Open Access
Fig. 2
Download original image
Schema of the training procedure in O’TRAIN: In one training epoch, the original data set is split into two sub sets to train/update the CNN model and validate the classification results on a new batch of images. By repeating N times these epochs, the model converges towards the desired classification performances.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.