Kenya Inflation Predictor using NNAR + Climate Change
# Inflation Prediction System
This is a final year project for building an inflation prediction system using machine learning and actual monthly CPI data.
## Features
- Predict annual average inflation using monthly data
- Preprocess and clean the provided CSV dataset
- Train a stronger regression model on real inflation statistics
- Streamlit app for interactive prediction
## Setup
1. Create virtual environment: `python -m venv venv`
2. Activate: `venv\Scripts\activate` (Windows)
3. Install dependencies: `pip install -r requirements.txt`
4. Run training: `python model-training.py`
5. Run app: `streamlit run app.py`
## Files
- `app.py`: Streamlit web app for prediction
- `model-training.py`: Script to preprocess the CSV and train the model
- `models/`: Saved trained model file
- `data/`: Dataset folder containing the inflation CSV
## Requirements
- Python 3.8+
- streamlit
- pandas
- numpy
- scikit-learn
- joblib
- matplotlib