convert — PyTorch 2.7 documentation (original) (raw)

class torch.ao.quantization.convert(module, mapping=None, inplace=False, remove_qconfig=True, is_reference=False, convert_custom_config_dict=None, use_precomputed_fake_quant=False)[source][source]¶

Converts submodules in input module to a different module according to mappingby calling from_float method on the target module class. And remove qconfig at the end if remove_qconfig is set to True.

Parameters

• module – prepared and calibrated module
• mapping – a dictionary that maps from source module type to target module type, can be overwritten to allow swapping user defined Modules
• inplace – carry out model transformations in-place, the original module is mutated
• convert_custom_config_dict – custom configuration dictionary for convert function
• use_precomputed_fake_quant – a flag to enable use of precomputed fake quant

Example of convert_custom_config_dict:

convert_custom_config_dict = { # user will manually define the corresponding quantized # module class which has a from_observed class method that converts # observed custom module to quantized custom module "observed_to_quantized_custom_module_class": { ObservedCustomModule: QuantizedCustomModule } }