ModelScope一执行就报内存溢出咋整？报错如下RuntimeError: Distribute

以下是 1.3B 模型的 pipeline 运行示例代码，您试一下是否会出现 oom 的问题？ from modelscope.pipelines import pipeline from modelscope.utils.constant import Tasks import torch

if name == 'main': if torch.multiprocessing.get_start_method(allow_none=True) is None: torch.multiprocessing.set_start_method('spawn') input = '程序员脱发用什么洗发水' model_id = 'damo/nlp_gpt3_text-generation_1.3B' pipe = pipeline(Tasks.text_generation, model=model_id) pipe.models = []

print(pipe(input))——此答案整理自钉群“魔搭ModelScope开发者联盟群 ①”