RuntimeError: Expected object of device type cuda but got device type cpu for argument #1 'self' in call to _th_mm

Hi,
I’m newbie of DGL, and try to make a working training model on it.
I installed dgl-cu102 using pip , but got an error as:

RuntimeError: Expected object of device type cuda but got device type cpu for argument #1 'self' in call to _th_mm

codes are based on here
Can I get any clue to resolve the above error? Thank you!

Duplicate of issue #1594