todo

# TODO:
    - Sensitive analysis on this with causal discovery: 10-20-50-100 bins discretization
    - Averaged mean for reward_action_values in the action selection process, decide if the action with lower score
        could be removed



source MyEnv/bin/activate
python setup.py install
export DISPLAY=':99.0'
Xvfb :99 -screen 0 1400x900x24 > /dev/null 2>&1 &

# headless

{0.08333333333333393:
    {0.0: 0.12, 1.0: 0.10678260869565218, 2.0: 0.11252173913043478, 3.0: 0.1137391304347826, 4.0: 0.10730434782608696, 5.0: 0.1088695652173913, 6.0: 0.11547826086956521, 7.0: 0.10504347826086957, 8.0: 0.1102608695652174},
 -1.1388888888888886:
    {0.0: 0.11999999999999998, 1.0: 0.10678260869565218, 2.0: 0.11252173913043478, 3.0: 0.1137391304347826, 4.0: 0.10730434782608696, 5.0: 0.1088695652173913, 6.0: 0.11547826086956521, 7.0: 0.10504347826086957, 8.0: 0.1102608695652174},
 -1.7499999999999998:
    {0.0: 0.11999999999999998, 1.0: 0.10678260869565216, 2.0: 0.11252173913043477, 3.0: 0.11373913043478258, 4.0: 0.10730434782608694, 5.0: 0.10886956521739129, 6.0: 0.1154782608695652, 7.0: 0.10504347826086954, 8.0: 0.11026086956521737}}