Skip to content

nandahkrishna/GeoLifeCLEF2019

Repository files navigation

GeoLifeCLEF2019

This repository contains Jupyter Notebooks with our team's code for the GeoLifeCLEF 2019 Species Recommendation Challenge.

The original code for the image patch extractor can be found at this repository and the datasets are available at the challenge's CrowdAI page.

Requirements can be installed by entering the following command in the Terminal:

pip install -r requirements.txt

We achieved a maximum Top30 score of 0.1342 and placed 3rd overall, with our top submission ranking 6th on the leaderboard. This was an XGBoost trained on Spatial Coordinates and Environmental Variable values.

Publication

You can find our paper here (CEUR-WS Vol. 2380) in the CLEF 2019 Working Notes.