Train a computer vision model for analysing marine biodiversity using SUBSIM on EDITO

This tutorial guides you through the process of training and testing an object detection model using the notebooks of the Swedish platform for subsea image analysis (www.subsim.se) deployed on the Digital Twin Ocean infrastructure EDITO. The video will show you how to analyse a small data set containing footage of invasive fish collected on the Swedish west coast. It will introduce you to the structure of input data, explain how to train a model, inspect the results and compare these against a stronger pre-trained model. IMPORTANT: This workflow is a demo workflow for users to learn using computer vision methods in EDITO, while the original SUBSIM workflows (available on the SUBSIM GitHub pages) are more elaborate. If you have any questions and want to use more elaborate computer vision workflows for analysing marine biodiversity, please contact us through the SBDI support page at https://biodiversitydata.se.