Detecting objects in dataset - none of the above option

I want to train a model that can identify cats vs dogs vs everything else. Assume that I only have data labeled for cats and dogs. So my model consists of an output column of size 2 - this means any image I give to the model, even if it’s a car, would automatically get slotted as a cat/dog. Is there any way to add a ‘none of the above’ option to predictor?

1 Like