This is a work in progress
My implementation of the paper "M3Fusion: A Deep Learning Architecture for Multiscale Multimodal Multitemporal Satellite Data Fusion" by Benedetti, Paola, Dino Ienco, Raffaele Gaetano, Kenji Ose, Ruggero G. Pensa, and Stephane Dupuy
The code is in Tensorflow 2.+
You'll need to write a custom data generator to read the spatial and temporal data as well as specify the hyperparameters.