README.md

metadata

license: mit
title: FillMyBlank.ai
sdk: streamlit
emoji: 😻
colorFrom: yellow
colorTo: blue

FillMyBlank.ai MVP: Local Lore & Proverb Collector

Project Overview

Welcome to the FillMyBlank.ai MVP: Local Lore & Proverb Collector! This application is a core component of the FillMyBlank.ai vision to build an AI that deeply understands India's rich linguistic and cultural diversity.

Objective: To create a "Corpus Collection Engine" disguised as a useful and engaging application. Users can easily contribute local proverbs, folk tales, historical facts, and other cultural lore from various Indian languages and regions.

How it helps FillMyBlank.ai: The text data collected through this app (lore, language, region) directly feeds into the dataset used for training advanced AI models. This raw, diverse, and authentic data is crucial for developing AI that can generate culturally relevant and linguistically nuanced fill-in-the-blank content for FillMyBlank.ai in the future.

Features

• Simple Data Submission: Easy-to-use form for contributing text lore.
• Metadata Tagging: Users can specify the language, region, and type of lore for better data organization.
• Lightweight & Accessible: Designed to be minimal and performant, making it more accessible in low-bandwidth environments.
• Direct Corpus Contribution: Every submission directly adds valuable data to our corpus.
• Recent Lore Display: Shows a few of the latest contributions to give users a sense of community.

AI Integration (MVP Focus)

In this Minimum Viable Product (MVP), the primary "AI integration" is the strategic collection of high-quality, diverse linguistic and cultural data. This data is the foundation upon which future, more complex AI models for FillMyBlank.ai will be built. The app serves as the crucial input mechanism for our AI training pipeline.

How to Deploy on Hugging Face Spaces

Follow these steps to deploy your own instance of the FillMyBlank.ai Local Lore & Proverb Collector:

1. Create a New Space:

   • Go to Hugging Face Spaces.
   • Click on "Create new Space".
   • Give your Space a name (e.g., your-username/local-lore-collector).
   • Select the Streamlit SDK.
   • Choose a Public license.
   • Click "Create Space".

2. Upload Your Files:

   • Once your Space is created, you'll see options to add files.
   • Click on "Add File" -> "Upload file".
   • Upload the app.py, requirements.txt, and lore_data.jsonl files. (You've already created the README.md now).

3. Automatic Deployment:

   • Once app.py and requirements.txt are uploaded, Hugging Face Spaces will automatically detect them and start building your application.
   • Give it a few minutes to build. You'll see a build log, and once it's done, your app will be live!

4. Access Your App:

   • Your live app will be accessible at https://huggingface.co/spaces/your-username/your-space-name (replace with your actual username and space name).

Local Development (Optional)

If you want to run this app locally on your machine:

1. Clone the repository: (If you downloaded the files, just navigate to the fillmyblank_mvp folder)

   git clone [https://github.com/your_username/your_repo_name.git](https://github.com/your_username/your_repo_name.git) # If you push it to GitHub first
   cd fillmyblank_mvp
   

2. Create lore_data.jsonl: Ensure this empty file exists in your project directory.

   touch lore_data.jsonl
   

3. Install dependencies:

   pip install -r requirements.txt
   

4. Run the Streamlit app:

   streamlit run app.py
   

   This will open the app in your local web browser.

Future Roadmap (Post-MVP)

• Advanced Data Validation: More robust checks for input data.
• User Accounts: Implement user registration and login for personalized contributions and saved lore.
• AI Suggestions: Integrate AI models to suggest categories, languages, or even fill in parts of the lore.
• Community Features: Upvoting, commenting, and sharing of submitted lore.
• Admin Dashboard: For reviewing and curating submitted data.
• Scalable Database: Migrate from JSONL file to a proper database (e.g., PostgreSQL, MongoDB) for production.