I can't output the object detection test. No labels or bounding boxes, just images

1 
pread_list = []
model.eval()
with torch.no_grad():

    for i in test_data:
        image = Image.open(i)
        image = transform(image)
        image = image.unsqueeze(0)
        image = image.to(device)
        output = model(image)  
        preds = torch.max(output, 600, 600)#, dim=1)[:, 1].tolist() 
        #id_list.append(_id)
        #pred_list.append(preds[0])

TypeError Traceback (most recent call last)
Input In [12], in <cell line: 3>()
9 image = image.to(device)
10 output = model(image)
—> 11 preds = torch.max(output, 600, 600)

TypeError: max() received an invalid combination of arguments - got (list, int, int), but expected one of:

  • (Tensor input)
  • (Tensor input, Tensor other, *, Tensor out)
  • (Tensor input, int dim, bool keepdim, *, tuple of Tensors out)
  • (Tensor input, name dim, bool keepdim, *, tuple of Tensors out)
2

torch.max expects the input, dim, and keepdim input arguments as explained in the docs while you are passing two integers to this call which seem to be a spatial size of the tensor.

1 Like