Fixed a warning issue when accessing the internal storage of the torch_geometric dataset and method iteration error.

[koguhnhyeok]

Here is my version of pytorch and torch_geometric:
Pytorch: 2.1.0+cu11.8
torch_geometric: 2.4.0

1. Fixed a warning issue when accessing the internal storage of the torch_geometric dataset.
When running the loader_pyg.py code in the above environment, the following warning message occurred.

UserWarning: It is not recommended to directly access the internal storage format data of an 'InMemoryDataset'. If you are absolutely certain what you are doing, access the internal storage via InMemoryDataset._data instead to suppress this warning. Alternatively, you can access stacked individual attributes of every graph via dataset.{attr_name}.

So, I just follow the instruction of the pyg and change all the dataset.data.{attr_name} to dataset._data.{attr_name}

2. Solved a method iteration error when allocating the dataset._data.keys to the iteration object of the for statement
TypeError occurs when running lines 206 and 210 for key in dataset._data.keys:, because the dataset._data.keys are the bound method BaseData.keys of data in my torch_geometric version. So I added parentheses to return the list of the keys.

This is my first PR, so if you have any problems, please let me know! and thank you for your great project.