Image Classification on Rectangles (test)
88.21Accuracy (%)DVIP 3
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DVIP 3Model Type=Deep Variational Implicit Process, Number of layers (L)=32026.06 | 88.21 | 0.295 | 95.3 | |
| DVIP 2Model Type=Deep Variational Implicit Process, Number of layers (L)=22026.06 | 87.84 | 0.306 | 95 | |
| DVIP 4Model Type=Deep Variational Implicit Process, Number of layers (L)=42026.06 | 87.43 | 0.309 | 94.7 | |
| DVIP 5Model Type=Deep Variational Implicit Process, Number of layers (L)=52026.06 | 86.49 | 0.32 | 93.9 | |
| VIPModel Type=Single-layer, Number of layers (L)=12026.06 | 85.63 | 0.348 | 93 | |
| DGP 3Model Type=Deep Gaussian Process, Number of layers (L)=3, Source=Salimbeni and Deisenroth (2017)2026.06 | 75.16 | 0.47 | 85.8 | |
| SGPModel Type=Single-layer, Number of layers (L)=12026.06 | 72.54 | 0.518 | 82.8 |