osmr / imgclsmob » Build #15

Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`
Paper not specified for task=`Image Classification` on dataset=`ImageNet`

Build Results

TOP 1 ACCURACY TOP 5 ACCURACY
SPEED
MODEL CODE PAPER
ε-REPR
CODE PAPER
ε-REPR
PAPER
108-MENet-8x1
(g=3)
56.1% -- 79.2% --
1.0-SqNxt-23
57.5% -- 80.9% --
1.0-SqNxt-23v5
59.2% -- 82.2% --
128-MENet-8x1
(g=4)
57.6% -- 80.4% --
1.5-SqNxt-23
65.1% -- 86.5% --
1.5-SqNxt-23v5
66.2% -- 87.0% --
160-MENet-8x1
(g=8)
56.2% -- 79.2% --
2.0-SqNxt-23
69.4% -- 89.0% --
2.0-SqNxt-23v5
70.4% -- 89.3% --
228-MENet-12x1
(g=3)
65.9% -- 86.8% --
256-MENet-12x1
(g=4)
67.3% -- 87.5% --
348-MENet-12x1
(g=3)
71.8% -- 90.4% --
352-MENet-12x1
(g=8)
68.4% -- 88.0% --
456-MENet-24x1
(g=3)
74.7% -- 92.0% --
AlexNet
59.0% -- 81.8% -- 417.5
AlexNet-b
58.4% -- 81.0% -- 418.1
BAM-ResNet-50
76.9% -- 93.4% --
BN-Inception
74.6% -- 92.3% -- 426.1
BN-VGG-11
71.0% -- 90.4% -- 420.9
BN-VGG-11b
70.4% -- 90.0% -- 415.6
BN-VGG-13
72.2% -- 90.9% -- 406.0
BN-VGG-13b
71.6% -- 90.4% -- 394.5
BN-VGG-16
74.3% -- 92.2% -- 400.7
BN-VGG-16b
72.8% -- 91.3% -- 394.4
BN-VGG-19
75.9% -- 92.9% -- 395.8
BN-VGG-19b
73.9% -- 91.6% -- 385.3
CBAM-ResNet-50
77.6% -- 93.9% --
CondenseNet-74
(C=G=4)
73.8% -- 91.7% --
CondenseNet-74
(C=G=8)
71.1% -- 89.9% --
DarkNet-53
78.3% -- 94.4% --
DARTS
73.3% -- 91.3% --
DenseNet-121
76.5% 76.4% 93.0% 93.3%
DenseNet-161
77.1% -- 93.6% --
DenseNet-169
77.6% 77.9% 93.7% 94.1%
DenseNet-201
76.9% 78.5% 93.4% 94.5%
DiracNetV2-18
68.5% -- 88.3% --
DiracNetV2-34
71.2% -- 90.1% --
DLA-102
78.0% -- 94.0% --
DLA-169
78.7% -- 94.3% --
DLA-34
74.6% -- 92.1% --
DLA-46-C
65.7% -- 86.8% --
DLA-60
77.0% -- 93.3% --
DLA-X-102
78.5% -- 94.2% --
DLA-X2-102
79.5% -- 94.6% --
DLA-X-46-C
66.7% -- 87.3% --
DLA-X-60
78.2% -- 94.0% --
DLA-X-60-C
69.0% -- 89.1% --
DPN-131
79.5% -- 94.5% --
DPN-68
76.8% -- 93.2% --
DPN-98
79.2% -- 94.5% --
DRN-C-26
75.1% -- 92.5% --
DRN-C-42
77.1% -- 93.4% --
DRN-C-58
78.3% -- 94.0% --
DRN-D-105
79.4% -- 94.5% --
DRN-D-22
74.2% -- 91.8% --
DRN-D-38
76.2% -- 93.1% --
DRN-D-54
78.8% -- 94.1% --
EfficientNet-B0
75.2% 76.3% 92.5% 93.2%
EfficientNet-B0b
76.1% -- 93.0% --
EfficientNet-B1
77.0% 78.8% 93.6% 94.4%
EfficientNet-B1b
78.4% -- 94.1% --
EfficientNet-B2b
79.7% -- 94.7% -- 368.5
EfficientNet-B3b
81.4% -- 95.6% -- 328.4
EfficientNet-B4b
82.8% -- 96.2% -- 221.8
EfficientNet-B5b
