b2aad723220e31bc8e950c112b557732b608b97a,PyPi/algorithms/td.py,DoubleQLearning,fit,#DoubleQLearning#Any#Any#,86

Before Change

        assert n_fit_iterations == 1

        state, action, reward, next_state, absorbing, _ =\
            parse_dataset(dataset[-1],
                          self.mdp_info["observation_space"].dim,
                          self.mdp_info["action_space"].dim)

        sa = (state, action)
        sa_idx = np.append(self.mdp_info["observation_space"].get_idx(state),
                           self.mdp_info["action_space"].get_idx(action))

After Change

        
        assert n_fit_iterations == 1

        state, action, reward, next_state, absorbing, _ = parse_dataset(
            dataset[-1])

        sa = [state, action]
        sa_idx = np.append(self.mdp_info["observation_space"].get_idx(state),
                           self.mdp_info["action_space"].get_idx(action))

Instances