I'm dealing with an image classification problem in which I have different types of labels for every observation. I'm trying to do multitask learning using R keras. The only problem is that I have to use the keras 'Data Loader' because the dataset is huge and I don't know how to use the flow_images_from_directory function for multi-output models. Please someone help me.
Asked
Active
Viewed 24 times
0
-
What have you tried? Read [this](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) and edit your answer so other people can help you. – John Polo Feb 12 '22 at 21:12
-
I've tried to save all the information in a tensor, but I cannot store all the images, so I have to work with Data Loader, but I don't know if I can do it in a multitask scenario. – Federico Feb 13 '22 at 12:36