| 12 | |
| 13 | |
| 14 | class LinearSchedule(object): |
| 15 | def __init__(self, schedule_timesteps, final_p, initial_p=1.0): |
| 16 | """Linear interpolation between initial_p and final_p over |
| 17 | schedule_timesteps. After this many timesteps pass final_p is |
| 18 | returned. |
| 19 | Parameters |
| 20 | ---------- |
| 21 | schedule_timesteps: int |
| 22 | Number of timesteps for which to linearly anneal initial_p |
| 23 | to final_p |
| 24 | initial_p: float |
| 25 | initial output value |
| 26 | final_p: float |
| 27 | final output value |
| 28 | """ |
| 29 | self.schedule_timesteps = schedule_timesteps |
| 30 | self.final_p = final_p |
| 31 | self.initial_p = initial_p |
| 32 | |
| 33 | def value(self, t): |
| 34 | """See Schedule.value""" |
| 35 | fraction = min(float(t) / self.schedule_timesteps, 1.0) |
| 36 | return self.initial_p + fraction * (self.final_p - self.initial_p) |
| 37 | |
| 38 | |
| 39 | class TimeLimit(gym.Wrapper): |