I am trying to load in MNIST dataset and although I know that there are many different ways to do so, I am having trouble in loading the dataset via the following method -
#image_location is already there in the variable path_digit[]
for i in range(10):
print("check")
stacked_training[i] = torch.stack([tensor(Image.open(o)) for o in (path_digit[i])]).float()/255
print(stacked_training[i].shape, '\n')
below is the error code -
check
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
<ipython-input-10-bb2269508aed> in <module>()
----> 4 stacked_training[i] = torch.stack([tensor(Image.open(o)) for o in (path_digit[i])]).float()/255
5 print(stacked_training[i].shape, '\n')
RuntimeError: expand(torch.FloatTensor{[5923, 28, 28]}, size=[28, 28]): the number of sizes provided (2) must be greater or equal to the number of dimensions in the tensor (3)