Skip to content

Latest commit

 

History

History
900 lines (485 loc) · 29 KB

File metadata and controls

900 lines (485 loc) · 29 KB

Public Models

All models were compiled using Hailo Dataflow Compiler v2.18.0.


Classification


Link Legend


Key / Icon Description
Networks used by Hailo-apps.
S Source – Link to the model's open-source repository.
PT Pretrained – Download the pretrained model file (ZIP format).
HEF, NV12, RGBX Compiled Models – Links to models in various formats: - HEF: RGB format - NV12: NV12 format - RGBX: RGBX format
PR Profiler Report – Download the model's performance profiling report.

Imagenet


Network Name float Accuracy (top1) Hardware Accuracy (top1) FPS (Batch Size=1) FPS (Batch Size=8) Links Input Resolution (HxWxC) Params (M) OPS (G)
cas_vit_m 81.2 80.9 47.8 157 384x384x3 12.42 10.89
cas_vit_s 79.9 79.7 71.1 226 384x384x3 5.5 5.4
cas_vit_t 81.9 81.6 31.6 95.9 384x384x3 21.76 20.85
davit_tiny 82.7 82.2 17.8 54.4 224x224x3 28.36 9.1
deit_base 80.9 80.4 28.5 107 224x224x3 80.26 35.22
deit_small 78.2 77.0 70.0 309 224x224x3 20.52 9.4
deit_tiny 69.1 68.5 89.1 406 224x224x3 5.3 2.57
efficientformer_l1 79.1 76.1 96.7 277 224x224x3 12.3 2.6
efficientnet_l 80.5 79.3 221 221 300x300x3 10.55 19.4
efficientnet_lite0 75.0 73.9 1952 1952 224x224x3 4.63 0.78
efficientnet_lite1 76.7 76.3 1668 1668 240x240x3 5.39 1.22
efficientnet_lite2 77.5 76.7 977 977 260x260x3 6.06 1.74
efficientnet_lite3 79.3 78.6 512 512 280x280x3 8.16 2.8
efficientnet_lite4 80.8 80.1 273 273 300x300x3 12.95 5.10
efficientnet_m 78.9 78.5 984 984 240x240x3 6.87 7.32
efficientnet_s 77.6 77.2 1232 1232 224x224x3 5.41 4.72
fastvit_sa12⭐ 79.8 76.7 1113 1113 224x224x3 11.99 3.59
hardnet39ds⭐ 73.4 73.1 1975 1975 224x224x3 3.48 0.86
hardnet68⭐ 75.5 75.2 137 400 224x224x3 17.56 8.5
inception_v1 69.7 69.5 2454 2454 224x224x3 6.62 3
levit128 78.4 76.0 128 675 224x224x3 9.2 0.8
levit192 79.7 77.2 134 707 224x224x3 10.9 1.3
levit256 81.4 79.1 100 522 224x224x3 18.9 2.3
levit384 82.3 78.6 65.6 325 224x224x3 39.1 4.7
mobilenet_v1 71.0 70.3 3305 3305 224x224x3 4.22 1.14
mobilenet_v2_1.0 71.8 70.9 2597 2597 224x224x3 3.49 0.62
mobilenet_v2_1.4 74.2 73.3 1652 1652 224x224x3 6.09 1.18
mobilenet_v3 72.2 71.8 3697 3697 224x224x3 4.07 2
regnetx_1.6gf 77.0 76.8 2551 2551 224x224x3 9.17 3.22
regnetx_800mf 75.2 74.9 3498 3498 224x224x3 7.24 1.6
repghost_1_0x 73.0 72.2 242 1071 224x224x3 4.1 0.28
repghost_2_0x 77.2 76.9 140 623 224x224x3 9.8 1.04
repvgg_a1 74.4 72.4 2707 2707 224x224x3 12.79 4.7
repvgg_a2 76.5 74.3 1132 1132 224x224x3 25.5 10.2
resmlp12_relu 75.3 75.0 1429 1429 224x224x3 15.77 6.04
resnet_v1_18⭐ 71.3 71.1 2718 2718 224x224x3 11.68 3.64
resnet_v1_34⭐ 72.7 72.2 1505 1505 224x224x3 21.79 7.34
resnet_v1_50⭐ 75.2 74.7 1372 1372 224x224x3 25.53 6.98
resnext26_32x4d 76.2 75.9 1630 1630 224x224x3 15.37 4.96
resnext50_32x4d 79.3 78.4 761 761 224x224x3 24.99 8.48
squeezenet_v1.1 59.8 59.4 3034 3034 224x224x3 1.24 0.78
swin_small 83.1 79.9 16.2 51.5 224x224x3 50 17.6
swin_tiny 81.3 79.3 28.7 81.2 224x224x3 29 9.1
vit_base 84.5 83.2 29.1 108 224x224x3 86.5 35.188
vit_base_bn 80.0 78.8 40.6 167 224x224x3 86.5 35.188
vit_small 81.5 80.0 70.1 309 224x224x3 21.12 8.62
vit_small_bn 78.1 77.1 131 640 224x224x3 21.12 8.62
vit_tiny 75.5 73.6 89.2 406 224x224x3 5.73 2.2
vit_tiny_bn 69.0 67.2 208 1081 224x224x3 5.73 2.2