:py:mod:`neural_compressor.strategy.hawq_v2` ============================================ .. py:module:: neural_compressor.strategy.hawq_v2 .. autoapi-nested-parse:: The HAWQ_V2 tuning strategy. Module Contents --------------- Classes ~~~~~~~ .. autoapisummary:: neural_compressor.strategy.hawq_v2.HAWQ_V2TuneStrategy .. py:class:: HAWQ_V2TuneStrategy(model, conf, q_dataloader=None, q_func=None, eval_dataloader=None, eval_func=None, resume=None, q_hooks=None) Bases: :py:obj:`neural_compressor.strategy.strategy.TuneStrategy` The HAWQ V2 tuning strategy. HAWQ_V2 implements the "Hawq-v2: Hessian aware trace-weighted quantization of neural networks". We made a small change to it by using the hessian trace to score the op impact and then fallback the OPs according to the scoring result. .. py:method:: next_tune_cfg() Generate and yield the next tuning config using HAWQ v2 search in tuning space. :Yields: *tune_config (dict)* -- A dict containing the tuning configuration for quantization.