83.6% -- 96.7% -- 115.4
EfficientNet-B6b
84.1% -- 96.9% -- 68.4
EfficientNet-B7b
84.3% -- 96.9% -- 39.5
ESPNetv2 x0.5
57.7% -- 79.8% --
ESPNetv2 x1.0
66.1% -- 86.5% --
ESPNetv2 x1.25
67.9% -- 87.8% --
ESPNetv2 x1.5
69.2% -- 88.7% --
ESPNetv2 x2.0
72.1% -- 90.4% --
FBNet-Cb
75.1% -- 92.4% -- 407.7
FD-MobileNet x0.25
44.1% -- 69.0% --
FD-MobileNet x0.5
56.9% -- 79.8% --
FD-MobileNet x0.75
61.6% -- 83.6% --
FD-MobileNet x1.0
65.8% -- 86.6% --
IBN(b)-ResNet-50
76.4% -- 93.1% --
IBN-DenseNet-121
75.5% -- 92.8% --
IBN-DenseNet-169
76.8% -- 93.5% --
IBN-ResNet-101
78.7% -- 94.4% --
IBN-ResNet-50
77.2% -- 93.6% --
IBN-ResNeXt-101
(32x4d)
79.1% -- 94.6% --
IGCV3 x0.25
46.3% -- 71.3% --
IGCV3 x0.5
60.2% -- 82.7% --
IGCV3 x0.75
68.9% -- 88.6% --
IGCV3 x1.0
72.1% -- 90.8% --
InceptionResNetV
80.1% -- 95.1% --
InceptionV3
78.9% -- 94.4% --
InceptionV4
79.4% -- 94.7% --
i-RevNet-301
74.0% -- 91.6% --
MixNet-L
78.9% 78.9% 94.2% 94.2% 400.3
MixNet-M
77.1% 77.0% 93.4% 93.3% 404.0
MixNet-S
76.0% 75.8% 92.8% 92.8% 413.1
MnasNet
68.4% -- 88.3% --
MobileNetV2 x0.25
51.7% -- 75.5% --
MobileNetV2 x0.5
64.0% -- 85.1% --
MobileNetV2 x0.75
69.8% -- 89.2% --
MobileNetV2 x1.0
73.0% -- 91.1% --
MobileNetV3 L/224/1.0
75.1% -- 92.2% --
MobileNet x0.25
53.7% -- 77.5% --
MobileNet x0.5
65.8% -- 86.5% --
MobileNet x0.75
69.9% -- 89.2% --
MobileNet x1.0
73.4% -- 91.0% --
74.3% -- 91.8% --
81.9% -- 95.8% --
PeleeNet
68.2% -- 88.5% --
PNASNet-5-Large
82.1% -- 95.7% --
PolyNet
80.9% 81.3% 95.5% 95.8%
PrepResNet-10
64.9% -- 85.8% --
PreResNet-101
78.3% -- 94.1% --
PreResNet-101b
79.0% -- 94.4% --
PreResNet-12
66.1% -- 86.5% --
PreResNet-14
67.4% -- 87.6% --
PreResNet-152
79.1% -- 94.5% --
PreResNet-152b
79.9% -- 94.8% --
PreResNet-16
69.5% -- 88.9% --
PreResNet-18
71.6% -- 90.3% --
PreResNet-200b
78.7% -- 94.1% --
PreResNet-26
73.7% -- 91.5% --
PreResNet-269b
79.1% -- 94.2% --
PreResNet-34
75.1% -- 92.3% --
PreResNet-50
77.6% -- 93.5% --
PreResNet-50b
77.5% -- 93.4% --
PreResNet-BC-26b
74.5% -- 92.0% --
PreResNet-BC-38b
77.1% -- 93.4% --
ProxylessNAS CPU
75.3% -- 92.4% --
ProxylessNAS GPU
75.2% -- 92.6% --
ProxylessNAS Mob-14
76.7% -- 93.4% --
ProxylessNAS Mobile
74.6% -- 92.2% --
PyramidNet-101
(a=360)
78.0% -- 93.8% --
ResNet-10
65.3% -- 85.6% --
ResNet-101
78.1% 78.2% 93.8% 94.0%
ResNet-101b
79.4% -- 94.7% --
ResNet-12
66.4% -- 86.7% --
ResNet-14
67.5% -- 87.5% --
