I was looking for ways to visualize model that we fine tuned in lesson one.
Is there any way to visualize what layers are final.
Is there any use and/or way to control how many final layers will be retrained?
I was looking for ways to visualize model that we fine tuned in lesson one.
Is there any way to visualize what layers are final.
Is there any use and/or way to control how many final layers will be retrained?
You can check class Learner()
in fastai/learner.py:
def freeze_to(self, n):
c=self.children
for l in c: set_trainable(l, False)
for l in c[n:]: set_trainable(l, True)
Yup we’ll be covering this in the next lesson.
have you solved this problem?if you did ,please show me the video link,Thank you