/A-physically-constrained-hybrid-deep-learning-model

Primary LanguageMATLAB

System-specific notes

Matlab R2016a, Tensorflow1.14 version, python3.7

How to use it?

Firstly, run 1DCNN_DataPreparation.m and miniGCN_DataPreparation.m to generate data
1DCNN_DataPreparation:Generate training, validation, and testing datastes for the one-dimensional convolutional neural network
miniGCN_DataPreparation:Generate training, validation, and testing datastes for the one-dimensional mini-batch graph convolutional network
Then run 1DCNN+miniGCN_loss.py

REFERENCE

CNN+miniGCN

Hong, D., Gao, L., Yao, J., Zhang, B., Plaza, A., Chanussot, J., 2021. Graph Convolutional Networks for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing 59(7), 5966-5978. If you have any question, you can contact me via email xy1021989464@163.com