No module named 'transformers_modules.sunzeyeah.pangu-2'

[MRKINKI]

tokenizer = AutoTokenizer.from_pretrained("sunzeyeah/pangu-2.6B", trust_remote_code=True) 这样加载时会报错

[sunzeyeah]

你好，由于importlib是根据.来间隔module, 如果model_name_or_path（即AutoTokenizer.from_pretrained()的第一个入参）的取值中出现.，就会报ModuleNotFoundError。现已更新huggingface hub的地址为sunzeyeah/pangu-2_6B，可使用如下命令进行加载：

tokenizer = AutoTokenizer.from_pretrained("sunzeyeah/pangu-2_6B", trust_remote_code=True)

[MRKINKI]

你好，由于importlib是根据.来间隔module, 如果model_name_or_path（即AutoTokenizer.from_pretrained()的第一个入参）的取值中出现.，就会报ModuleNotFoundError。现已更新huggingface hub的地址为sunzeyeah/pangu-2_6B，可使用如下命令进行加载：

tokenizer = AutoTokenizer.from_pretrained("sunzeyeah/pangu-2_6B", trust_remote_code=True)

谢谢！已经可以加载tokenizer和模型，但使用以下配置训练时 出现报错ValueError: Unexpected keyword arguments: hidden_states,layer_past,attention_mask,head_mask,custom_query

training_args = TrainingArguments(
    save_total_limit=1,
    output_dir=output_dir,
    evaluation_strategy="steps",
    eval_accumulation_steps=1,
    learning_rate=learning_rate,
    per_device_train_batch_size=train_batch_size,
    per_device_eval_batch_size=eval_batch_size,
    gradient_checkpointing=True,
    half_precision_backend=True,
    fp16=True,
    adam_beta1=0.9,
    adam_beta2=0.95,
    gradient_accumulation_steps=gradient_accumulation_steps,
    num_train_epochs=num_train_epochs,
    warmup_steps=100,
    eval_steps=eval_steps,
    save_steps=save_steps,
    load_best_model_at_end=True,
    logging_steps=50,
    deepspeed="./config.json",
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
    eval_dataset=dev_dataset,
    compute_metrics=compute_metrics,
    data_collator=default_data_collator,
    preprocess_logits_for_metrics=preprocess_logits_for_metrics,
)

[xiaoranchenwai]

你好，由于importlib是根据.来间隔module, 如果model_name_or_path（即AutoTokenizer.from_pretrained()的第一个入参）的取值中出现.，就会报ModuleNotFoundError。现已更新huggingface hub的地址为sunzeyeah/pangu-2_6B，可使用如下命令进行加载：

tokenizer = AutoTokenizer.from_pretrained("sunzeyeah/pangu-2_6B", trust_remote_code=True)

谢谢！已经可以加载tokenizer和模型，但使用以下配置训练时 出现报错ValueError: Unexpected keyword arguments: hidden_states,layer_past,attention_mask,head_mask,custom_query

training_args = TrainingArguments(
    save_total_limit=1,
    output_dir=output_dir,
    evaluation_strategy="steps",
    eval_accumulation_steps=1,
    learning_rate=learning_rate,
    per_device_train_batch_size=train_batch_size,
    per_device_eval_batch_size=eval_batch_size,
    gradient_checkpointing=True,
    half_precision_backend=True,
    fp16=True,
    adam_beta1=0.9,
    adam_beta2=0.95,
    gradient_accumulation_steps=gradient_accumulation_steps,
    num_train_epochs=num_train_epochs,
    warmup_steps=100,
    eval_steps=eval_steps,
    save_steps=save_steps,
    load_best_model_at_end=True,
    logging_steps=50,
    deepspeed="./config.json",
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
    eval_dataset=dev_dataset,
    compute_metrics=compute_metrics,
    data_collator=default_data_collator,
    preprocess_logits_for_metrics=preprocess_logits_for_metrics,
)

你好，这个问题解决了吗，我也遇到了

[sunzeyeah]

你好，能否提供完整的报错日志以及启动脚本？