Repeat the experiment for serveral times
repExperiment(sname, aname, conf, nrep = 5L, nepi, value_fun = NULL, ...)
| sname | The scenario name of Gym environment |
|---|---|
| aname | The name of the Agent |
| conf | Configuration object |
| nrep | Number of repetitions |
| nepi | Number of episode to learn |
| value_fun | customized neural network as value function approximator, default NULL |
| ... | Other Parameters to pass to GymEnv |
list of ggplot2 object for performance and list of reward per experiment per episode