
=== Start adding workers ===
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=0,shuffle=True)'}
=> Add worker SGDMWorker(index=0, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=1,shuffle=True)'}
=> Add worker SGDMWorker(index=1, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=2,shuffle=True)'}
=> Add worker SGDMWorker(index=2, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=3,shuffle=True)'}
=> Add worker SGDMWorker(index=3, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=4,shuffle=True)'}
=> Add worker SGDMWorker(index=4, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=5,shuffle=True)'}
=> Add worker SGDMWorker(index=5, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=6,shuffle=True)'}
=> Add worker SGDMWorker(index=6, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=7,shuffle=True)'}
=> Add worker SGDMWorker(index=7, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=8,shuffle=True)'}
=> Add worker SGDMWorker(index=8, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=9,shuffle=True)'}
=> Add worker SGDMWorker(index=9, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=10,shuffle=True)'}
=> Add worker SGDMWorker(index=10, momentum=0.9)
{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': True, 'download': True, 'batch_size': 64, 'shuffle': None, 'sampler': 'DistributedSampler(num_replicas=12,rank=11,shuffle=True)'}
=> Add worker ByzantineWorker(index=11)

=== Start adding graph ===
TwoCliquesWithByzantine(m=5,b=1)

{'Type': 'Setup', 'Dataset': 'cifar10', 'data_dir': '/home/he/dero/datasets/', 'train': False, 'download': True, 'batch_size': 128, 'shuffle': False, 'sampler': None}
Train epoch 1
[E 1B0  |    768/50000 (  2%) ] Loss: 2.3068 top1=  9.3750

=== Peeking data label distribution E1B0 ===
Worker 0 has targets: tensor([0, 0, 1, 3, 0], device='cuda:0')
Worker 1 has targets: tensor([1, 1, 9, 2, 9], device='cuda:0')
Worker 2 has targets: tensor([2, 2, 4, 5, 9], device='cuda:0')
Worker 3 has targets: tensor([3, 3, 1, 2, 3], device='cuda:0')
Worker 4 has targets: tensor([4, 3, 3, 2, 4], device='cuda:0')
Worker 5 has targets: tensor([4, 4, 9, 8, 1], device='cuda:0')
Worker 6 has targets: tensor([5, 5, 4, 1, 1], device='cuda:0')
Worker 7 has targets: tensor([6, 6, 7, 0, 0], device='cuda:0')
Worker 8 has targets: tensor([7, 7, 0, 2, 1], device='cuda:0')
Worker 9 has targets: tensor([8, 8, 0, 3, 1], device='cuda:0')
Worker 10 has targets: tensor([8, 8, 3, 9, 7], device='cuda:0')
Worker 11 has targets: tensor([9, 9, 2, 4, 1], device='cuda:0')



=== Log global consensus distance @ E1B0 ===
consensus_distance=0.000



=== Log clique consensus distance @ E1B0 ===
clique1_consensus_distance=0.000
clique2_consensus_distance=0.000



=== Log mixing matrix @ E1B0 ===
[[0.167 0.167 0.167 0.167 0.167 0.    0.    0.    0.    0.    0.167 0.   ]
 [0.167 0.233 0.2   0.2   0.2   0.    0.    0.    0.    0.    0.    0.   ]
 [0.167 0.2   0.233 0.2   0.2   0.    0.    0.    0.    0.    0.    0.   ]
 [0.167 0.2   0.2   0.233 0.2   0.    0.    0.    0.    0.    0.    0.   ]
 [0.167 0.2   0.2   0.2   0.233 0.    0.    0.    0.    0.    0.    0.   ]
 [0.    0.    0.    0.    0.    0.233 0.2   0.2   0.2   0.167 0.    0.   ]
 [0.    0.    0.    0.    0.    0.2   0.233 0.2   0.2   0.167 0.    0.   ]
 [0.    0.    0.    0.    0.    0.2   0.2   0.233 0.2   0.167 0.    0.   ]
 [0.    0.    0.    0.    0.    0.2   0.2   0.2   0.233 0.167 0.    0.   ]
 [0.    0.    0.    0.    0.    0.167 0.167 0.167 0.167 0.167 0.167 0.   ]
 [0.167 0.    0.    0.    0.    0.    0.    0.    0.    0.167 0.642 0.025]
 [0.    0.    0.    0.    0.    0.    0.    0.    0.    0.    0.025 0.975]]


