neural_compressor.torch.utils.environ
=====================================

.. py:module:: neural_compressor.torch.utils.environ

.. autoapi-nested-parse::

   Intel Neural Compressor PyTorch environment check.



Functions
---------

.. autoapisummary::

   neural_compressor.torch.utils.environ.is_ipex_imported
   neural_compressor.torch.utils.environ.is_transformers_imported
   neural_compressor.torch.utils.environ.is_package_available
   neural_compressor.torch.utils.environ.is_hpex_available
   neural_compressor.torch.utils.environ.is_ipex_available
   neural_compressor.torch.utils.environ.get_ipex_version
   neural_compressor.torch.utils.environ.get_torch_version
   neural_compressor.torch.utils.environ.get_accelerator
   neural_compressor.torch.utils.environ.device_synchronize


Module Contents
---------------

.. py:function:: is_ipex_imported() -> bool

   Check whether intel_extension_for_pytorch is imported.


.. py:function:: is_transformers_imported() -> bool

   Check whether transformers is imported.


.. py:function:: is_package_available(package_name)

   Check if the package exists in the environment without importing.

   :param package_name: package name
   :type package_name: str


.. py:function:: is_hpex_available()

   Returns whether hpex is available.


.. py:function:: is_ipex_available()

   Return whether ipex is available.


.. py:function:: get_ipex_version()

   Return ipex version if ipex exists.


.. py:function:: get_torch_version()

   Return torch version if ipex exists.


.. py:function:: get_accelerator(device_name='auto')

   Return the recommended accelerator based on device priority.


.. py:function:: device_synchronize(raw_func)

   Function decorator that calls accelerated.synchronize before and after a function call.