You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for the awesome work.
We have to incrementally fine-tine AST model because our collected data are changed over time, or we have a new class of data.
In our case, we fine-tune the AST model with a 3-class dataset, and then fine-tune the resulting AST model with a 4-class dataset, or with the collected 3-class data which are changed over time.
Here, the first 3 classes of the 4-class dataset are the same as the 3-class dataset.
Is it possible to incrementally fine-tune AST model even if we have a 4-class dataset ?
Many thanks.
The text was updated successfully, but these errors were encountered:
jmren168
changed the title
How to fine-tune twice on the data collected continuously over time?
How to fine-tune AST model twice on the data collected continuously over time?
Aug 8, 2024
jmren168
changed the title
How to fine-tune AST model twice on the data collected continuously over time?
How to incrementally fine-tune AST model with the new data?
Aug 8, 2024
Hi @YuanGongND,
Thanks for the awesome work.
We have to incrementally fine-tine AST model because our collected data are changed over time, or we have a new class of data.
In our case, we fine-tune the AST model with a 3-class dataset, and then fine-tune the resulting AST model with a 4-class dataset, or with the collected 3-class data which are changed over time.
Here, the first 3 classes of the 4-class dataset are the same as the 3-class dataset.
Is it possible to incrementally fine-tune AST model even if we have a 4-class dataset ?
Many thanks.
The text was updated successfully, but these errors were encountered: