# 报错处理 ‘torch.dtype‘ object has no attribute ‘element_size‘

在导入大模型时，有时会报错：'torch.dtype' object has no attribute 'element_size'。

可以修改报错文件 /miniconda3/lib/python3.8/site-packages/transformers/modeling_utils.py中的信息，只需注释5行代码，并新增9行代码即可顺利运行，修改如下：

        for param in total_parameters:
            if param.requires_grad or not only_trainable:
                # For 4bit models, we need to multiply the number of parameters by 2 as half of the parameters are
                # used for the 4bit quantization (uint8 tensors are stored)
                if is_loaded_in_4bit and isinstance(param, bnb.nn.Params4bit):
    #                quant_storage = self.hf_quantizer.quantization_config.bnb_4bit_quant_storage
    #                # For compatibility with older PT version - see: https://github.com/huggingface/peft/pull/1635
    #                nb_params = (
    #                    quant_storage.itemsize if hasattr(quant_storage, "itemsize") else quant_storage.element_size()
    #                )
    #               total_numel.append(param.numel() * 2 * nb_params)
                    if hasattr(param, "element_size"):
                        num_bytes = param.element_size()
                    elif not hasattr(param, "quant_storage"):
                        num_bytes = 1
                    else:
                        num_bytes = param.quant_storage.itemsize
                    total_numel.append(
                        param.numel() * 2 * num_bytes
                    )
                else:
                    total_numel.append(param.numel())

参考资料：

fix for itemsize => element_size() for torch backwards compat by winglian · Pull Request #30133 · huggingface/transformers · GitHub