PyTorch is consistently adding a boost to the field of Computer Vision and Deep Learning by providing a number of powerful tools and techniques. In the field of computer vision where the deep learning based executions are to be dealt with heavy image datasets, an accelerated environment is needed to fasten the execution process with an acceptable accuracy level. PyTorch provides this feature through the XLA (Accelerated Linear Algebra), a compiler for linear algebra that can target multiple types of hardware, including GPU, and TPU. The PyTorch/XLA environment is integrated with the Google Cloud TPU and an accelerated speed of execution is achieved.

Read more: https://analyticsindiamag.com/hands-on-guide-to-implement-resnet50-in-pytorch-with-tpu/

#pytorch #tpu