afe.core.calibrate_networks

Functions

calibrate_network(β†’Β None)

Save quantization configuration in an AwesomeNet, then calibrate it.

Module Contents

afe.core.calibrate_networks.calibrate_network(net: afe.ir.net.AwesomeNet, opt_config: afe.core.configs.OptimizationConfigs, input_generator: sima_utils.data.data_generator.DataGenerator | Iterable[Dict[afe.ir.defines.NodeName, numpy.ndarray]]) → None[source]

Save quantization configuration in an AwesomeNet, then calibrate it. This function is used in legacy tests. Do not call this function.

Parameters:

  • net – an AwesomeNet

  • opt_config – A OptimizationConfigs instance containing quantization scheme information.

  • input_generator – Source of input values for calibration.