solar

AWS X ZELESTRA HACKATHON PROJECT

Approach:

This project is focused on predicting solar panel efficiency using machine learning models. The workflow involves:

1. Data cleaning and preprocessing
2. Exploratory Data Analysis (EDA)
3. Feature engineering to extract meaningful variables
4. Model training and evaluation using multiple algorithms (e.g., Random Forest, XGBoost)

Feature Engineering:

• Handling missing values through imputation
• Encoding categorical variables using one-hot encoding
• Scaling numeric features
• Creating new features such as temperature difference, irradiance ratios, etc.

Tools Used:

• Python (Jupyter Notebook)
• pandas, numpy for data manipulation
• matplotlib, seaborn for visualization
• scikit-learn for ML models
• xgboost for gradient boosting