No class identified by the model

[deven367 (Deven Mistry)]

I have multi-label image classification model. When I use to predict on a test set, the model does not return any labels. I understand that the probabilities predicted in the last layer don’t cross the set threshold. How can I mitigate this? Also, is my understanding correct?

Here is the class distribution of the test data vs the predicted .

Column 1, in the second graph contains almost 500 images which are not classified as anything.

[meanpenguin (Edward)]

Maybe try to see what categories are ending up in the unknown?

interp = ClassificationInterpretation.from_learner(learn)
interp.plot_confusion_matrix()

Also, this link might help in adjusting the threshold

How to change the threshold of a prediction of multi-label classification

I am trying to create a model based in the content of chapter 6 of the book. I have a multi-label dataset that I’m using to train my model, using, as metrics, an accuracy function such as: def accuracy_multi1(inp, targ, thresh=0.5, sigmoid=True): "Compute accuracy when 'inp' and 'targ' are the same size" if sigmoid: inp=inp.sigmoid() return ((inp>thresh) == targ.bool()).float().mean() And my learner is like: learn = cnn_learner(dls, resnet50, metrics=partial(accuracy_multi1,thresh…