Error while printing a random sample from a dataLoader object containing videos

I created a data loader of the ucf101 video dataset

dataset = torchvision.datasets.UCF101('/content/drive/My Drive/Colab Notebooks/ucf101', annotation_path='/content/drive/My Drive/Colab Notebooks/ucfTrainTestlist', frames_per_clip=16, step_between_clips=1, frame_rate=None, fold=1, train=True, transform=transforms.Compose([transforms.ToTensor(),]))
data =, batch_size=512, shuffle=True)

I want to see the shape and a full tensor of a random sample of this object, I tryed using

dataiter = iter(data)
for i in dataiter:

but I get the error

TypeError                                 Traceback (most recent call last)
<ipython-input-32-daa759a0394f> in <module>()
      1 dataiter = iter(data)
----> 2 for i in dataiter:
      3   print(i)
      4   print(i.shape())

7 frames
/usr/local/lib/python3.6/dist-packages/torchvision/transforms/ in to_tensor(pic)
     40     """
     41     if not(_is_pil_image(pic) or _is_numpy(pic)):
---> 42         raise TypeError('pic should be PIL Image or ndarray. Got {}'.format(type(pic)))
     44     if _is_numpy(pic) and not _is_numpy_image(pic):

TypeError: pic should be PIL Image or ndarray. Got <class 'torch.Tensor'>

This dataset should already return tensors:

video (Tensor[T, H, W, C]): the T video frames
audio(Tensor[K, L]): the audio frames, where K is the number of channels and L is the number of points
label (int): class of the video clip

So I guess you would have to remove the ToTensor transformation.