Skip to content

Added linear scheduling option to epsilon greedy exploration #66

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
cdevin wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Added linear scheduling option to epsilon greedy exploration #66

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 11 additions & 4 deletions rlkit/exploration_strategies/epsilon_greedy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,17 +1,24 @@
import random

from rlkit.exploration_strategies.base import RawExplorationStrategy
from rlkit.util.ml_util import LinearSchedule


class EpsilonGreedy(RawExplorationStrategy):
"""
Take a random discrete action with some probability.
"""
def __init__(self, action_space, prob_random_action=0.1):
self.prob_random_action = prob_random_action
def __init__(self, action_space, prob_random_action=0.1, prob_end=None, steps=1e6):
"""
If prob_end is None, this will default to a fixed schedule.
"""
if prob_end is not None:
self.prob_random_action = LinearSchedule(prob_random_action, prob_end, steps)
else:
self.prob_random_action = LinearSchedule(prob_random_action, prob_random_action, steps)
self.action_space = action_space

def get_action_from_raw_action(self, action, **kwargs):
if random.random() <= self.prob_random_action:
def get_action_from_raw_action(self, action,t=0 **kwargs):
if random.random() <= self.prob_random_action.get_value(t):
return self.action_space.sample()
return action