Open Access
Table A.3
Hyperparameters for the GP model’s feature extractor.
| Property | Value |
|---|---|
| Feature extractor type | Convolutional neural network |
| Number of hidden layers | 3 |
| 1st layer type | Convolutional (1D) |
| 1st layer input channels | 1 |
| 1st layer output channels | 16 |
| 1st layer kernel size | 3 |
| 1st layer padding | same |
| 1st layer activation | ReLU |
| 2nd layer type | Max pooling (1D) |
| 2nd layer kernel size | 2 |
| 2nd layer stride | 2 |
| 3rd layer type | Fully connected |
| 3rd layer output size | Original input size |
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.