Skip to content

Latest commit

 

History

History
242 lines (239 loc) · 31.9 KB

results_imagenet.md

File metadata and controls

242 lines (239 loc) · 31.9 KB

Flowvision Model Zoo

model top1 top1_err top5 top5_err parameters img_size crop_pct interpolation
alexnet 56.554 43.446 79.082 20.918 61.1M 224 0.875 bilinear
vgg11 69.040 30.960 88.634 11.366 132.9M 224 0.875 bilinear
vgg13 69.940 30.060 89.256 10.742 133.1M 224 0.875 bilinear
vgg16 71.586 28.414 90.390 9.610 138.4M 224 0.875 bilinear
vgg19 72.394 27.606 90.886 9.114 143.7M 224 0.875 bilinear
vgg11_bn 70.378 29.622 89.814 10.186 132.9M 224 0.875 bilinear
vgg13_bn 71.554 28.446 90.374 9.626 133.1M 224 0.875 bilinear
vgg16_bn 73.370 26.630 91.502 8.468 138.4M 224 0.875 bilinear
vgg19_bn 74.242 25.758 91.846 8.154 143.7M 224 0.875 bilinear
resnet18 69.760 30.240 89.082 10.918 11.7M 224 0.875 bilinear
resnet34 73.302 26.698 91.420 8.580 21.8M 224 0.875 bilinear
resnet50 76.146 23.854 92.872 7.128 25.6M 224 0.875 bilinear
resnet101 77.366 22.634 93.562 6.438 44.6M 224 0.875 bilinear
resnet152 78.314 21.686 94.060 5.940 60.2M 224 0.875 bilinear
squeezenet1_0 58.096 41.904 80.420 19.580 1.2M 224 0.875 bilinear
squeezenet1_0 58.188 41.813 80.616 19.384 1.2M 224 0.875 bilinear
densenet121 74.438 25.562 91.970 8.030 8.0M 224 0.875 bilinear
densenet169 75.588 24.412 92.802 7.198 14.2M 224 0.875 bilinear
densenet201 76.886 23.114 93.372 6.628 20.0M 224 0.875 bilinear
densenet161 77.112 22.888 93.556 6.444 28.7M 224 0.875 bilinear
googlenet 69.772 30.228 89.534 10.466 6.6M 224 0.875 bilinear
inception_v3 77.508 22.492 93.656 6.344 27.2M 299 1.000 bilinear
shufflenet_v2_x0_5 60.552 39.448 81.736 18.264 1.4M 224 0.875 bilinear
shufflenet_v2_x1_0 69.356 30.644 88.314 11.686 2.3M 224 0.875 bilinear
mobilenet_v2 71.870 28.130 90.294 9.706 3.5M 224 0.875 bilinear
mobilenet_v3_small 67.668 32.332 87.408 12.592 2.5M 224 0.875 bilinear
mobilenet_v3_large 74.054 25.946 91.340 8.660 5.5M 224 0.875 bilinear
resnext50_32x4d 77.608 22.392 93.688 6.312 88.8M 224 0.875 bilinear
resnext101_32x8d 79.308 20.692 94.530 5.470 25.0M 224 0.875 bilinear
wide_resnet50_2 78.480 21.520 94.084 5.916 68.9M 224 0.875 bilinear
wide_resnet101_2 78.842 21.158 94.280 5.780 126.9M 224 0.875 bilinear
senet154 81.324 18.676 95.508 4.492 115.1M 224 0.875 bilinear
se_resent50 77.642 22.358 93.748 6.252 28.1M 224 0.875 bilinear
se_resnet101 78.390 21.610 94.254 5.746 49.3M 224 0.875 bilinear
se_resnet152 78.666 21.334 94.380 5.620 66.8M 224 0.875 bilinear
se_resnext50_32x4d 79.080 20.920 94.432 5.568 27.6M 224 0.875 bilinear
se_resnext101_32x4d 80.236 19.764 95.034 4.966 49.0M 224 0.875 bilinear
res2net50_14w_8s 78.152 21.848 93.842 6.158 25.1M 224 0.875 bilinear
res2net50_26w_4s 77.946 22.054 93.852 6.148 25.7M 224 0.875 bilinear
