Пакет: libtorch-cuda-dev (2.6.0+dfsg-6) [contrib]

[ Източник: pytorch-cuda  ]

Tensors and Dynamic neural networks in Python (Development Files)

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 CUDA version of PyTorch (Development files).

Други пакети, свързани с libtorch-cuda-dev

• dep: libprotobuf-dev

  protocol buffers C++ library (development files) and proto files

• dep: libtorch-cuda-2.6 (= 2.6.0+dfsg-6)

  Tensors and Dynamic neural networks in Python (Shared Objects)

• dep: python3-dev

  header files and a static library for Python (default)

Изтегляне на libtorch-cuda-dev

Изтегляне за всички налични архитектури

Архитектура	Големина на пакета	Големина след инсталиране	Файлове
amd64	2 368,5 кБ	30 273,0 кБ	[списък на файловете]
arm64	2 367,8 кБ	30 269,0 кБ	[списък на файловете]
ppc64el	2 368,0 кБ	30 268,0 кБ	[списък на файловете]