AttributeError: 'generator' object has no attribute 'data' when manually loading MNIST arrf file

Question

I cant use fetch_openml() in aws, and have to download external datasets manually. I have downloaded the mnist_784.arff file from OpenML (https://www.openml.org/search?type=data&sort=runs&id=554&status=active), and trying to load it in my jupyter notebook. I run

import arff
mnist_dataset = arff.load(open('./mnist_784.arff'))


max_value = 255
mean = 0.1307
std = 0.3081


data = (mnist_dataset.data) / max_value
data = ((data - mean) / std).round(decimals=4)

and am getting this error,

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[17], line 6
      2 mean = 0.1307
      3 std = 0.3081
----> 6 data = (mnist_dataset.data) / max_value
      7 data = ((data - mean) / std).round(decimals=4)
      9 target = mnist_dataset.target.astype("int")

AttributeError: 'generator' object has no attribute 'data'