A motor has a two rotors which are independently operated using a single inverter and without being supplied with a complex current. Further the motor is capable of shifting a rotation direction of each of two rotors by performing a magnetizing operation one time and reversing polarities of the shifting magnets when a mode of the motor is shifted from a synchronous mode in which the first rotor and the second rotor rotate in the same direction to a counter mode in which the first rotor and the second rotor rotate in opposite directions, and shifting the mode of the motor to the counter mode while the magnetizing operation is performed one time times and relative positions of the first rotor and the second rotor are changed when the first rotor and the second rotor inertially rotate in the synchronous mode.

A machine learning device which learns a current command for a motor, the machine learning device including a state observation unit which observes a state variable including a motor rotation speed or a motor torque command of the motor and at least one of a motor torque, a motor current, and a motor voltage of the motor; and a learning unit which learns the current command for the motor based on the state variable.