-
Notifications
You must be signed in to change notification settings - Fork 1.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[bin] support model saving for both per-step & per-epoch mode #2248
Conversation
@@ -76,19 +78,35 @@ def train(self, model, optimizer, scheduler, data_loader, writer, configs, | |||
info_dict = update_parameter_and_lr(model, optimizer, | |||
scheduler, scaler, | |||
info_dict) | |||
save_interval = info_dict.get('save_interval', 10000) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
just like log_interval
, save_interval
is set in train.yaml?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这里我认为 : log save_interval等为训练行为 应该放到train的args里边
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
是的,都在train.yaml, 需要的话可以改到args
fix #1839