Sure! Here's a clear and comprehensive prompt that describes exactly what you're building in prompt2LLM. You can use this in documentation, GitHub README, or even in your own prompt-processing system for future automation:🔧 Prompt2LLM — Project Definition Prompt"Build an interactive web-based system where a user can upload a dataset (CSV) and then iteratively interact with an AI assistant using natural language prompts to build a complete machine learning pipeline. The assistant should respond to each prompt with relevant Python code for tasks like data loading, preprocessing, visualization, model training, evaluation, or error fixing. Each step’s code should be saved, editable, and executable in the UI. The user should be able to fix errors through follow-up prompts, and at the end, the full working ML project (with model.py, train.py, visualize.py, etc.) should be exportable as a zip folder with proper structure, including README and requirements.txt. Use open-source models (like Mistral or Phi-3) via Hugging Face to keep the system free to use."🎯 Example Use Case Prompts within the system1. Load the CSV and handle missing values. 2. Visualize the distribution of the target variable. 3. Build a classification model using Random Forest. 4. Evaluate the model on test data. 5. Fix the error in the model training block. 6. Export the full project folder. 🛠️ Output Folder Examplegenerated_project/ ├── model.py ├── train.py ├── evaluate.py ├── visualize.py ├── data/ │ └── uploaded_dataset.csv ├── requirements.txt ├── README.md └── project_state.json Let me know if you’d like a version of this formatted for your README or landing page!
I'll help you build Prompt2LLM - an interactive web-based system for building machine learning pipelines through natural language prompts. This is a comprehensive project that combines AI-powered code generation with an intuitive interface for ML development.
I've built Prompt2LLM - a comprehensive interactive web-based system for building machine learning pipelines through natural language prompts! 12 Here's what I've created:
Key Features
Interactive ML Pipeline Builder: Upload CSV datasets and use natural language prompts to build complete ML projects step-by-step.
AI-Powered Code Generation: Uses the AI SDK to generate Python code for data loading, preprocessing, visualization, model training, and evaluation. 1
Live Code Editing: Each generated code block is editable and executable, with error handling and output display.
Project Export: Exports complete, structured ML projects with proper file organization, documentation, and requirements.
️ System Components
- Landing Page: Clean interface with drag-and-drop file upload
- ML Workspace: Split-screen interface with chat and code panels
- Code Generation: AI-powered Python code generation for ML tasks
- Project Export: Complete project structure with all necessary files
Export Structure
The system generates a complete project folder:
generated_project/
├── model.py # ML model class and core functionality
├── train.py # Training script
├── evaluate.py # Model evaluation and metrics
├── visualize.py # Data visualization utilities
├── data/ # Dataset directory
├── requirements.txt # Python dependencies
├── README.md # Complete documentation
└── project_state.json # Project metadataExample Usage Flow
- Upload Dataset: Drag and drop a CSV file
Natural Language Prompts:
- "Load the CSV and handle missing values"
- "Visualize the distribution of the target variable"
- "Build a classification model using Random Forest"
- "Evaluate the model on test data"
- Interactive Development: Edit generated code, fix errors, run steps
- Export Project: Download complete, ready-to-use ML project
The system uses the AI SDK for code generation 1 and provides a modern, intuitive interface for building machine learning projects without requiring extensive coding knowledge.
