A tensor processing unit (TPU) is a proprietary processor designed by Google in 2016 for use in neural networks inference. It has been deployed to all Google data centers and powers applications such as Google Search, Street View, Google Photos and Google Translate.
Read this article to understand the intuition behind Tensor Processing Unit (TPU)
Norm Jouppi was the Technical leader of the TPU project.
In short, TPU has demonstrated significant performance gain in the domain of inference and is used along with CPUs and GPUs.
Have a doubt or thought? Join the discussion now
This is a companion discussion topic for the original entry at http://iq.opengenus.org/tensor-processing-unit-tpu-explained/