res2net50_26w_6s 78.574 21.426 94.126 5.874 37.1M 224 0.875 bilinear
res2net50_26w_8s 79.210 20.790 94.362 5.638 48.4M 224 0.875 bilinear
res2net50_48w_2s 77.514 22.486 93.548 6.452 25.3M 224 0.875 bilinear
res2net101_26w_4s 79.196 20.804 94.440 5.560 45.2M 224 0.875 bilinear
mnasnet0_5 67.734 32.266 87.490 12.510 2.2M 224 0.875 biilnear
mnasnet1_0 73.456 26.544 91.510 8.490 4.4M 224 0.875 biilnear
ghostnet 73.982 26.018 91.462 8.538 5.2M 224 0.875 biilnear
resnest50 80.958 19.042 95.382 4.618 27.5M 224 0.875 bilinear
resnest101 82.890 17.110 96.324 3.676 48.3M 256 0.875 bilinear
resnest200 83.850 16.150 96.882 3.118 70.2M 320 0.909 bicubic
resnest269 84.510 15.490 96.992 3.008 110.9M 416 0.928 bicubic
efficientnet_b0 77.692 22.308 93.532 6.468 5.3M 224 0.875 bicubic
efficientnet_b1 78.792 21.208 94.342 5.658 7.8M 256 1.000 bicubic
efficientnet_b2 80.608 19.392 95.310 4.690 9.1M 288 1.000 bicubic
efficientnet_b3 82.240 17.760 96.116 3.884 12.2M 320 1.000 bicubic
efficientnet_b4 83.418 16.582 96.598 3.402 19.3M 384 1.000 bicubic
efficientnet_b5 83.444 16.556 96.628 3.372 30.4M 456 1.000 bicubic
efficientnet_b6 84.008 15.992 96.916 3.084 43.0M 528 1.000 bicubic
efficientnet_b7 84.112 15.757 96.908 3.092 66.3M 600 1.000 bicubic
regnet_y_400mf 73.084 26.916 91.294 8.706 4.3M 224 0.875 bilinear
regnet_y_800mf 75.734 24.266 92.710 7.290 6.4M 224 0.875 bilinear
regnet_y_1_6gf 77.276 22.724 93.574 6.426 11.2M 224 0.875 bilinear
regnet_y_3_2gf 78.334 21.666 94.196 5.804 19.4M 224 0.875 bilinear
regnet_y_8gf 79.342 20.658 94.740 5.260 39.4M 224 0.875 bilinear
regnet_y_16gf 80.056 19.944 95.044 4.956 83.6M 224 0.875 bilinear
regnet_y_32gf 80.604 19.396 95.160 4.840 145.0M 224 0.875 bilinear
regnet_x_400mf 72.544 27.456 90.860 9.140 5.5M 224 0.875 bilinear
regnet_x_800mf 74.884 25.116 92.428 7.572 7.3M 224 0.875 bilinear
regnet_x_1_6gf 76.338 23.662 93.050 6.950 9.2M 224 0.875 bilinear
regnet_x_3_2gf 78.136 21.864 93.920 6.080 15.3M 224 0.875 bilinear
regnet_x_8gf 79.240 20.760 94.548 5.452 39.6M 224 0.875 bilinear
regnet_x_16gf 79.898 20.102 94.858 5.142 54.3M 224 0.875 bilinear
regnet_x_32gf 80.226 19.774 95.086 4.914 107.8M 224 0.875 bilinear
rexnetv1_1_0 77.860 22.140 93.880 6.120 4.8M 224 0.875 bicubic
rexnetv1_1_3 79.492 20.508 94.674 5.326 7.6M 224 0.875 bicubic
rexnetv1_1_5 80.314 19.686 95.162 4.838 9.7M 224 0.875 bicubic
rexnetv1_2_0 81.626 18.374 95.666 4.334 16.4M 224 0.875 bicubic
rexnetv1_3_0 82.792 17.208 96.242 3.758 34.7M 224 0.875 bicubic
rexnet_lite_1_0 76.176 23.824 92.824 7.176 4.8M 224 0.875 bicubic
rexnet_lite_1_3 77.726 22.274 93.788 6.212 6.7M 224 0.875 bicubic
rexnet_lite_1_5 78.588 21.412 94.192 5.828 8.3M 224 0.875 bicubic
rexnet_lite_2_0 80.134 19.866 95.006 4.994 13.0M 224 0.875 bicubic
