Hugging Face Trainer：模型初始化错误答案

【问题标题】：Hugging Face Trainer: Error in the model initHugging Face Trainer：模型初始化错误
【发布时间】：2021-06-23 23:26:13
【问题描述】：

我收到以下错误：

You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
Traceback (most recent call last):
  File "./run_hyperparameter_search.py", line 74, in <module>
    trainer = Trainer(
  File "/ext3/miniconda3/lib/python3.8/site-packages/transformers/trainer.py", line 273, in __init__
    model = self.call_model_init()
  File "/ext3/miniconda3/lib/python3.8/site-packages/transformers/trainer.py", line 737, in call_model_init
    raise RuntimeError("model_init should have 0 or 1 argument.")
RuntimeError: model_init should have 0 or 1 argument.
~

这是我在拥抱面部训练器中所说的：

#Initialising the model
trainer = Trainer(
    args = training_args,
    tokenizer = tokenizer,
    train_dataset = train_data,
    eval_dataset = val_data,
    # maybe there is a () in the init, but not in compute metrics for sure. Will test
    model_init = finetuning_utils.model_init(),
    compute_metrics = finetuning_utils.compute_metrics,
)

问题显然出在 model_init 中。

finetuning_utils.model_init() 包含以下内容：

def model_init():
    """Returns an initialized model for use in a Hugging Face Trainer."""
    ## TODO: Return a pretrained RoBERTa model for sequence classification.
    ## See https://huggingface.co/transformers/model_doc/roberta.html#robertaforsequenceclassification.
    model = RobertaForSequenceClassification.from_pretrained("roberta-base")
    #model = model.to('cuda')
    return model

请帮助解决错误。

【问题讨论】：

来自huggingface.co/transformers/_modules/transformers/trainer.html 看起来 model_init 需要一个可调用对象。您是否尝试过删除括号，即 finetuning_utils.model_init ？
或者也许使用model = RobertaForSequenceClassification.from_pretrained("roberta-base")而不是model_init
@ML_Engine 谢谢！
我已将其添加为答案，如果有帮助将不胜感激：-)

标签： nlp huggingface-transformers

【解决方案1】：

从Huggingface trainer docs 看来，model_init 需要一个可调用对象。因此，与其实例化参数，不如采用可调用对象本身，即不带括号：

model_init = finetuning_utils.model_init

或者，您可以删除model_init 并使用model 参数，其效果与finetuning_utils.model_init 中包含的代码相同，如下所示：

model = RobertaForSequenceClassification.from_pretrained("roberta-base")

【讨论】：