unit8co/darts

Improve numerical stability of the NLL computation for some distributions

Open

#621 ouverte le 1 déc. 2021

Voir sur GitHub
 (0 commentaires) (0 réactions) (0 assignés)Python (762 forks)batch import
good first issueimprovement

Métriques du dépôt

Stars
 (6 832 stars)
Métriques de merge PR
 (Merge moyen 143j 16h) (16 PRs mergées en 30 j)

Description

Not all Likelihoods have numerically-stable computaiton of the NLL. By default, we rely on the log_prob() function of the PyTorch Distribution objects, but it might not be the best way to get numerically stable gradients. I think it’s fairly easy to build cases where Negative Binomial fails for instance.

See: #663

Also for the Weibull distribution, the outputs changed when we changed the shape of the network outputs, which is weird and ideally should be investigated. Not clear if that’s due to numerical instability…

Guide contributeur