neural_compressor.torch.quantization.algorithm_entry
Module Contents
Functions
|
The main entry to apply rtn quantization. |
- neural_compressor.torch.quantization.algorithm_entry.rtn_entry(model: torch.nn.Module, configs_mapping: Dict[Tuple[str, callable], neural_compressor.torch.quantization.RTNConfig], mode: neural_compressor.torch.utils.Mode = Mode.QUANTIZE, *args, **kwargs) torch.nn.Module [source]
The main entry to apply rtn quantization.