This is a tensorflow and keras based implementation of SSRNs in the IEEE T-GRS paper "Spectral-Spatial Residual Network for Hyperspectral Image Classification: A 3-D Deep Learning Framework".
Thank you for your work. How do you split the training data set, validation data set and test data set? First, if the number of categories in a certain category is small, is there a chance that you will not be able to select it, and you will not be able to learn its features by SSRN? second,does the random selection destroy the distribution of the data?
Thank you for your work. How do you split the training data set, validation data set and test data set? First, if the number of categories in a certain category is small, is there a chance that you will not be able to select it, and you will not be able to learn its features by SSRN? second,does the random selection destroy the distribution of the data?