diff options
Diffstat (limited to 'training')
| -rw-r--r-- | training/strategy/ti.py | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/training/strategy/ti.py b/training/strategy/ti.py index d735dac..720ebf3 100644 --- a/training/strategy/ti.py +++ b/training/strategy/ti.py | |||
| @@ -114,11 +114,12 @@ def textual_inversion_strategy_callbacks( | |||
| 114 | return torch.stack(params) if len(params) != 0 else None | 114 | return torch.stack(params) if len(params) != 0 else None |
| 115 | 115 | ||
| 116 | @torch.no_grad() | 116 | @torch.no_grad() |
| 117 | def on_after_optimize(w, lr: float): | 117 | def on_after_optimize(w, lrs: dict[str, float]): |
| 118 | if ema_embeddings is not None: | 118 | if ema_embeddings is not None: |
| 119 | ema_embeddings.step(text_encoder.text_model.embeddings.token_override_embedding.params.parameters()) | 119 | ema_embeddings.step(text_encoder.text_model.embeddings.token_override_embedding.params.parameters()) |
| 120 | 120 | ||
| 121 | if use_emb_decay and w is not None: | 121 | if use_emb_decay and w is not None: |
| 122 | lr = lrs["emb"] or lrs["0"] | ||
| 122 | lambda_ = emb_decay * lr | 123 | lambda_ = emb_decay * lr |
| 123 | 124 | ||
| 124 | if lambda_ != 0: | 125 | if lambda_ != 0: |