fawazsanu/Solar-Energy-Prediction — reverse-engineered prompt

Build me a Python project that predicts how much DC power a solar plant is generating based on weather and time of day.

I want it to train on the Kaggle Solar Power Generation Data for Plant 1, combine the generation and weather sensor files by timestamp, create useful time and irradiation features, and train a strong forecasting model using XGBoost or a similar approach. Please evaluate it properly for time based data, using the earlier records for training and the later records for testing, and print clear results like R2, RMSE, and MAE so I can see how good it is.

Also make a simple live prediction script where I can enter a location and an OpenWeatherMap API key, fetch the current weather, and get an estimated solar power output in kW. Include a saved model, requirements file, clear setup steps, and friendly console output. Look up current docs online if you need to.