-
Notifications
You must be signed in to change notification settings - Fork 254
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
Small Fixes #14
Small Fixes #14
Conversation
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
def before_fit(self, learn): self.schedo = self.sched(learn.opt) | ||
def before_fit(self, learn): | ||
self.schedo = self.sched(learn.opt) | ||
learn.schedo = self.schedo |
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.
What's this for?
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.
To have access to the scheduler inside the Learner object. This way, I can monitor the learning rate or other scheduler params.
Iniside the MonitorCB
now I can access the learn.schedo
.
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.
Can you do that using the optimiser hyperparams instead?
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.
oh, you are right =)
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.
opt.param_groups[0]["lr"]
returns the current learning rate from the optimizer
object
Thanks! Can you make this a more minimal diff without all the changes to images and other outputs, and remove the stacktrace that appears in one notebook? |
Ok, I did |
Unfortunately that deleted all the outputs! |
Ok, re run everything, but lot's of warning on M1Pro |
I will close this then... don't know how to resolve the conflicts |
@@ -61,7 +61,7 @@ def subplots( | |||
if figsize is None: figsize=(ncols*imsize, nrows*imsize) | |||
fig,ax = plt.subplots(nrows, ncols, figsize=figsize, **kwargs) | |||
if suptitle is not None: fig.suptitle(suptitle) | |||
if nrows*ncols==1: ax = array([ax]) | |||
if nrows*ncols==1: ax = np.array([ax]) |
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.
This change seems to already be in master
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.
=)
I'll fix |
This PR fixes the missing args for
show_images
and adds thescheduler
to the learner object