:py:mod:`neural_compressor.experimental.nas` ============================================ .. py:module:: neural_compressor.experimental.nas .. autoapi-nested-parse:: NAS module. Subpackages ----------- .. toctree:: :titlesonly: :maxdepth: 3 dynast/index.rst Submodules ---------- .. toctree:: :titlesonly: :maxdepth: 1 basic_nas/index.rst dynas/index.rst nas/index.rst nas_utils/index.rst search_algorithms/index.rst Package Contents ---------------- Classes ~~~~~~~ .. autoapisummary:: neural_compressor.experimental.nas.BasicNAS neural_compressor.experimental.nas.DyNAS neural_compressor.experimental.nas.NAS .. py:class:: BasicNAS(conf_fname_or_obj, search_space=None, model_builder=None) Bases: :py:obj:`neural_compressor.experimental.nas.nas.NASBase`, :py:obj:`neural_compressor.experimental.component.Component` Basic NAS approach. Defining the pipeline for basic NAS approach. :param conf_fname: The path to the YAML configuration file. :type conf_fname: string :param search_space: A dictionary for defining the search space. :type search_space: dict :param model_builder: A function to build model instance with the specified model architecture parameters. :type model_builder: function obj .. py:method:: execute() Execute the search process. :returns: Best model architectures found in the search process. .. py:method:: estimate(model) Estimate performance of the model. Depends on specific NAS algorithm. Here we use train and evaluate. :returns: Evaluated metrics of the model. .. py:method:: init_by_cfg(conf_fname_or_obj) Initialize the configuration. .. py:method:: pre_process() Initialize the train and evaluation settings. .. py:class:: DyNAS(conf_fname_or_obj) Bases: :py:obj:`neural_compressor.experimental.nas.nas.NASBase` DyNAS approach. Defining the pipeline for DyNAS approach. :param conf_fname_or_obj: The path to the YAML configuration file or the object of NASConfig. :type conf_fname_or_obj: string or obj .. py:method:: estimate(individual) Estimate performance of the model. :returns: Evaluated metrics of the model. .. py:method:: init_for_search() Initialize the search configuration. .. py:method:: search() Execute the search process. :returns: Best model architectures found in the search process. .. py:method:: select_model_arch() Select the model architecture. .. py:method:: create_acc_predictor() Create the accuracy predictor. .. py:method:: create_macs_predictor() Create the MACs predictor. .. py:method:: create_latency_predictor() Create the latency predictor. .. py:method:: init_cfg(conf_fname_or_obj) Initialize the configuration. .. py:class:: NAS Bases: :py:obj:`object` Create object of different NAS approaches. :param conf_fname_or_obj: The path to the YAML configuration file or the object of NASConfig. :type conf_fname_or_obj: string or obj :returns: An object of specified NAS approach.