ML_bioclimatic_predictions

The present folder contains the instructions and the code used for building the dataset and performing the climatic predictions in: 

Linchamps, P., Stoetzel, E., Robinet, F., Hanon, R., Latouche, P., Cornette, R. 

Bioclimatic inference based on mammal community using machine learning regression models: perspectives for paleoecological studies. (submitted)

Data acquisition and computations were done using R 4.0.2 (R Core Team, 2020) and Python 3.8.8 (van Rossum & Drake, 2009). The distinct software packages required to run the experiments can be found in the header of the codes. 

Linear models were fitted and interpreted using R, with the following code:

- linear_models.R

Non-linear models (except neural networks) were fitted and interpreted using Python, with the following code:

- Non_linear_models.py

Neural networks were fitted using the following code:

- NNET.py