Are labels associated with a model or a dataset?

I'm not sure if I have this backwards or not, so I'll explain a bit of what is going on.

I want to use Unity's Barracuda api to use an onnx model for classification and detection (depending on the model). Some example projects I've found have a model and labels handy, so it's easy to map the outputs to the labels. ie, if the 5th element has the highest score, i can look up the 5th label and find it is a cat.

However, in my explorations, namely the Onnx Model Zoo as well as downloading and inspecting datasets using Fiftyone, I can only find labels corresponding to a dataset. Perhaps I've answered my own question, but I'll pose it.

When a model is trained, is there a way to find the labels associated with the outputs? Or is that dependant on the dataset it was trained on? (meaning, I should find the dataset it was trained on and use those labels instead)

Topic machine-learning-model image-classification

Category Data Science


If I understood you correctly; you're wondering how to get the label corresponding to the output from a model?

There's usually some mapping between the output vector and the label, ref.

You'll have to find this mapping for the specific model. Perhaps in the documentation of the model you are using?

Hope this is of some help to you.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.