Wind farm production analysis

The purpose of this repository is to explore different approaches for modeling time series data on wind farm energy production.

I created 3 main pieces of code: a Jupyter notebook, a Python file for the estimator class and a pytest test file.

1. The notebook called Wind farm production forecast contains the initial data exploration, data preparation, model selection and performance assessment parts.

2. The estimator class MyLRNNRegression.py implements an ensemble model based on two models explored in the notebook.

3. In the test file test_MyLRNNRegression.py I show an example of a unit test ran with pytest.

Furthermore, the code in the notebook Get weather data retrieves weather data from the weather station Hemsby in the UK, through the meteostat API. The weather data is stored in hemsby_hourly_resampled.csv.