I have multiple DNN that can extract features vector from images. Those can be used for two main goals:
- Use them for transfer learning ang faster trainings
- Use them as feature extractors, and train only additional, small "heads" (in particular if my dataset is really small)
But how can I choose which trained network to use? For the moment, I use:
- A UMAP projection on a labelled dataset (that the networks have never seen), and a compare visually how well my various classes are "clustered".
- A "Raw 10-NN Mean" score. For each point, I count how many of their 10 nearest neighbours are of the same class. I Divide by 10 to get the score.
- A "Class 10-NN Mean" score. Same as before, but each class has the same weight, regardless of their proportion in the dataset.
Are there any other interesting metrics I could use to compare my feature extractors?
Many thanks!