vit_tiny_patch16_224 75.458 24.542 92.850 7.150 5.7M 224 0.900 bicubic
vit_tiny_patch16_384 78.440 21.560 94.546 5.454 5.7M 384 1.000 bicubic
vit_small_patch32_224 75.990 24.010 93.274 6.726 22.9M 224 0.900 bicubic
vit_small_patch32_384 80.484 19.516 95.600 4.400 22.9M 384 1.000 bicubic
vit_small_patch16_224 81.380 18.620 96.136 3.864 22.1M 224 0.900 bicubic
vit_small_patch16_384 83.790 16.210 97.106 2.894 22.2M 384 1.000 bicubic
vit_base_patch32_224 80.720 19.280 95.566 4.434 88.2M 224 0.900 bicubic
vit_base_patch32_384 83.350 16.650 96.842 3.158 88.3M 384 1.000 bicubic
vit_base_patch16_224 84.528 15.472 97.294 2.706 86.6M 224 0.900 bicubic
vit_base_patch16_384 85.992 14.008 98.002 1.998 86.9M 384 1.000 bicubic
vit_base_patch8_224 85.800 14.200 97.790 2.210 86.6M 224 0.900 bicubic
vit_large_patch16_224 85.840 14.160 97.818 2.182 304.3M 224 0.900 bicubic
vit_large_patch16_384 87.084 12.916 98.302 1.698 304.7M 384 1.000 bicubic
vit_large_patch32_384 81.512 18.488 96.090 3.910 306.6M 384 1.000 bicubic
vit_base_patch16_224_miil 84.266 15.734 96.804 3.196 86.6M 224 0.875 bilinear
vit_base_patch16_224_sam 80.240 19.760 94.758 5.242 86.5M 224 0.900 bicubic
vit_base_patch32_224_sam 73.696 26.304 91.016 8.984 88.2M 224 0.900 bicubic
deit_tiny_patch16_224 72.176 27.824 91.114 8.886 5.7M 224 0.900 bicubic
deit_small_patch16_224 79.852 20.148 95.044 4.956 22.1M 224 0.900 bicubic
deit_base_patch16_224 81.990 18.010 95.736 4.264 86.6M 224 0.900 bicubic
deit_base_patch16_384 83.102 16.898 96.368 3.632 86.9M 384 1.000 bicubic
deit_tiny_distilled_patch16_224 74.504 25.496 91.890 8.110 5.9M 224 0.900 bicubic
deit_small_distilled_patch16_224 81.216 18.784 95.386 4.614 22.4M 224 0.900 bicubic
deit_base_distilled_patch16_224 83.386 16.614 96.488 3.512 87.3M 224 0.900 bicubic
deit_base_distilled_patch16_384 85.426 14.574 97.332 2.668 87.6M 384 1.000 bicubic
pvt_tiny 75.096 24.904 92.420 7.580 13.2M 224 0.875 bicubic
pvt_small 79.762 20.238 94.942 5.058 24.5M 224 0.875 bicubic
pvt_medium 81.196 18.804 95.642 4.358 44.2M 224 0.875 bicubic
pvt_large 81.694 18.306 95.852 4.148 61.4M 224 0.875 bicubic
swin_tiny_patch4_window7_224 81.378 18.622 95.540 4.460 28.3M 224 0.900 bicubic
swin_small_patch4_window7_224 83.226 16.774 96.320 3.680 49.6M 224 0.900 bicubic
swin_base_patch4_window7_224 83.602 16.398 96.448 3.552 87.8M 224 0.900 bicubic
swin_base_patch4_window12_384 84.486 15.514 96.892 3.108 87.9M 384 1.000 bicubic
swin_base_patch4_window7_224_in22k_to_1k 85.274 14.726 97.568 2.432 87.8M 224 0.900 bicubic
swin_base_patch4_window12_384_in22k_to_1k 86.434 13.566 98.078 1.922 87.9M 384 1.000 bicubic
swin_large_patch4_window7_224_in22k_to_1k 86.318 13.682 97.898 2.102 196.5M 224 0.900 bicubic
swin_large_patch4_window12_384_in22k_to_1k 87.134 12.866 98.232 1.768 196.7M 384 1.000 bicubic
