Think of a tensor as a multi-dimensional array. In TensorFlow, all data is represented as tensors, which are the primary data structures that are used to represent and manipulate data in TensorFlow.

What is next for the Google Tensor processor? Will we see a competitor to the likes of the Snapdragon line? We examine the possibilities.

We look at Google’s TPUs – tensor processing units – and ask what makes them different to CPUs, GPUs and DPUs, as well as how you can take advantage of them in AI processing.