deb_control_files:
- control
- md5sums
- postinst
- prerm
deb_fields:
Architecture: arm64
Conflicts: python3-torch-cuda
Depends: libtorch2.1 (= 2.1.2+dfsg-4), python3 (<< 3.12), python3 (>= 3.11~), python3-astunparse,
python3-expecttest, python3-filelock, python3-fsspec, python3-hypothesis, python3-jinja2,
python3-networkx, python3-numpy (>= 1:1.25.0), python3-numpy-abi9, python3-pkg-resources,
python3-psutil, python3-requests, python3-sympy, python3-typing-extensions, python3-yaml,
python3.11:any, python3:any, libc6 (>= 2.34), libgcc-s1 (>= 3.0), libonnx1t64
(>= 1.14.1), libprotobuf32t64 (>= 3.21.12), libstdc++6 (>= 13.1), libtorch-test
(= 2.1.2+dfsg-4)
Description: |-
Tensors and Dynamic neural networks in Python (Python Interface)
PyTorch is a Python package that provides two high-level features:
.
(1) Tensor computation (like NumPy) with strong GPU acceleration
(2) Deep neural networks built on a tape-based autograd system
.
You can reuse your favorite Python packages such as NumPy, SciPy and Cython
to extend PyTorch when needed.
.
This is the CPU-only version of PyTorch (Python interface).
Homepage: https://pytorch.org/
Installed-Size: '57138'
Maintainer: Debian Deep Learning Team <debian-ai@lists.debian.org>
Package: python3-torch
Priority: optional
Provides: python3-torch-api-2.1
Recommends: libtorch-dev (= 2.1.2+dfsg-4), build-essential, ninja-build, pybind11-dev
Replaces: python3-torch-cuda
Section: python
Source: pytorch
Suggests: python3-pytest
Version: 2.1.2+dfsg-4
srcpkg_name: pytorch
srcpkg_version: 2.1.2+dfsg-4