A Study on the Impact of Class Distribution on Deep Learning - The Case of Histological Images and Cancer Detection
Résumé
Studies on deep learning tuning mostly focus on the neural network architectures and algorithms hyperparameters. Another core factor for accurate training is the class distribution of the training dataset. This paper contributes to understanding the optimal class distribution on the case for histological images used in cancer diagnosis. We formulate several hypotheses, which are then tested considering experiments with hundreds of trials. We considered both segmentation and classification tasks considering the U-net and group equivariant CNN (G-CNN). This paper is an extended abstract of another paper published by the authors 1 .
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |
