PyTorch vs TensorFlow

Easy

deep-learning-libraries Meta Google

What is the main philosophical difference between PyTorch and TensorFlow (prior to TF 2.x eager mode)?

Multiple Choice

PyTorch only supports CPUs; TensorFlow only supports GPUs PyTorch uses dynamic computation graphs (define-by-run), while original TensorFlow used static graphs (define-and-run) TensorFlow is written in Python; PyTorch is written in C++ with no Python API They are identical frameworks maintained by the same organization

Correct!

Incorrect — try again next time!

Upgrade to Pro to unlock solutions and explanations

PyTorch vs TensorFlow

Multiple Choice

Related problems

Deep-dive guides