Fix the wrong default loss name for GAN models

2026-09-09 11:16:00 +08:00 · 2022-02-25 10:48:03 +01:00
parent 04202da1ac
commit a71a013276
1 changed files with 1 additions and 1 deletions
@@ -99,7 +99,7 @@ class BaseGANVocoderConfig(BaseVocoderConfig):
                "mel_fmax": None,
            }`
        target_loss (str):
-            Target loss name that defines the quality of the model. Defaults to `avg_G_loss`.
+            Target loss name that defines the quality of the model. Defaults to `G_avg_loss`.
        grad_clip (list):
            A list of gradient clipping theresholds for each optimizer. Any value less than 0 disables clipping.
            Defaults to [5, 5].