[E 1B10 |   8448/50000 ( 17%) ] Loss: 2.2873 top1= 18.6080

=== Log global consensus distance @ E1B10 ===
consensus_distance=0.006



=== Log clique consensus distance @ E1B10 ===
clique1_consensus_distance=0.000
clique2_consensus_distance=0.025


[E 1B20 |  16128/50000 ( 32%) ] Loss: 2.2616 top1= 19.8864

=== Log global consensus distance @ E1B20 ===
consensus_distance=0.044



=== Log clique consensus distance @ E1B20 ===
clique1_consensus_distance=0.000
clique2_consensus_distance=0.175


[E 1B30 |  23808/50000 ( 48%) ] Loss: 2.1744 top1= 22.5852

=== Log global consensus distance @ E1B30 ===
consensus_distance=0.128



=== Log clique consensus distance @ E1B30 ===
clique1_consensus_distance=0.000
clique2_consensus_distance=0.510


[E 1B40 |  31488/50000 ( 63%) ] Loss: 2.1313 top1= 21.0227

=== Log global consensus distance @ E1B40 ===
consensus_distance=0.203



=== Log clique consensus distance @ E1B40 ===
clique1_consensus_distance=0.000
clique2_consensus_distance=0.804


[E 1B50 |  39168/50000 ( 78%) ] Loss: 2.1531 top1= 21.5909

=== Log global consensus distance @ E1B50 ===
consensus_distance=0.327



=== Log clique consensus distance @ E1B50 ===
clique1_consensus_distance=0.002
clique2_consensus_distance=1.305


[E 1B60 |  46848/50000 ( 94%) ] Loss: 2.0818 top1= 24.8580

=== Log global consensus distance @ E1B60 ===
consensus_distance=0.686



=== Log clique consensus distance @ E1B60 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=2.750



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3166 top1= 10.0060


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.4972 top1= 10.0060


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.3571 top1= 10.0160

Train epoch 2
[E 2B0  |    768/50000 (  2%) ] Loss: 2.1598 top1= 26.5625

=== Log global consensus distance @ E2B0 ===
consensus_distance=1.056



=== Log clique consensus distance @ E2B0 ===
clique1_consensus_distance=0.009
clique2_consensus_distance=4.244


[E 2B10 |   8448/50000 ( 17%) ] Loss: 2.1038 top1= 25.5682

=== Log global consensus distance @ E2B10 ===
consensus_distance=2.436



=== Log clique consensus distance @ E2B10 ===
clique1_consensus_distance=0.011
clique2_consensus_distance=9.864


[E 2B20 |  16128/50000 ( 32%) ] Loss: 2.1834 top1= 23.5795

=== Log global consensus distance @ E2B20 ===
consensus_distance=5.680



=== Log clique consensus distance @ E2B20 ===
clique1_consensus_distance=0.010
clique2_consensus_distance=23.126


[E 2B30 |  23808/50000 ( 48%) ] Loss: 2.1072 top1= 22.5852

=== Log global consensus distance @ E2B30 ===
consensus_distance=10.541



=== Log clique consensus distance @ E2B30 ===
clique1_consensus_distance=0.012
clique2_consensus_distance=42.789


[E 2B40 |  31488/50000 ( 63%) ] Loss: 2.1253 top1= 22.3011

=== Log global consensus distance @ E2B40 ===
consensus_distance=15.328



=== Log clique consensus distance @ E2B40 ===
clique1_consensus_distance=0.010
clique2_consensus_distance=62.064


[E 2B50 |  39168/50000 ( 78%) ] Loss: 2.1220 top1= 19.1761

=== Log global consensus distance @ E2B50 ===
consensus_distance=19.398



=== Log clique consensus distance @ E2B50 ===
clique1_consensus_distance=0.008
clique2_consensus_distance=78.411


[E 2B60 |  46848/50000 ( 94%) ] Loss: 2.0245 top1= 21.8750

=== Log global consensus distance @ E2B60 ===
consensus_distance=23.047



=== Log clique consensus distance @ E2B60 ===
clique1_consensus_distance=0.011
clique2_consensus_distance=93.087



