I had this problem today, too. My solution is changing the „np.float“ into „np.float64“ in the code which causes the error. Which code caused the problem needs to check the error message (traceback). After renaming it, the error message has been fixed.