How Do I Convert Train Dataset Frames Into 5d Tensor While Maintaining Label Of Frames Dimension?

I have used the image_dataset_from_directory() to create my train(529003 frames), validation(29388 frames) and test(28875 frames) data: train_dataset = image_dataset_from_director

Solution 1:

I found the solution I need to make a custom generator that generates 5D Tensors from video input which considers the sequence length as the 5th element of the 5D Tensor. The one I am using from Keras, image_dataset_from_directory() produces a 4D Tensor.