perun.backend.nvml
==================

.. py:module:: perun.backend.nvml

.. autoapi-nested-parse::

   Nvidia Mangement Library Source definition.



Attributes
----------

.. autoapisummary::

   perun.backend.nvml.log


Classes
-------

.. autoapisummary::

   perun.backend.nvml.NVMLBackend


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

.. py:data:: log

.. py:class:: NVMLBackend

   Bases: :py:obj:`perun.backend.backend.Backend`


   NVMLSource class.

   Setups connection to NVML and creates relevant devices


   .. py:attribute:: id
      :value: 'nvlm'



   .. py:attribute:: name
      :value: 'NVIDIA ML'



   .. py:attribute:: description
      :type:  str
      :value: 'Access GPU information from NVML python bindings'



   .. py:method:: setup()

      Init pynvml and gather number of devices.



   .. py:method:: close()

      Backend shutdown code.



   .. py:method:: availableSensors() -> Dict[str, Tuple]

      Return string ids of visible devices.

      Returns
      -------
      Set[str]
          Set with sensor ids.



   .. py:method:: getSensors(deviceList: Set[str]) -> List[perun.data_model.sensor.Sensor]

      Gather sensor object based on a set of device ids.

      Parameters
      ----------
      deviceList : Set[str]
          Set containing divice ids.

      Returns
      -------
      List[Sensor]
          List with Sensor objects.



   .. py:method:: _getPowerSensor(device_idx: int) -> perun.data_model.sensor.Sensor


   .. py:method:: _getPowerCallback(handle) -> Callable[[], numpy.number]


   .. py:method:: _getMemorySensor(device_idx: int) -> perun.data_model.sensor.Sensor


   .. py:method:: _getUsedMemCallback(handle) -> Callable[[], numpy.number]


   .. py:method:: _getClockSensor(device_idx: int, clock_type: str) -> perun.data_model.sensor.Sensor


   .. py:method:: _getClockCallback(handle, clock_type) -> Callable[[], numpy.number]