Combined dataset with structurred data related

hi everyone,

iam thinking how to solve the training of a nn, when the dataset contains unstructured data related, like video, audio, an other unstructured data.
What architecture would i have to use?. since the begining data in training data set is from same kind, pictures, audio, but nothing combined.

Hi mako,

You may want to look into multimodel neural networks. This is an advanced topic that goes beyond this specialization. To give you a taste, you can for instance have a look here or here.

thanks i will take a look.