Image Retrieval On Muge Retrieval
Metriken
Mean Recall
R@1
R@10
R@5
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | Mean Recall | R@1 | R@10 | R@5 |
|---|---|---|---|---|
| chinese-clip-contrastive-vision-language | 83.6 | 68.9 | 93.1 | 88.7 |
| chinese-clip-contrastive-vision-language | 69.2 | 48.6 | 84.0 | 75.1 |
| chinese-clip-contrastive-vision-language | 77.4 | 58.4 | 90.0 | 83.6 |
| wukong-100-million-large-scale-chinese-cross | 72.1 | 52.7 | 85.6 | 77.9 |
| chinese-clip-contrastive-vision-language | 80.1 | 63.3 | 91.3 | 85.6 |
| wukong-100-million-large-scale-chinese-cross | 61.2 | 39.2 | 77.4 | 66.9 |
| zero-and-r2d2-a-large-scale-chinese-cross | 77.5 | 60.1 | 89.4 | 82.9 |
| zero-and-r2d2-a-large-scale-chinese-cross | 68.7 | 47.4 | 83.5 | 75.1 |
| chinese-clip-contrastive-vision-language | 81.3 | 65.3 | 92.1 | 86.7 |