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2896 top1= 10.0060


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.4029 top1= 17.7484


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.2950 top1= 11.3582

Train epoch 3
[E 3B0  |    768/50000 (  2%) ] Loss: 2.1087 top1= 21.7330

=== Log global consensus distance @ E3B0 ===
consensus_distance=24.954



=== Log clique consensus distance @ E3B0 ===
clique1_consensus_distance=0.013
clique2_consensus_distance=100.759


[E 3B10 |   8448/50000 ( 17%) ] Loss: 2.0358 top1= 24.0057

=== Log global consensus distance @ E3B10 ===
consensus_distance=28.867



=== Log clique consensus distance @ E3B10 ===
clique1_consensus_distance=0.010
clique2_consensus_distance=116.545


[E 3B20 |  16128/50000 ( 32%) ] Loss: 2.0779 top1= 22.0170

=== Log global consensus distance @ E3B20 ===
consensus_distance=32.652



=== Log clique consensus distance @ E3B20 ===
clique1_consensus_distance=0.011
clique2_consensus_distance=131.795


[E 3B30 |  23808/50000 ( 48%) ] Loss: 2.0322 top1= 21.4489

=== Log global consensus distance @ E3B30 ===
consensus_distance=36.653



=== Log clique consensus distance @ E3B30 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=147.938


[E 3B40 |  31488/50000 ( 63%) ] Loss: 1.9864 top1= 25.0000

=== Log global consensus distance @ E3B40 ===
consensus_distance=41.370



=== Log clique consensus distance @ E3B40 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=166.999


[E 3B50 |  39168/50000 ( 78%) ] Loss: 2.0206 top1= 25.7102

=== Log global consensus distance @ E3B50 ===
consensus_distance=45.935



=== Log clique consensus distance @ E3B50 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=185.404


[E 3B60 |  46848/50000 ( 94%) ] Loss: 1.8975 top1= 29.6875

=== Log global consensus distance @ E3B60 ===
consensus_distance=50.921



=== Log clique consensus distance @ E3B60 ===
clique1_consensus_distance=0.008
clique2_consensus_distance=205.541



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2572 top1= 19.7115


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.2582 top1= 19.0805


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.2208 top1= 15.4347

Train epoch 4
[E 4B0  |    768/50000 (  2%) ] Loss: 2.0189 top1= 25.0000

=== Log global consensus distance @ E4B0 ===
consensus_distance=53.735



=== Log clique consensus distance @ E4B0 ===
clique1_consensus_distance=0.009
clique2_consensus_distance=216.886


[E 4B10 |   8448/50000 ( 17%) ] Loss: 1.9250 top1= 30.9659

=== Log global consensus distance @ E4B10 ===
consensus_distance=59.973



=== Log clique consensus distance @ E4B10 ===
clique1_consensus_distance=0.008
clique2_consensus_distance=242.082


[E 4B20 |  16128/50000 ( 32%) ] Loss: 2.0185 top1= 24.7159

=== Log global consensus distance @ E4B20 ===
consensus_distance=66.924



=== Log clique consensus distance @ E4B20 ===
clique1_consensus_distance=0.009
clique2_consensus_distance=270.158


[E 4B30 |  23808/50000 ( 48%) ] Loss: 1.9636 top1= 27.1307

=== Log global consensus distance @ E4B30 ===
consensus_distance=75.357



=== Log clique consensus distance @ E4B30 ===
clique1_consensus_distance=0.008
clique2_consensus_distance=304.210


[E 4B40 |  31488/50000 ( 63%) ] Loss: 1.9943 top1= 27.2727

=== Log global consensus distance @ E4B40 ===
consensus_distance=84.047



=== Log clique consensus distance @ E4B40 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=339.279


[E 4B50 |  39168/50000 ( 78%) ] Loss: 1.9563 top1= 25.8523

=== Log global consensus distance @ E4B50 ===
consensus_distance=92.295



=== Log clique consensus distance @ E4B50 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=372.539


[E 4B60 |  46848/50000 ( 94%) ] Loss: 1.8158 top1= 33.2386

=== Log global consensus distance @ E4B60 ===
consensus_distance=100.092



=== Log clique consensus distance @ E4B60 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=403.988



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2696 top1= 18.8401


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.1618 top1= 16.2159


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1515 top1= 21.9251

Train epoch 5
[E 5B0  |    768/50000 (  2%) ] Loss: 2.0418 top1= 25.0000

=== Log global consensus distance @ E5B0 ===
consensus_distance=104.714



=== Log clique consensus distance @ E5B0 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=422.663


