Skip to content

Fix for the moving average scores #35

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

Merged
Eugene-hu merged 2 commits into from
Jun 9, 2023
Merged

Fix for the moving average scores #35

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
20b7a3a
fixes the moving average scores and scatters
Eugene-hu Jun 8, 2023
9ad044a
Combine followup and answer rewards
opentaco Jun 9, 2023
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
13 changes: 5 additions & 8 deletions openvalidators/forward.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -339,14 +339,11 @@ async def forward(self):
rewards=answer_rewards,
)

# Compute forward pass rewards.
scattered_followup_rewards = (
torch.zeros((self.metagraph.n), dtype=torch.float32).to(self.device).scatter(0, followup_uids, followup_rewards)
)
scattered_answer_rewards = (
torch.zeros((self.metagraph.n), dtype=torch.float32).to(self.device).scatter(0, answer_uids, answer_rewards)
)
rewards = scattered_followup_rewards + scattered_answer_rewards
# Compute forward pass rewards, assumes followup_uids and answer_uids are mutually exclusive.
rewards = self.moving_averaged_scores.scatter(0, followup_uids, followup_rewards)
rewards = rewards.scatter(0, answer_uids, answer_rewards)

# Update moving_averaged_scores with rewards.
self.moving_averaged_scores = self.config.neuron.moving_average_alpha * rewards.to(self.device) + (
1 - self.config.neuron.moving_average_alpha
) * self.moving_averaged_scores.to(self.device)
Expand Down