MLOps overview#
Authors: Sam Budd, Gonzalo Mateo-García
The MLOps section have tutorials for training, testing and running inference of flood extent segmentation models for Sentinel-2. Models are trained in the WorldFloods dataset which is freely accessible. Each of the tutorials is self-contained and can be run on Google Colab.
Tutorials on models of Portalés-Julià et al 2023.
Inference with clouds aware flood segmentation model: Run inference of the multioutput binary classification model. This model is able to predict land/water under the clouds.
Inference on time series of Sentinel-2 images: Download a time series of Sentinel-2 images over an area of interest and run inference on them.
Run the clouds-aware flood segmentation model in Sentinel-2 and Landsat and vectorise the flood maps
Tutorials on models of Mateo-García et al 2021.
Train models: trains the WorldFloods model on the WorldFloods V1 dataset from scratch
Model metrics: loads a worldfloods trained model and run inference on all images of the WorldFloods V1 test dataset. It computes displays some standard metrics and the PR and ROC curves for water detection.
Inference on Sentinel-2 images: loads a worldfloods pretrained model and runs inference on a Sentinel-2 image from the WorldFloods V1 dataset. It shows the predictions vs the ground truth on that image.
Exploratory work:
Probabilistic Neural Networks: Run inference of the U-Nets trained with dropout. We apply Bayesian dropout at inference time to obtain an ensemble of predictions.
Mateo-Garcia, G. et al. Towards global flood mapping onboard low cost satellites with machine learning. Scientific Reports 11, 7249 (2021). DOI: 10.1038/s41598-021-86650-z.
Portalés-Julià E., Mateo-García G., Purcell C. and Gómez-Chova L. Global flood extent segmentation in optical satellite images. Scientific Reports 13, 20316 (2023). DOI: 10.1038/s41598-023-47595-7.