Free Access
Table 3.
Number of epochs needed to train the model and computing time in second required to process the test data for different choices of loss functions.
| Loss function | Training | Testing |
|---|---|---|
| (number of epochs) | (Computation time (s)) | |
| Binary entropy | 24 | 149.7 |
| Hinge | 100 | 148.9 |
| KL | 100 | 148.3 |
| Wasserstein | 100 | 145.7 |
| Lerr | 20 | 146.8 |
| (Lerr + Ltv) | 23 | 146.8 |
Notes. Each epoch consists of 32 samples from data set 1.
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.