[FIXED] transforms.ToTensor() and Numpy Issue

Issue

I'm working on MNIST datasets using Pytorch and I'm trying to scale the images, I ran into problems associated with Numpy

train_dataset = datasets.MNIST(root='data',
                                train=True,
                                transform=transforms.ToTensor(),
                                download=True)

And here's my error, really confused on how to solve this

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-5-ef2899ce3492> in <module>
      6                                 train=True,
      7                                 transform=transforms.ToTensor(),
----> 8                                 download=True)
      9 
     10 test_dataset = datasets.MNIST(root='data',

~\miniconda3\envs\6.86x\lib\site-packages\torchvision\datasets\mnist.py in __init__(self, root, train, transform, target_transform, download)
     91                                ' You can use download=True to download it')
     92 
---> 93         self.data, self.targets = self._load_data()
     94 
     95     def _check_legacy_exist(self):

~\miniconda3\envs\6.86x\lib\site-packages\torchvision\datasets\mnist.py in _load_data(self)
    110     def _load_data(self):
    111         image_file = f"{'train' if self.train else 't10k'}-images-idx3-ubyte"
--> 112         data = read_image_file(os.path.join(self.raw_folder, image_file))
    113 
    114         label_file = f"{'train' if self.train else 't10k'}-labels-idx1-ubyte"

~\miniconda3\envs\6.86x\lib\site-packages\torchvision\datasets\mnist.py in read_image_file(path)
    507 
    508 def read_image_file(path: str) -> torch.Tensor:
--> 509     x = read_sn3_pascalvincent_tensor(path, strict=False)
    510     assert(x.dtype == torch.uint8)
    511     assert(x.ndimension() == 3)

~\miniconda3\envs\6.86x\lib\site-packages\torchvision\datasets\mnist.py in read_sn3_pascalvincent_tensor(path, strict)
    496     parsed = np.frombuffer(data, dtype=m[1], offset=(4 * (nd + 1)))
    497     assert parsed.shape[0] == np.prod(s) or not strict
--> 498     return torch.from_numpy(parsed.astype(m[2])).view(*s)
    499 
    500 

RuntimeError: Numpy is not available

Solution

Figured it out, my Numpy version was outdated compared to pytorch. Updated it basically!

Answered By - Muhammad Bello