Which line do you need to add to train on 8 GPUs?
Convert horses to zebras with this CycleGAN model by @A_K_Nain, now on http://keras.io :
https://keras.io/examples/generative/cyclegan/ …
Definitely the most concise and elegant implementation of CycleGAN I've seen anywhere -- around 350 lines end-to-end. Train on 8 GPUs by just adding 1 line.pic.twitter.com/ZgHg0TTaHC
-
-
-
If you have 8 GPUs on your machine, just add the line `with MirroredStrategy().scope():` around model creation & compilation. If you don't, then the line is `http://tfc.run (...)`, see:https://github.com/tensorflow/cloud/blob/master/src/python/tensorflow_cloud/core/README.md …
- Show replies
New conversation -
-
-
Nice work Akash!!!..
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
pix2pix is really powerful
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.