Loading MNIST as Pytorch Dataset

## Describe the bug
Conversion of MNIST dataset to pytorch fails with bug

## Steps to reproduce the bug
```python
from datasets import load_dataset
dataset = load_dataset("mnist", split="train")
dataset.set_format('torch')
dataset[0]
print()
```

## Expected results
Expect to see torch tensors image and label

## Actual results
Traceback (most recent call last):
File "C:\Program Files\JetBrains\PyCharm 2020.3.3\plugins\python\helpers\pydev\pydevd.py", line 1491, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
File "C:\Program Files\JetBrains\PyCharm 2020.3.3\plugins\python\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "C:/Users/chapm/PycharmProjects/multiviewdata/multiviewdata/huggingface/mnist.py", line 13, in
dataset[0]
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\arrow_dataset.py", line 2154, in __getitem__
return self._getitem(
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\arrow_dataset.py", line 2139, in _getitem
formatted_output = format_table(
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\formatting.py", line 532, in format_table
return formatter(pa_table, query_type=query_type)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\formatting.py", line 281, in __call__
return self.format_row(pa_table)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\torch_formatter.py", line 58, in format_row
return self.recursive_tensorize(row)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\torch_formatter.py", line 54, in recursive_tensorize
return map_nested(self._recursive_tensorize, data_struct, map_list=False)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\utils\py_utils.py", line 356, in map_nested
mapped = [
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\utils\py_utils.py", line 357, in
_single_map_nested((function, obj, types, None, True, None))
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\utils\py_utils.py", line 309, in _single_map_nested
return {k: _single_map_nested((function, v, types, None, True, None)) for k, v in pbar}
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\utils\py_utils.py", line 309, in
return {k: _single_map_nested((function, v, types, None, True, None)) for k, v in pbar}
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\utils\py_utils.py", line 293, in _single_map_nested
return function(data_struct)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\torch_formatter.py", line 51, in _recursive_tensorize
return self._tensorize(data_struct)
File "C:\Users\chapm\PycharmProjects\multiviewdata\venv\lib\site-packages\datasets\formatting\torch_formatter.py", line 38, in _tensorize
if np.issubdtype(value.dtype, np.integer):
AttributeError: 'bytes' object has no attribute 'dtype'
python-BaseException

## Environment info

- `datasets` version: 2.3.2
- Platform: Windows-10-10.0.22579-SP0
- Python version: 3.9.2
- PyArrow version: 8.0.0
- Pandas version: 1.4.1