can not reproduce experiment shown in figure

[zcyang]

Hi, can you share a configuration that can reproduce the results you showed on the figure? I run the default M1 configuration and only get average episodic reward at around 3.

I tried to change the configurations like setting action_repeat = 4, change learning_rate, add double_q and duel_q, there is no much change.

Many thanks!

[carpedm20]

I think commit before Dec 2016 cause some problem. I'll dig into this and this is highly related to #21.

[zcyang]

any update on this?