ML bias 101: your pretrained convnet does not have a general, nor objective knowledge of the visual world. It's mostly dogs and some birds, and then the occasional seatbelt. It's crazy how many people apparently expect otherwise
Show this thread
Everything is a dog when you use ImageNet pretrained weights
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Hm. I must be using the wrong weights — my model says everything is a “Traceback (most recent call last): File "test_imagenet.py", line 40, in model = VGG16(weights="imagenet")”
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
"On the ImageNet, nobody knows everything is a dog"
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Just call them neutral networks
- Show replies
-
-
-
I think it does happen when the rules of transfer learning are not properly followed.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Any good progress for reinitializing the weights like half the network for a custom dataset?
-
Xavier initialization?
End of conversation
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
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.