I am working in a binary classification problem and I have two inputs to the network (a dataset and images).
In the first branch I use a Multi-layer Perceptron (MLP) to handle the dataset and in the second branch I use a CNN to process all the image data. After that, I concatenate the outputs of both branches.
The labels are unbalanced and I have been wondering how I can balance the data. I know that I could use some techniques like SMOTE for the dataset and dataaugmentation for the images if the network had only one input, but I don't know if there is some other technique that can handle multi-input data or I must apply this techniques separately and then put them together.
Thank you in advance.