TypeError: can‘t convert cuda:0 device type tensor to numpy.

Traceback (most recent call last):
  File "train.py", line 511, in <module>
    train(hyp, tb_writer, opt, device)
  File "train.py", line 368, in train
    save_dir=log_dir)
  File "../yolov5-master/test.py", line 176, in test
    plot_images(img, output_to_target(output, width, height), paths, str(f), names)  # predictions
  File "../yolov5-master/utils/utils.py", line 914, in output_to_target
    return np.array(targets)
  File "../lib/python3.7/site-packages/torch/tensor.py", line 492, in __array__
    return self.numpy()
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

Traceback (most recent call last):
  File "train.py", line 511, in <module>
    train(hyp, tb_writer, opt, device)
  File "train.py", line 368, in train
    save_dir=log_dir)
  File "../yolov5-master/test.py", line 176, in test
    plot_images(img, output_to_target(output, width, height), paths, str(f), names)  # predictions
  File "../yolov5-master/utils/utils.py", line 914, in output_to_target
    return np.array(targets)
  File "../miniconda3/envs/yolov5/lib/python3.7/site-packages/torch/tensor.py", line 492, in __array__
    return self.numpy()
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

解决老版本yolo v5 test时报上述错误：

修改/yolov5-master/utils/utils.py

第895行： output_to_target(output, width, height) 函数中的元素o，将其转到cpu上: o = o.cpu().numpy()

 output是GPU上的list ,list元素的类型是tensor，需要先转为cpu上的numpy()类型

def output_to_target(output, width, height):
    # Convert model output to target format [batch_id, class_id, x, y, w, h, conf]
    if isinstance(output, torch.Tensor):
        output = output.cpu().numpy()
    
    targets = []
    for i, o in enumerate(output):
        if o is not None:
            # output是gpu上的list ,list元素的类型是tensor，需要先转为cpu删的numpy()类型
            o = o.cpu().numpy()
            for pred in o:
                box = pred[:4]
                w = (box[2] - box[0]) / width
                h = (box[3] - box[1]) / height
                x = box[0] / width + w / 2
                y = box[1] / height + h / 2
                conf = pred[4]
                cls = int(pred[5])

                targets.append([i, cls, x, y, w, h, conf])

    return np.array(targets)