Table 2.
CNN architectures trained on the CFHT dataset.
| Fully connected (FC) layers |
|||||||
|---|---|---|---|---|---|---|---|
| Train | Number | Feature maps | Number | Number | Dropout | Total | |
| Model | size | F. maps | M × (F, N, S) | FC layers | neurons | % | weights |
| 11 | 250 | 3 | 2 × (5, 256, 2) | 3 | 328 | 0 | 22 848 778 |
| 1 × (5, 128, 2) | 192 | 50 | |||||
| 2 | – | ||||||
| 12 | 1000 | 3 | 1 × (9, 256, 3) | 3 | 152 | 95 | 4 551 906 |
| 2 × (5, 128, 2) | 88 | 95 | |||||
| 2 | – | ||||||
| 13 | 500 | 3 | 1 × (9, 256, 3) | 3 | 152 | 95 | 4 551 906 |
| 2 × (5, 128, 2) | 88 | 95 | |||||
| 2 | – | ||||||
| 14 | 250 | 3 | 1 × (9, 256, 3) | 3 | 152 | 95 | 4 551 906 |
| 2 × (5, 128, 2) | 88 | 95 | |||||
| 2 | – | ||||||
| 15 | 1000 | 3 | 1 × (9, 16, 3) | 3 | 152 | 50 | 875 586 |
| 2 × (5, 32, 2) | 88 | 50 | |||||
| 2 | – | ||||||
Notes. Models are defined by the feature maps, number of fully connected layers, and % dropout per fully connected layer, where F is the length of each filter, N is the number of filters used, S is the stride, and M is the number of feature maps. The ‘Number F. Maps’ column lists the total number of feature maps in each model. The ‘Total Weights’ column lists the total number of trainable weights in each model.
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.