Too Long; Didn't Read
The Tensor Processing Unit (TPU) is an accelerator — custom-made by Google Brain hardware engineers — that specialises in training deep and computationally expensive ML models. After this, you’ll never want to touch your clunky CPU ever again. This article shows you how easy it is to train any TensorFlow model on a TPU with very few changes to your code. The TPU is a distributed processor that is not a single-core processor and is not like a traditional-core training processor.