PyTorch is a Python package that provides two high-level features: - Tensor computation (like NumPy) with strong GPU acceleration - 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.
| Uploaded | Mon Mar 31 23:49:11 2025 |
| md5 checksum | 2b3f8d0049abc0b444a685ff85d76b8a |
| arch | x86_64 |
| build | gpu_cuda113py310h926b89d_1 |
| build_number | 1 |
| depends | _openmp_mutex, blas 1.0 mkl, cffi, cudatoolkit >=11.3.1,<11.4.0a0, cudnn >=8.2,<9.0a0, cudnn >=8.2.1,<9.0a0, cupti, future, intel-openmp >=2023.1.0,<2024.0a0, libgcc-ng >=9.3.0, libprotobuf >=3.20.3,<3.21.0a0, libstdcxx-ng >=9.3.0, magma, mkl 2023.*.*, mkl >=2023.1.0,<2024.0a0, mkl-service >=2.3.0,<3.0a0, ninja, numpy >=1.21,<2, python >=3.10,<3.11.0a0, pyyaml, typing-extensions |
| license | BSD-3-Clause |
| license_family | BSD |
| md5 | 2b3f8d0049abc0b444a685ff85d76b8a |
| name | pytorch |
| platform | linux |
| sha1 | 16f933a7b756ab7fdafead8b499bf09f80d9bc1c |
| sha256 | bd31aeb618874bb6265cb73c5dd3a63a87618dad018c7115d255f398742d3261 |
| size | 448030206 |
| subdir | linux-64 |
| timestamp | 1685781872337 |
| version | 1.13.1 |