INNER CODE UNIT · Python

value

rlcode/reinforcement-learning · 1-grid-world/2-value_iteration.py:45

            value = (reward + self.discount_factor * next_value)

            if value > max_value:
                action_list.clear()
                action_list.append(action)
                max_value = value
            elif value == max_value:
                action_list.append(action)

        return action_list

    def get_value(self, state):
        return round(self.value_table[state[0]][state[1]], 2)

if __name__ == "__main__":
    env = Env()
    value_iteration = ValueIteration(env)
    display = GraphicDisplay(value_iteration, title="Value Iteration")

View source record →