Appropriately reweight loss function based on arena size?

neurostatslab / vocalocator

Deep neural networks for sound source localization and vocalization attribution.

MIT License

2 stars 0 forks source link

Appropriately reweight loss function based on arena size? #33

Closed amanchoudhri closed 3 months ago

amanchoudhri commented 1 year ago

Currently, errors along the long axis of the arena are effectively penalized less than errors along the short side. Is there some way we can fix this?