up

README.md

@@ -6,40 +6,3 @@ sdk: docker
 pinned: false
 license: unknown
 short_description: doodle digits to identified number
-Digit Doodle App
-Overview
-The Digit Doodle App is an interactive web application that allows users to draw digits and get real-time predictions using a trained machine learning model. The app provides a simple and engaging way to explore AI-powered handwriting recognition.
-
-Features
-Interactive Drawing Canvas: Users can doodle digits on a web-based canvas.
-Real-Time Prediction: The app processes the drawing and predicts the digit using a trained model.
-Machine Learning Integration: Uses a deep learning model trained on handwritten digit datasets.
-Easy Deployment: Hosted on PythonAnywhere for online access.
-Tech Stack
-Frontend: HTML, CSS, JavaScript (Canvas API)
-Backend: Python Flask
-Machine Learning: PyTorch/
-Installation
-To set up the project locally:
-
-Clone the repository:
-git clone https://huggingface.co/spaces/sivakum4/numberclassifier
-cd numberclassifier
-Create a virtual environment and install dependencies:
-python -m venv venv
-source venv/bin/activate
-pip install -r requirements.txt
-Run the app locally:
-python server.py
-Open your browser and go to http://127.0.0.1:7860/ to start doodling.
-Usage
-Open the app in a browser.
-Draw a digit (0-9) in the provided canvas.
-Click the "Predict" button to get the model’s prediction.
-Clear the canvas and try again!
-Future Enhancements
-Support for multi-digit recognition.
-Enhance UI for better user experience.
-Contributing
-Contributions are welcome! Feel free to fork the repo, create feature branches, and submit pull requests.
-