This repository contains the code to reproduce the results in Pichler and Hartig, Machine Learning and Deep Learning -- A review for Ecologists.
You can find classification and regression examples for common ML (+DL) algorithms in the docs subfolder or under the following link: https://maximilianpi.github.io/Pichler-and-Hartig-2022/
ML algorithms:
| ML algorithm | Task | Language | Link |
|---|---|---|---|
| Elastic-net/LASSO/Ridge | Classification, Regression | R, Python, Julia | link |
| Support Vector Machine | Classification, Regression | R, Python, Julia | link |
| k-nearest-neighbor | Classification, Regression | R, Python, Julia | link |
| Random Forest | Classification, Regression | R, Python, Julia | link |
| Boosted Regression Trees | Classification, Regression | R, Python, Julia | link |
| Deep Neural Networks | Classification, Regression | R, Python, Julia | link |
| Convolutional Neural Networks | Multi-class/label | R, Python, Julia | link |
| Recurrent Neural Network | Time-series forecasting | R, Python, Julia | link |
| Graph Neural Network | Node classicifaction | R, Python | link |
Link to the pre-print:
https://arxiv.org/abs/2204.05023
Scripts to fetch/create the trend data and make the figures (trend figures and wordclouds) can be found in the 'Figures' folder:
source("Figures/fetch_data.R") # get trend data
source('Figures/Figures.R') # create figures
Script to create the simulation and run the models can be found in the 'Simulation' folder.
source("Simulation/simulation.R")