TensorFlow is called ‘TensorFlow’ because it handles the flow (node/mathematical operation) of Tensors, which are data structures that you can think of as multi-dimensional arrays. Tensors are represented as n-dimensional arrays of base dataypes such as a string or integer — they provide a way to generalize vectors and matrices to higher dimensions.
shape of a Tensor defines its number of dimensions and the size of each dimension. The
rank of a Tensor provides the number of dimensions (n-dimensions) — you can also think of this as the Tensor’s order or degree.
If you run into something that seems like it couldn’t possibly happen:
1) 90% of the time, it’s not actually happening; the observation is wrong. Perhaps a caching or sync issue. Perhaps a measurement error.
2) 10% of the time, your entire mental model of the system is wrong.