(Fun with lesson8) Rotation adjustment -- things you can do without annotated dataset!

(YangLu) #1

There are a lot of fun things to do with un-annotated dataset, this is one of them.

I have taken the PASCAL dataset and thrown away all annotation. Next I wrote a few custom transformers, so that the pictures will be rotated, and the y values are rotations.

Add custom head, find optimum LR, freeze and unfreeze head and fit a few times.
After 5 minutes of training, I got some somewhat satisfying results.

… I thought it was cool…

(DOTAN A) #2

So you have learned the rotations? Very nice!
which architecture did you use? are those results of test data that wasn’t part of the training set?

(Jeremy Howard) #3

That’s terrific! If you’re interested, it might be worth training that on a larger dataset (e.g. you could combine COCO + Pascal 2007 and 2012) and trying to get a model that works really well, then maybe create a little web app or client-side app that people can use to fix up their photos! :slight_smile:

I also think this would make a great blog post.

(YangLu) #4

Pretty much exactly the same as Lesson one architecture, just with custom head.

(YangLu) #5


I can totally see this as python flask web app, deployed on pythonanywhere for 3 months for free.

For now, I’ll just write a quick blog post maybe, and move on to lecture 9mod7=2 — my plan is to make a small project/ for each lecture, and I’m already tad behind…

(Jeremy Howard) #6

Lesson 9mod7 perhaps should be the new official naming scheme.

(Dominique Gilleman) #7

Hi, maybe you can share your notebook ?