Skip to content

Fix: prevent variance underflow causing NaN losses #537

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
Kapadiaaa wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix: prevent variance underflow causing NaN losses #537

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
2 changes: 2 additions & 0 deletions neural_lam/models/base_graph_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -351,7 +351,9 @@ def predict_step(self, prev_state, prev_prev_state, forcing):
# NOTE: The predicted std. is not scaled in any way here
# linter for some reason does not think softplus is callable
# pylint: disable-next=not-callable
eps = 1e-6
pred_std = torch.nn.functional.softplus(pred_std_raw)
pred_std = torch.clamp(pred_std, min=eps)
else:
pred_delta_mean = net_output
pred_std = None
Expand Down