Vue d’ensemble
A deep learning framework with tensor computation, automatic differentiation and GPU acceleration. The repository is maintained under pytorch on GitHub. Its primary language is Python. See the [project README](https://github.com/pytorch/pytorch#readme) for the supported workflows.
Fonctionnalités clés
- A deep learning framework with tensor computation, automatic differentiation and GPU acceleration.
Prérequis, installation et démarrage rapide
Follow the official installation guide for your Python environment and compute backend. Use an isolated environment and the dependency versions specified upstream.
[Read the upstream installation and quickstart instructions](https://github.com/pytorch/pytorch#readme).
[Read the upstream installation and quickstart instructions](https://github.com/pytorch/pytorch#readme).
Utilisation
Begin with an upstream example and a small dataset; validate the results before moving to larger training or evaluation workloads.
[Usage examples and configuration reference](https://github.com/pytorch/pytorch#readme).
[Usage examples and configuration reference](https://github.com/pytorch/pytorch#readme).
Compatibilité des modèles et cas d’usage
La compatibilité des modèles n’est pas indiquée dans les métadonnées du dépôt.
Licence et notes sur les risques
GitHub has not identified a single standard SPDX license for this repository. Read the upstream license files before reuse or redistribution.
Source review 2026-09-05: GitHub search metadata and repository README. No runtime benchmark performed. License metadata: NOASSERTION.
Publication et maintenance
[View upstream releases](https://github.com/pytorch/pytorch/releases).