract OCR** for text recognition.\n- **HuggingFace Transformers** for layout and content analysis.\n- **LaTeX** for equation formatting.\n\nThank you to the open-source community for their invaluable contributions!\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmansurpro%2Fdocuparse","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmansurpro%2Fdocuparse","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmansurpro%2Fdocuparse/lists"}