cswin_tiny_224 82.812 17.188 96.300 3.700 22.3M 224 0.900 bicubic
cswin_small_224 83.596 16.404 96.584 3.416 34.6M 224 0.900 bicubic
cswin_base_224 84.228 15.772 96.912 3.088 77.4M 224 0.900 bicubic
cswin_base_384 85.510 14.490 97.484 2.516 77.4M 384 1.000 bicubic
cswin_large_224 86.522 13.478 97.992 2.008 173.3M 224 0.900 bicubic
cswin_large_384 87.486 12.514 98.346 1.654 173.3M 384 1.000 bicubic
crossformer_tiny_patch4_group7_224 81.542 18.458 95.582 4.418 27.8M 224 0.900 bicubic
crossformer_small_patch4_group7_224 82.472 17.528 96.086 3.914 30.7M 224 0.900 bicubic
crossformer_base_patch4_group7_224 83.520 16.480 96.540 3.460 52.0M 224 0.900 bicubic
crossformer_large_patch4_group7_224 83.888 16.112 96.558 3.442 92.0M 224 0.900 bicubic
mlp_mixer_b16_224 76.598 23.402 92.228 7.772 59.9M 224 0.875 bicubic
mlp_mixer_l16_224 72.060 27.940 87.668 12.332 208.2M 224 0.875 bicubic
mlp_mixer_b16_224_miil 82.308 17.692 95.720 4.280 59.9M 224 0.875 bilinear
gmlp_s16_224 79.644 20.356 94.626 5.374 19.4M 224 0.875 bicubic
resmlp_12_224 76.650 23.350 93.182 6.818 15.4M 224 0.875 bicubic
resmlp_12_distilled_224 77.954 22.046 93.560 6.440 15.4M 224 0.875 bicubic
resmlp_24_224 79.378 20.622 94.546 5.454 30.0M 224 0.875 bicubic
resmlp_24_distilled_224 80.760 19.240 95.220 4.780 30.0M 224 0.875 bicubic
resmlp_36_224 79.772 20.228 94.886 5.114 44.7M 224 0.875 bicubic
resmlp_36_distilled_224 81.150 18.850 95.478 4.522 44.7M 224 0.875 bicubic
resmlp_big_24_224 81.038 18.962 95.020 4.980 129.1M 224 0.875 bicubic
resmlp_big_24_224_in22k_to_1k 84.396 15.604 97.112 2.888 129.1M 224 0.875 bicubic
resmlp_big_24_distilled_224 83.590 16.410 96.650 3.350 129.1M 224 0.875 bicubic
convmixer_768_32_relu 80.076 19.924 94.992 5.008 21.1M 224 0.875 bilinear
convmixer_1024_20 77.012 22.988 93.384 6.616 24.4M 224 0.875 bilinear
convmixer_1536_20 81.056 18.944 95.620 4.380 51.6M 224 0.875 bilinear
convnext_tiny_224 82.066 17.934 95.856 4.144 28.6M 224 0.875 bicubic
convnext_small_224 83.146 16.854 96.432 3.568 50.2M 224 0.875 bicubic
convnext_base_224 83.828 16.172 96.746 3.254 88.6M 224 0.875 bicubic
convnext_base_384 84.896 15.104 97.272 2.728 88.6M 384 1.000 bicubic
convnext_large_224 84.292 15.708 96.894 3.106 197.8M 224 1.000 bicubic
convnext_large_384 84.162 15.838 96.968 3.032 197.8M 384 1.000 bicubic
convnext_base_224_22k_to_1k 85.820 14.180 97.868 2.132 88.6M 224 0.875 bicubic
convnext_base_384_22k_to_1k 86.558 13.442 98.194 1.806 88.6M 384 1.000 bicubic
convnext_large_224_22k_to_1k 86.636 13.364 98.032 1.968 197.8M 224 0.875 bicubic
convnext_large_384_22k_to_1k 87.390 12.610 98.364 1.636 197.8M 384 1.000 bicubic
convnext_xlarge_224_22k_to_1k 87.006 12.994 98.204 1.796 350.2M 224 0.875 bicubic
