TOOLORA / Local reading edition
PyTorch
Build, train and run deep-learning models with tensor operations, automatic differentiation and neural-network modules.
Overview
PyTorch is an open-source framework for deep-learning research and model development. Developers represent data as tensors, compute gradients through automatic differentiation and build training loops with neural-network modules, optimizers and data loaders. Supported CPUs and accelerators run models, with compilation and distributed capabilities extending training. Models, data, weights and a compatible runtime must be supplied separately.
- Company country / region
- United States · Company headquarters
Official sources and references
- Official websitehttps://pytorch.org/
- Documentationhttps://docs.pytorch.org/docs/stable/index.html
Descriptions summarize public information; consult official websites for features, pricing and regional availability.