search

Farama-Foundation / D4RL

A collection of reference environments for offline reinforcement learning
Apache License 2.0
1.35k stars 285 forks source link

possible to get a whole expert trajectory? #157

Open yiwc opened 2 years ago

yiwc commented 2 years ago

hi team,

I know we now provide a dataset of [s,a,r,s' ] may I know if it's possible to get the whole expert trajectory? for example [s,a,r, s',a'r', s'',a'',r'' ....]

Appreciate it!

nuomizai commented 2 years ago

The samples in dataset are time-continuous, which means that the datasets are consist of several whole expert trajectories. Each trajectory is seperated by the timeout variable. At the end of each trajectory, timeout=True.