brozek/rltorch
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Implemented epsilon as a scheduler

Browse source
This commit is contained in:
Brandon Rozek 2019-02-10 23:56:21 -05:00
parent 04e54cddc2
commit b2ab2ee132
1 changed files with 3 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

9
rltorch/action_selector/EpsilonGreedySelector.py
View file

@ -1,15 +1,12 @@
from .ArgMaxSelector import ArgMaxSelector
import numpy as np
class EpsilonGreedySelector(ArgMaxSelector):
def __init__(self, model, action_size, device = None, epsilon = 0.1, epsilon_decay = 1, epsilon_min = 0.1):
def __init__(self, model, action_size, device = None, epsilon = 0.1):
super(EpsilonGreedySelector, self).__init__(model, action_size, device = device)
self.epsilon = epsilon
self.epsilon_decay = epsilon_decay
self.epsilon_min = epsilon_min
# random_act is already implemented in ArgMaxSelector
# best_act is already implemented in ArgMaxSelector
def act(self, state):
action = self.random_act() if np.random.rand() < self.epsilon else self.best_act(state)
if self.epsilon > self.epsilon_min:
self.epsilon = self.epsilon * self.epsilon_decay
eps = next(self.epsilon) if isinstance(self.epsilon, collections.Iterable) else self.epsilon
action = self.random_act() if np.random.rand() < epsilon else self.best_act(state)
return action