:orphan:

:py:mod:`neural_compressor.adaptor.query`
=========================================

.. py:module:: neural_compressor.adaptor.query


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

Classes
~~~~~~~

.. autoapisummary::

   neural_compressor.adaptor.query.QueryBackendCapability




.. py:class:: QueryBackendCapability

   Base class that defines Query Interface.
   Each adaption layer should implement the inherited class for specific backend on their own.

   .. py:method:: get_version()
      :abstractmethod:

      Get the current backend's version string.



   .. py:method:: get_precisions()
      :abstractmethod:

      Get the supported low precisions, e.g ['int8', 'bf16']



   .. py:method:: get_op_types()
      :abstractmethod:

      Get the op types for specific backend per low precision.
      e.g {'2.3.0': {'int8': ['Conv2D', 'MatMuL']}}


   .. py:method:: get_fuse_patterns()
      :abstractmethod:

      Get the fusion patterns for specified op type for every specific precision




   .. py:method:: set_quantization_config(q_config)
      :abstractmethod:

      Set the quantization config to backend.
      :param q_config: set the organized quantization configuration to backend.
      :type q_config: yaml content?


   .. py:method:: get_quantization_capability()
      :abstractmethod:

      Get the quantization capability of low precision op types.
      e.g, granularity, scheme and etc.



   .. py:method:: get_mixed_precision_combination(unsupported_precisions)
      :abstractmethod:

      Get the valid precision combination base on hardware and user' config.
      e.g['fp32', 'bf16', 'int8']