You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+1Lines changed: 1 addition & 0 deletions
Original file line number
Diff line number
Diff line change
@@ -214,6 +214,7 @@ $ accelerate launch train.py
214
214
-[x] remove any magic being done with automatic rgbs processing, and have it explicitly passed in - offer functions on the discriminator that can process real images into the right multi-scales
215
215
-[x] add horizontal flip for starters
216
216
217
+
-[ ] move all modulation projections into the adaptive conv2d class
217
218
-[ ] add accelerate
218
219
-[x] works single machine
219
220
-[x] works for mixed precision (make sure gradient penalty is scaled correctly), take care of manual scaler saving and reloading, borrow from imagen-pytorch
0 commit comments