ResNet-152
79.0% 78.6%
94.5% 94.3%
ResNet-152b
80.1% -- 95.0% --
ResNet-16
69.5% -- 88.8% --
ResNet-18
71.5% 72.1% 90.2% -- 412.9
ResNet-26
73.7% -- 91.5% --
ResNet-34
75.2% -- 92.2% --
ResNet-50
77.7% 77.1%
93.7% 93.3%
ResNet-50b
77.6% -- 93.6% --
ResNet-BC-26b
74.9% -- 92.0% --
ResNet-BC-38b
76.3% -- 93.0% --
ResNeXt-101
(32x4d)
78.2% -- 93.9% --
ResNeXt-101
(64x4d)
79.0% -- 94.3% --
ResNeXt-14
(16x4d)
68.1% -- 87.5% --
ResNeXt-14
(32x2d)
67.4% -- 87.2% --
ResNeXt-14
(32x4d)
69.7% -- 88.5% --
ResNeXt-26
(32x2d)
73.4% -- 91.1% --
ResNeXt-26
(32x4d)
75.9% -- 92.5% --
SENet-154
81.4% 82.7% 95.4% 96.2%
SENet-16
74.4% -- 91.8% --
SENet-28
78.1% -- 94.0% --
SE-PreResNet-10
66.0% -- 86.6% --
SE-PreResNet-18
71.9% -- 90.4% --
SE-PreResNet-BC-26b
76.8% -- 93.4% --
SE-PreResNet-BC-38b
78.4% -- 94.2% --
SE-ResNet-10
66.1% -- 86.3% --
SE-ResNet-101
78.1% -- 94.1% --
SE-ResNet-152
78.5% -- 94.2% --
SE-ResNet-18
71.8% -- 90.4% --
SE-ResNet-26
74.3% -- 91.8% --
SE-ResNet-50
77.5% -- 93.6% --
SE-ResNet-50b
79.2% -- 94.6% --
SE-ResNet-BC-26b
76.4% -- 93.0% --
SE-ResNet-BC-38b
78.4% -- 94.1% --
SE-ResNeXt-101
(32x4d)
80.0% -- 94.9% --
SE-ResNeXt-50
(32x4d)
79.0% -- 94.5% --
ShuffleNetV2b x0.5
59.7% -- 81.8% --
ShuffleNetV2b x1.0
69.4% -- 88.8% --
ShuffleNetV2b x1.5
72.7% -- 90.9% --
ShuffleNetV2b x2.0
74.4% -- 91.7% --
ShuffleNetV2 x0.5
59.0% -- 81.4% --
ShuffleNetV2 x1.0
68.6% -- 88.4% --
ShuffleNetV2 x1.5
72.5% -- 90.6% --
ShuffleNetV2 x2.0
74.1% -- 91.6% --
ShuffleNet x0.25
(g=1)
37.6% -- 62.7% --
ShuffleNet x0.25
(g=3)
38.3% -- 63.5% --
ShuffleNet x0.5
(g=1)
53.4% -- 77.4% --
ShuffleNet x0.5
(g=3)
55.8% -- 79.2% --
ShuffleNet x0.75
(g=1)
60.4% -- 82.9% --
ShuffleNet x0.75
(g=3)
61.8% -- 83.5% --
ShuffleNet x1.0
(g=1)
65.1% -- 86.1% --
ShuffleNet x1.0
(g=2)
65.8% -- 86.4% --
ShuffleNet x1.0
(g=3)
65.6% -- 86.5% --
ShuffleNet x1.0
(g=4)
65.8% -- 86.6% --
ShuffleNet x1.0
(g=8)
65.9% -- 86.6% --
SPNASNet
74.1% -- 91.8% -- 412.2
SqueezeNet v1.0
60.7% -- 82.3% --
SqueezeNet v1.1
60.7% -- 82.3% --
SqueezeResNet v1.0
60.2% -- 81.9% --
SqueezeResNet v1.1
59.9% -- 81.8% --
VGG-11
70.1% 70.4% 89.6% 89.6% 411.2
VGG-13
71.2% 71.3% 90.2% 90.1% 406.1
VGG-16
73.0% 74.4% 91.3% 91.9% 396.4
VGG-19
73.8% 74.5% 91.6% 92.0% 390.5
WRN-50-2
77.5% -- 93.6% --
Xception
79.0% 79.0% 94.5% 94.5%
ZFNet
60.2% 64.0% 82.7% 85.3% 428.2
ZFNet-b
63.6% -- 85.1% -- 426.3