Experiment examplesο
The repository contains examples of how to use Malpolon for different scenarios. The examples are organized by the type of user and the type of dataset used.
I. Ecologists scenarioο
I have a dataset of pre-extracted image patches or raster files (e.g. bioclimatic data, satellite rastersβ¦) of my own and I want to train a deep image neural network model on it. I want to be able to easily customize the training process and the model architecture.
Drop and play : I have a file (.csv) of Presence/Absence (PA) or Presence Only (PO) observations and I want to train a model on different environmental variables (rasters, satellite imagery) without having to worry about the data loading and on-the-fly extraction.
Custom dataset : I have my own dataset consisting of pre-extracted image patches and/or rasters and I want to train a model on it.
Sentinel-2Aο
MicroGeoLifeCLEF2022ο
II. Inference scenarioο
I have a file (.csv) of Presence/Absence (PA) or Presence Only (PO) observations and I want to predict the presence of species on a given area using a model I trained previously and a selected dataset I would provide.
Sentinel-2Aο
MicroGeoLifeCLEF2022ο
III. Kaggle scenarioο
I am a potential kaggle participant on the GeoLifeClef challenge. I want to train a model on the provided datasets without having to worry about the data loading, starting from a plug-and-play example.
GeoLifeClef2022 : contains a fully functionnal example of a model training on the GeoLifeClef2022 dataset, from data download, to training and prediction.
GeoLifeClef2023 : contains dataloaders for the GeoLifeClef2023 dataset (different from the GLC2022 dataloaders). The training and prediction scripts are not provided.