Is difference merging to a new different model from a trained model the same as training the new model? #1314
-
|
will difference merging give inferior results than directly training? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
|
It really depends on the model and the trained subject, you'll have to do some testing to find what best suits your needs. |
Beta Was this translation helpful? Give feedback.
-
|
ok thanks |
Beta Was this translation helpful? Give feedback.
-
|
I did difference merging once or twice and got lots of artifacts, so I prefer to train directly |
Beta Was this translation helpful? Give feedback.
-
|
oh ok thanks |
Beta Was this translation helpful? Give feedback.
It really depends on the model and the trained subject, you'll have to do some testing to find what best suits your needs.