frankkramer-lab / MIScnn

A framework for Medical Image Segmentation with Convolutional Neural Networks and Deep Learning
GNU General Public License v3.0
402 stars 116 forks source link

Implementing Multi Model #53

Closed Deathlymad closed 1 year ago

Deathlymad commented 3 years ago

Add a Pipeline Element that applies merges multiple models into a result. To merge different folds for example.

Potentially allow for multiple functions like Linear combination, Polynominal combination or class based combination.

For example have sevelar networks trained on the same image set and have an additional one trained on the divergence of each of the models predictions to predict the ideal model used in each point.