🌞 Solaris

Solaris is a machine learning project that predicts solar power output using multimodal data, including time series, tabular sensor data, and potentially vision data (e.g., sky images).

The goal is to build accurate and scalable models to forecast solar energy production for better energy planning and management.

---

📌 Features

• Load and preprocess solar plant data
• Train and evaluate multiple machine learning models
• Compare model performance using metrics like MAE, MSE, and R²
• Visualize results with plots and metrics dashboards

---

🚀 Getting Started

1. Clone the repo and navigate to the project directory
2. Install dependencies (recommended: use uv or pip)
3. Download the dataset into the data/ folder
4. Run the notebooks to explore, train, and evaluate models

---

📁 Project Structure

solaris/
├── data/           # Dataset files
├── notebooks/      # Jupyter notebooks
├── solaris/        # Core source code
└── README.md       # Project description

---

📦 Requirements

• Python 3.12+
• pandas, scikit-learn, matplotlib, seaborn
• tensorflow, xgboost, kagglehub

---

🛠 In Progress

• Adding time series forecasting
• Adding support for image data
• Building a simple dashboard for live predictions

---

`