in my application, I continuously simulate the data that are fed into my NN. I would like to implement this using
IterableDataset. All examples I found online assume all the training data to be already present. Can you give me a short example on how to continuously generate new batches of data. Imagine one training example consists of 10 random normal numbers (
torch.randn(10)), so that a final batch should have shape (batch_size, 10).
Thanks in advance for your help!