Image Classification on ImageNet-1K 2009 (val)
87.5Top-1 AccuracyConvNeXt-L
Evaluation results
| Method | Links | |
|---|---|---|
| ConvNeXt-LImage Size (Px)=384, #Param (M)=198.0, FLOPs (G)=101.0, Throughput (Img/Sec)=172, Pre-trained=ImageNet-21K2023.06 | 87.5 | |
| FasterViT-4Image Size (Px)=384, #Param (M)=424.6, FLOPs (G)=119.2, Throughput (Img/Sec)=281, Pre-trained=ImageNet-21K2023.06 | 87.5 | |
| Swin-LImage Size (Px)=384, #Param (M)=197.0, FLOPs (G)=103.9, Throughput (Img/Sec)=206, Pre-trained=ImageNet-21K2023.06 | 87.3 | |
| ConvNeXt-LImage Size (Px)=224, #Param (M)=198.0, FLOPs (G)=34.4, Throughput (Img/Sec)=508, Pre-trained=ImageNet-21K2023.06 | 86.6 | |
| FasterViT-4Image Size (Px)=224, #Param (M)=424.6, FLOPs (G)=36.6, Throughput (Img/Sec)=849, Pre-trained=ImageNet-21K2023.06 | 86.6 | |
| Swin-LImage Size (Px)=224, #Param (M)=197.0, FLOPs (G)=34.5, Throughput (Img/Sec)=787, Pre-trained=ImageNet-21K2023.06 | 86.3 | |
| ViT-L/16Image Size (Px)=384, #Param (M)=307.0, FLOPs (G)=190.7, Throughput (Img/Sec)=149, Pre-trained=ImageNet-21K2023.06 | 85.2 | |
| MEGALODONNumber of Parameters=90M2024.04 | 83.1 | |
| MEGANumber of Parameters=90M2024.04 | 82.3 | |
| DeiT-BNumber of Parameters=86M2024.04 | 81.8 | |
| ResNet-152Number of Parameters=60M2024.04 | 78.3 | |
| ViT-BNumber of Parameters=86M2024.04 | 77.9 |