Skip to content

Fix negative probability values #18

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
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Fix negative probability values #18

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
8439d96
Fix actions probabilities negative values issue
Dec 12, 2019
3e86023
Merge pull request #1 from AminaBa/AminaBa-patch-1
Dec 12, 2019
68c09fb
Fix negative probability values issue
Dec 12, 2019
bcba139
Update policy_output.py
Dec 12, 2019
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
14 changes: 9 additions & 5 deletions agent/asyncrl/policy_output.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,11 +18,15 @@ def _sample_discrete_actions(batch_probs):
List consisting of sampled actions
"""
action_indices = []

# Subtract a tiny value from probabilities in order to avoid
# "ValueError: sum(pvals[:-1]) > 1.0" in numpy.multinomial
batch_probs = batch_probs - np.finfo(np.float32).epsneg

# Prevent having a vector which sum is not in [0, 1]
while not 0 < np.sum(batch_probs) < 1:
# Subtract a tiny value from probabilities in order to avoid
# "ValueError: sum(pvals[:-1]) > 1.0" in numpy.multinomial
batch_probs = batch_probs - np.finfo(np.float32).epsneg
# Apply abs function to keep probability values positive to avoid
# "ValueError: pvals < 0, pvals > 1 or pvals contains NaNs" in numpy.multinomial
batch_probs = np.absolute(batch_probs)

for i in range(batch_probs.shape[0]):
histogram = np.random.multinomial(1, batch_probs[i])
action_indices.append(int(np.nonzero(histogram)[0]))
Expand Down