@citnaj I was very impressed with your DeOldify project that Jeremy showed today in the last fastai lecture ! Great job !
May I ask what are your plans for the future of the project ?
Thanks, and glad you asked! Along with working on this week's gan material that you saw today, the work on that behind the scenes we did is going to go into DeOldify. I'm starting work tomorrow. I'm psyched because we made some pretty big improvements.
-
-
That's great to hear ! I suppose you will be porting the project so
@fastdotai 1.0 some time in the future ?Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
For one- I've never seen pretraining of gans like what we did there, and I'm pretty sure that's going to improve things noticeably in DeOldify. Two- the model itself is improved with things like that DenseBlock stuff he talked about.
-
Three- native http://Fast.AI support is just plain awesome to have for this stuff. Didn't have that before. Will have that when I port to http://Fast.AI 1.0 in the next week or so (to answer your question about that).
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.