convnext_xlarge_384_22k_to_1k 87.548 12.452 98.486 1.514 350.2M 384 1.000 bicubic
convnext_iso_small_224 79.742 20.258 94.798 5.208 22.3M 224 0.875 bicubic
convnext_iso_base_224 81.818 18.182 95.674 4.326 87.1M 224 0.875 bicubic
convnext_iso_large_224 82.576 17.424 95.956 4.044 306.0M 224 0.875 bicubic
poolformer_s12 77.184 22.816 93.476 6.524 12.0M 224 0.950 bicubic
poolformer_s24 80.362 19.638 95.116 4.884 21.0M 224 0.950 bicubic
poolformer_s36 81.356 18.644 95.422 4.578 31.0M 224 0.950 bicubic
poolformer_m36 82.110 17.890 95.690 4.310 56.0M 224 0.950 bicubic
poolformer_m48 82.456 17.544 95.962 4.038 73.0M 224 0.950 bicubic
uniformer_small 82.958 17.042 96.290 3.710 22.0M 224 0.900 bicubic
uniformer_small_plus 82.938 17.062 96.406 3.594 50.0M 224 0.900 bicubic
uniformer_base 83.814 16.186 96.738 3.262 50.0M 224 0.900 bicubic
uniformer_base_ls 83.952 16.048 96.728 3.272 50.0M 224 0.900 bicubic
levit-128s 76.518 23.482 92.904 7.096 7.8M 224 0.875 bicubic
levit-128 78.586 21.414 93.944 6.056 9.2M 224 0.875 bicubic
levit-192 79.874 20.126 94.750 5.250 10.9M 224 0.875 bicubic
levit-256 81.594 18.406 95.456 4.544 18.9M 224 0.875 bicubic
levit-384 82.594 17.406 95.956 4.044 39.1M 224 0.875 bicubic
regionvit_tiny_224 80.100 19.900 95.174 4.826 13.8M 224 0.875 bicubic
regionvit_small_224 82.558 17.442 96.112 3.888 30.6M 224 0.875 bicubic
regionvit_medium_224 83.002 16.998 96.220 3.780 41.2M 224 0.875 bicubic
regionvit_base_224 83.116 16.884 96.096 3.904 72.7M 224 0.875 bicubic
van_tiny 75.408 24.592 93.030 6.970 4.1M 224 0.9 bicubic
van_small 81.012 18.988 95.626 4.374 13.9M 224 0.9 bicubic
van_base 82.806 17.194 96.206 3.794 26.6M 224 0.9 bicubic
van_large 83.860 16.140 96.728 3.272 44.8M 224 0.9 bicubic
deit_small_patch16_LS_224 81.306 18.694 95.356 4.644 22.1M 224 0.875 bicubic
deit_small_patch16_LS_384 83.274 16.726 96.574 3.426 22.2M 384 0.875 bicubic
deit_small_patch16_LS_224_in21k 82.764 17.236 96.700 3.300 22.1M 224 0.875 bicubic
deit_small_patch16_LS_384_in21k 84.458 15.542 97.240 2.760 22.2M 384 0.875 bicubic
deit_base_patch16_LS_224 83.686 16.314 96.542 3.458 86.6M 224 0.875 bicubic
deit_base_patch16_LS_384 84.740 15.260 97.108 2.892 86.9M 384 0.875 bicubic
deit_base_patch16_LS_224_in21k 85.482 14.518 97.568 2.432 86.6M 224 0.875 bicubic
deit_base_patch16_LS_384_in21k 86.354 13.646 97.946 2.054 86.9M 384 0.875 bicubic
deit_large_patch16_LS_224 84.628 15.372 96.946 3.054 304.4M 224 0.875 bicubic
deit_large_patch16_LS_384 85.602 14.398 97.510 2.490 304.8M 384 0.875 bicubic
deit_large_patch16_LS_224_in21k 86.812 13.188 98.122 1.878 304.4M 224 0.875 bicubic
deit_large_patch16_LS_384_in21k 87.440 12.560 98.346 1.654 304.8M 384 0.875 bicubic
deit_huge_patch14_LS_224 85.158 14.842 97.248 2.752 632.1M 224 0.875 bicubic
