bank777777的博客

多GPU训练的模型，转换加载到单GPU测试或预训练报错：RuntimeError: Attempting to deserialize object on CUDA device 3 but torch.cuda.device_count() is 1.加载的模型是3个GPU进行训练的，而你的计算机只有1个GPU state_dict = model.state_dict() weights_path = torch.load(opt.weights_path)['model_sta