facebookresearch / facebookresearch/AudioDec

Why do the feature matching loss and adversarial loss continue to increase during training?

Open
#42 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
515
Forks
33
PR merge metrics
No merged PRs in 30d

Description

Dear Author, I trained AudioDec_v2 without modifying any experimental settings, and the training loss changes are as follows. Why are the feature matching loss and adversarial loss continuously increasing? Is this behavior normal?

Image

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.