Can't train the algorithm with pytho2.7

[ktakanopy]

After some training, this error message is thrown. Please can you view the fix for this?

I modiefied the line

    # Get stored values.
    old_state_m, action_m, reward_m, new_state_m, terminal_m = memory

To:

    # Get stored values.
    old_state_m, action_m, reward_m, new_state_m, terminal_m,_ = memory

But i think the training is horrible.

`
none
Game time 0.842555046082
** Update target **

MAX MOVES

Game #: 70
Moves this round 150
Items collected 1
none
Game time 0.872440099716

MAX MOVES

Game #: 71
Moves this round 150
Items collected 2
none
Game time 0.815211057663
** Update target **

MAX MOVES

Game #: 72
Moves this round 150
Items collected 2
hit boundry
Game time 0.862430095673
** Update target **

MAX MOVES

Game #: 73
Moves this round 150
Items collected 3
none
Game time 0.874290943146

MAX MOVES

Game #: 74
Moves this round 150
Items collected 1
none
Game time 0.831968784332
** Update target **
Traceback (most recent call last):
File "train.py", line 88, in
parameters = Functions.train(parameters)
File "/home/kevintakano/UFAM/ReinforcementLearning/DeepQ/RL_Functions.py", line 334, in train
dbldqn=parameters['dqn'])
File "/home/kevintakano/UFAM/ReinforcementLearning/DeepQ/RL_Functions.py", line 405, in process_minibatch
old_state_m, action_m, reward_m, new_state_m, terminal_m = memory
ValueError: too many values to unpack

`