GoldPredictAPI 📈🤖

This project implements a gold price prediction system using machine learning. The system is trained on historical gold price data and provides a FastAPI-based API for making predictions.

Files 📂

• gld_predication.py: FastAPI app with a pre-trained model, offering an API endpoint for gold price predictions.

• Gold Price Prediction.ipynb: Jupyter Notebook for initial gold price data exploration, preprocessing, and training of the RandomForestRegressor model.

Dependencies 🛠️

• FastAPI
• Scikit-learn
• NumPy
• Pandas
• Matplotlib
• Seaborn

How to Use 🚀

1. Install dependencies:

   pip install -r requirements.txt

2. Run the FastAPI application:

   uvicorn gld_predication:app --reload

   This will start the FastAPI server locally.

3. Make predictions using Swagger:

   Open your web browser and go to http://127.0.0.1:8000/docs to access the Swagger UI.

   • Click on the /predict endpoint.

   • Click on the "Try it out" button.

   • Input the sample request data:

     {
       "SPX": 2671.91992,
       "USO": 14.0600,
       "SLV": 15.5100,
       "EUR_USD": 1.186789
     }

   • Click on the "Execute" button to make a prediction.

4. Make predictions using curl:

   Alternatively, you can use curl to make predictions:

   curl -X 'POST' \
     'http://127.0.0.1:8000/predict' \
     -H 'accept: application/json' \
     -H 'Content-Type: application/json' \
     -d '{
     "SPX": 2671.91992,
     "USO": 14.0600,
     "SLV": 15.5100,
     "EUR_USD": 1.186789
   }'

Additional Information ℹ️

• Colab Notebook: The Gold Price Prediction.ipynb file in Google Colab contains the initial exploration and model training.

• Model Saving: The trained model is saved as gld_data.pkl using the pickle library and loaded by the FastAPI application for predictions.

License 📄

This project is licensed under the MIT License - see the LICENSE file for details.

---