我这也是,运行nlp模块的时候不会有问题,但运行多模态时也会报这个错,报错的运行代码如下:
import torch
import cv2
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
task = Tasks.text_to_image_synthesis
model_id = 'damo/multi-modal_chinese_stable_diffusion_v1.0'
# 基础调用
pipe = pipeline(task=task, model=model_id)
output = pipe({'text': '中国山水画'})
cv2.imwrite('result.png', output['output_imgs'][0])