deit_huge_patch14_LS_224_in21k 86.890 13.110 98.140 1.860 632.1M 224 0.875 bicubic
cait_XS24_384 84.060 15.940 96.886 3.114 26.7M 384 1 bicubic
cait_S24_224 83.452 16.548 96.572 3.428 46.9M 224 1 bicubic
cait_S24_384 85.048 14.952 97.344 2.656 47.1M 384 1 bicubic
cait_S36_384 86.058 14.546 97.480 2.520 68.4M 384 1 bicubic
cait_M36_384 84.060 13.942 97.732 2.268 271.2M 384 1 bicubic
cait_M48_448 86.492 13.508 97.752 2.248 356.5M 384 1 bicubic
dla34 74.470 25.530 92.010 7.990 15.8M 224 0.875 bicubic
dla46_c 64.482 35.518 86.088 13.912 1.3M 224 0.875 bicubic
dla46x_c 65.642 34.358 86.658 13.342 1.1M 224 0.875 bicubic
dla60x_c 67.546 32.454 88.202 11.798 1.3M 224 0.875 bicubic
dla60 76.824 23.176 93.176 6.824 22.3M 224 0.875 bicubic
dla60x 78.074 21.926 93.948 6.052 17.6M 224 0.875 bicubic
dla102 77.706 22.294 93.776 6.224 33.7M 224 0.875 bicubic
dla102x 78.272 21.728 94.102 5.898 26.8M 224 0.875 bicubic
dla102x2 79.362 20.638 94.634 5.366 41.7M 224 0.875 bicubic
dla169 78.528 21.472 94.276 5.724 54.0M 224 0.875 bicubic
genet_small 75.342 24.658 92.238 7.762 8.2M 192 0.875 bicubic
genet_normal 79.616 20.384 94.512 5.488 21.1M 192 0.875 bicubic
genet_large 81.300 18.700 95.386 4.614 31.1M 256 0.875 bicubic
hrnet_w18_small 72.078 27.922 90.470 9.530 13.2M 224 0.875 bicubic
hrnet_w18_small_v2 74.816 25.184 92.264 7.736 15.6M 224 0.875 bicubic
hrnet_w18 76.484 23.516 93.322 6.678 21.3M 224 0.875 bicubic
hrnet_w30 78.074 21.926 94.136 5.864 37.7M 224 0.875 bicubic
hrnet_w32 78.188 21.812 94.034 5.966 41.2M 224 0.875 bicubic
hrnet_w40 78.652 21.348 94.378 5.622 57.6M 224 0.875 bicubic
hrnet_w44 78.646 21.354 94.250 5.750 67.1M 224 0.875 bicubic
hrnet_w48 78.972 21.028 94.344 5.656 77.5M 224 0.875 bicubic
hrnet_w64 79.210 20.790 94.558 5.442 128.1M 224 0.875 bicubic
fan_tiny_12_p16_224 79.098 20.902 94.594 5.406 7.3M 224 0.875 bicubic
fan_small_12_p16_224 82.454 17.546 96.220 3.780 28.3M 224 0.875 bicubic
fan_base_18_p16_224 83.466 16.534 96.516 3.484 54.4M 224 0.875 bicubic
fan_tiny_8_p4_hybrid 80.064 19.936 95.042 4.958 7.5M 224 0.875 bicubic
fan_small_12_p4_hybrid 83.504 16.496 96.564 3.436 26.1M 224 0.875 bicubic
fan_base_16_p4_hybrid 83.828 16.172 96.686 3.314 50.5M 224 0.875 bicubic
fan_base_16_p4_hybrid_in22k_1k 85.474 14.526 97.524 2.476 50.5M 224 0.875 bicubic
fan_base_16_p4_hybrid_in22k_1k_384 85.952 14.048 97.790 2.210 50.5M 384 0.875 bicubic
fan_large_16_p4_hybrid_in22k_1k 86.390 13.610 97.942 2.058 76.9M 224 0.875 bicubic
fan_large_16_p4_hybrid_in22k_1k_384 86.718 13.282 98.040 1.960 76.9M 384 0.875 bicubic

Spetial Normalized Case

  • vit, mlp-mixer, gmlp based model: test images normalized using mean=(0.5, 0.5, 0.5) and std=(0.5, 0.5, 0.5)
  • vit_base_patch16_224_miil: test images normalized using mean=(0, 0, 0) and std=(1, 1, 1)