Name: DataLearning
Type: Organization
Bio: DL-WG is developed by the Data Learning group, an interdisciplinary group developing pioneering research on fundamental Data Science and Machine Learning
Location: Imperial College London
Blog: https://www.imperial.ac.uk/earth-science/research/research-groups/datalearning/
DataLearning's Projects
Using adversarial training to improve forecasts of data-driven surrogate models
introduction to DA
MSc Research project (6 months). Data Assimilation using Deep Learning (AEs). Imperial College Machine Learning MSc 2018-19
Code repository for the data learning paper
Fire-Image-DenseNet - A model for predicting the final burnt area of wildfire
In this project, we introduce the Generalised Latent Assimilation (GLA) approach which can be used in Machine learning surrogate modelling for dynamical systems
Python and bash script for training on HPC
Code base for Latent-GAN training algorithm.
This repository includes the Reduced Order Model code and the Variational Data Assimilation code developed for the EPSRC MAGIC (EP/N010221/1) project
ML surrogate model which combines ROM, LSTM and DA for dynamical systems such as wildfires