neural_compressor.strategy.basic

The basic tuning strategy.

Module Contents

Classes

BasicTuneStrategy

The basic tuning strategy.
class neural_compressor.strategy.basic.BasicTuneStrategy(model, conf, q_dataloader=None, q_func=None, eval_dataloader=None, eval_func=None, resume=None, q_hooks=None)

Bases: neural_compressor.strategy.strategy.TuneStrategy

The basic tuning strategy.

next_tune_cfg()

Generate and yield the next tuning config with below order.

  1. OP Type Wise Tuning

  2. Fallback OP One by One

  3. Fallback Multiple OPs Accumulated

Yields:

tune_config (dict) – A dict containing the tuning configuration for quantization.