This document provides technical information for migration from Chainer to PyTorch. General InformationConcepts and components in both frameworksArray LibraryChainer uses NumPy/CuPy (xp.ndarray) as an array library, and wraps them as chainer.Variable to support autograd. Similarly, PyTorch uses ATen (at::Tensor (C++)) as an array library ("tensor library" in PyTorch terms), and wraps it as torch::