:py:mod:`neural_compressor.strategy.utils.utility`
==================================================

.. py:module:: neural_compressor.strategy.utils.utility

.. autoapi-nested-parse::

   Tuning utility.



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

Classes
~~~~~~~

.. autoapisummary::

   neural_compressor.strategy.utils.utility.QuantType
   neural_compressor.strategy.utils.utility.QuantOptions
   neural_compressor.strategy.utils.utility.OrderedDefaultDict
   neural_compressor.strategy.utils.utility.ClassRegister



Functions
~~~~~~~~~

.. autoapisummary::

   neural_compressor.strategy.utils.utility.preprocess_user_cfg
   neural_compressor.strategy.utils.utility.extract_data_type
   neural_compressor.strategy.utils.utility.reverted_data_type
   neural_compressor.strategy.utils.utility.get_adaptor_name
   neural_compressor.strategy.utils.utility.build_slave_faker_model



.. py:class:: QuantType




   Quantization type.


.. py:class:: QuantOptions(quant_type=1)


   Option Class for Quantization.

   This class is used for configuring global variable related to quantization.
   The global variable quant_options is created with this class.

   :param quant_type: Quantization type. Default value is 1.
   :type quant_type: int


.. py:function:: preprocess_user_cfg(op_user_cfg: Dict)

   Preprocess the op user config for weight only.

   :param op_user_cfg: The original user config.

   .. rubric:: Example

   op_user_cfg = {'activation': {'bits': [4]}}
   op_user_cfg_modified = {'activation': {'bits': [4], 'group_size': [32]}}

   :returns: The modified config.


.. py:class:: OrderedDefaultDict




   Ordered default dict.


.. py:function:: extract_data_type(data_type: str) -> str

   Extract data type and signed from data type.

   :param data_type: The original data type such as uint8, int8.

   :returns: (signed or unsigned, data type without signed)


.. py:function:: reverted_data_type(signed_flag: str, data_type: str) -> str

   Revert the data type.


.. py:function:: get_adaptor_name(adaptor)

   Get adaptor name.

   :param adaptor: adaptor instance.


.. py:function:: build_slave_faker_model()

   Slave does not have a model, so construct a fake model.

   :returns: a class object where all properties and methods are virtual.
   :rtype: object


.. py:class:: ClassRegister


   Class register.