Data from: Interpretable machine learning reveals context-dependent decision-making by migratory songbirds

This dataset contains data and analysis code used to generate the results presented in our paper "Context-dependent decision-making in migratory birds: The roles of regional climate and preceding weather". It is structured as follows: data: preproceseed datasets used to train, test, and analyse FluxRGNN+ bird migration models preprocessed: hourly NEXRAD weather radar data combined with ERA5 reanalysis data and H3 tessellation details shapes: geographical shapes used for plotting raw: raw NEXRAD weather radar data used to generate the preprocessed datasets. nexrad: vertically integrated bird densities and average velocities used to train and evaluate FluxRGNN+ models nexrad_vp_avg: average vertical profiles per radar and season used to determine ERA5 model levels analysis: all relevant data and code used for the final analysis of FluxRGNN+ predictions. The environment.yml file can be used to install all required Python packages with conda. artifacts: trained models, predictions, and Shapley values. The .csv files in the analysis folder can be used to find specific files in this folder (e.g. the file final_model_artifacts.csv contains the path to the trained models for spring and fall migration seasons respectively) shap_values: processed (filtered and grouped) Shapley values generated by the script process_explanations.py plots: figures generated by the Jupyter notebooks evaluate_models.ipynb, plot_predictions.ipynb, and analyse_explanations.ipynb Model training and Shapley value computations were performed using the FluxRGNN and shap Github repositories.