What to do with training data if model has been reject? #178
Unanswered
adambelniak
asked this question in
Q&A
Replies: 2 comments
-
|
I don't see any reason why doing so. If the model has been rejected doesn't mean that the most recent training data is bad. Its more the opposite: It means that the training data is a valuable source for getting the agent smarter. |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
That's clear. I'm wondering, how we indicate that collected moves, during self-play phase, are bad? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
What do you thing about removing last training history if model has been rejected ?
Beta Was this translation helpful? Give feedback.
All reactions