Metadata-Version: 2.1
Name: cutensor-cu12
Version: 0.0.1.dev0
Summary: NVIDIA cuTENSOR
Home-page: https://developer.nvidia.com/cutensor
Author: NVIDIA Corporation
Author-email: cuda_installer@nvidia.com
License: NVIDIA Proprietary Software
Keywords: cuda,nvidia,machine learning,tensor network,high-performance computing
Platform: UNKNOWN
Classifier: Topic :: Scientific/Engineering
Classifier: Environment :: GPU :: NVIDIA CUDA
Description-Content-Type: text/x-rst

###############################################################
cuTENSOR: A High-Performance CUDA Library For Tensor Primitives
###############################################################

`cuTENSOR <https://developer.nvidia.com/cutensor>`_ is a high-performance CUDA library for tensor primitives.

Key Features
============

* Extensive mixed-precision support:

  * FP64 inputs with FP32 compute.
  * FP32 inputs with FP16, BF16, or TF32 compute.
  * Complex-times-real operations.
  * Conjugate (without transpose) support.

* Support for up to 64-dimensional tensors.
* Arbitrary data layouts.
* Trivially serializable data structures.
* Main computational routines:

  * Direct (i.e., transpose-free) tensor contractions.
  * Tensor reductions (including partial reductions).
  * Element-wise tensor operations:

    * Support for various activation functions.
    * Arbitrary tensor permutations.
    * Conversion between different data types.

Documentation
=============

Please refer to https://docs.nvidia.com/cuda/cutensor/index.html for the cuTENSOR documentation.

Installation
============

The cuTENSOR wheel can be installed as follows:

.. code-block:: bash

   pip install cutensor-cuXX

where XX is the CUDA major version (currently CUDA 11 is supported).
The package ``cutensor`` (without the ``-cuXX`` suffix) is considered deprecated.


