Adapter Transformers 模型训练

构建一个带adapter结构的模型

from transformers import BertForSequenceClassification
from transformers import AdapterType, MultiLingAdapterArguments, HfArgumentParser, AdapterConfig
model = BertForSequenceClassification.from_pretrained(model_name, num_labels=n_label)
parser = HfArgumentParser((MultiLingAdapterArguments))
adapter_args = parser.parse_args_into_dataclasses()[0]
adapter_config = AdapterConfig.load(
adapter_args.adapter_config,
non_linearity=adapter_args.adapter_non_linearity,
reduction_factor=adapter_args.adapter_reduction_factor,
)
model.add_adapter('test', AdapterType.text_task, config=adapter_config)
print(model.config.adapters.adapter_list(AdapterType.text_task))
model.set_active_adapters(['test'])

最后

以上就是苗条泥猴桃为你收集整理的Adapter Transformers 模型训练的全部内容，希望文章能够帮你解决Adapter Transformers 模型训练所遇到的程序开发问题。

如果觉得靠谱客网站的内容还不错，欢迎将靠谱客网站推荐给程序员好友。