Tensor
In plain terms
A multi-dimensional array, the core data structure in deep learning.
Going deeper
A tensor generalizes scalars, vectors, and matrices to any number of dimensions, holding the data and parameters neural networks operate on. Frameworks compute over tensors.
Under the hood
GPU-accelerated tensor operations underpin training; shape mismatches are a common bug source.
Examples
Neural networks, TensorFlow/PyTorch, GPU computation, image batches.
Think of it like...
Like a stack of spreadsheets extended into as many dimensions as needed.