[E 5B10 |   8448/50000 ( 17%) ] Loss: 1.8903 top1= 29.1193

=== Log global consensus distance @ E5B10 ===
consensus_distance=114.834



=== Log clique consensus distance @ E5B10 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=463.498


[E 5B20 |  16128/50000 ( 32%) ] Loss: 1.9468 top1= 26.4205

=== Log global consensus distance @ E5B20 ===
consensus_distance=123.680



=== Log clique consensus distance @ E5B20 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=499.139


[E 5B30 |  23808/50000 ( 48%) ] Loss: 1.9330 top1= 31.6761

=== Log global consensus distance @ E5B30 ===
consensus_distance=133.523



=== Log clique consensus distance @ E5B30 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=538.935


[E 5B40 |  31488/50000 ( 63%) ] Loss: 2.0336 top1= 26.8466

=== Log global consensus distance @ E5B40 ===
consensus_distance=144.444



=== Log clique consensus distance @ E5B40 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=583.006


[E 5B50 |  39168/50000 ( 78%) ] Loss: 1.9232 top1= 25.9943

=== Log global consensus distance @ E5B50 ===
consensus_distance=153.018



=== Log clique consensus distance @ E5B50 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=617.465


[E 5B60 |  46848/50000 ( 94%) ] Loss: 1.7824 top1= 34.2330

=== Log global consensus distance @ E5B60 ===
consensus_distance=160.583



=== Log clique consensus distance @ E5B60 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=647.954



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3011 top1= 19.7115


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.0377 top1= 17.5280


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1443 top1= 22.2356

Train epoch 6
[E 6B0  |    768/50000 (  2%) ] Loss: 1.9359 top1= 27.4148

=== Log global consensus distance @ E6B0 ===
consensus_distance=165.030



=== Log clique consensus distance @ E6B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=665.913


[E 6B10 |   8448/50000 ( 17%) ] Loss: 1.8558 top1= 31.6761

=== Log global consensus distance @ E6B10 ===
consensus_distance=175.506



=== Log clique consensus distance @ E6B10 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=708.157


[E 6B20 |  16128/50000 ( 32%) ] Loss: 1.9816 top1= 29.5455

=== Log global consensus distance @ E6B20 ===
consensus_distance=187.806



=== Log clique consensus distance @ E6B20 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=757.904


[E 6B30 |  23808/50000 ( 48%) ] Loss: 1.8497 top1= 33.2386

=== Log global consensus distance @ E6B30 ===
consensus_distance=201.724



=== Log clique consensus distance @ E6B30 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=814.099


[E 6B40 |  31488/50000 ( 63%) ] Loss: 1.9368 top1= 33.2386

=== Log global consensus distance @ E6B40 ===
consensus_distance=216.712



=== Log clique consensus distance @ E6B40 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=874.647


[E 6B50 |  39168/50000 ( 78%) ] Loss: 1.8933 top1= 31.2500

=== Log global consensus distance @ E6B50 ===
consensus_distance=231.697



=== Log clique consensus distance @ E6B50 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=935.069


[E 6B60 |  46848/50000 ( 94%) ] Loss: 1.8604 top1= 35.5114

=== Log global consensus distance @ E6B60 ===
consensus_distance=245.985



=== Log clique consensus distance @ E6B60 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=992.743



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3334 top1= 16.8670


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.1004 top1= 17.7183


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1328 top1= 19.3810

Train epoch 7
[E 7B0  |    768/50000 (  2%) ] Loss: 2.0386 top1= 26.9886

=== Log global consensus distance @ E7B0 ===
consensus_distance=254.848



=== Log clique consensus distance @ E7B0 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=1028.559


[E 7B10 |   8448/50000 ( 17%) ] Loss: 1.8309 top1= 32.9545

=== Log global consensus distance @ E7B10 ===
consensus_distance=271.849



=== Log clique consensus distance @ E7B10 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=1097.129


[E 7B20 |  16128/50000 ( 32%) ] Loss: 1.9797 top1= 33.0966

=== Log global consensus distance @ E7B20 ===
consensus_distance=287.455



=== Log clique consensus distance @ E7B20 ===
clique1_consensus_distance=0.007
clique2_consensus_distance=1160.082


[E 7B30 |  23808/50000 ( 48%) ] Loss: 1.8941 top1= 34.0909

=== Log global consensus distance @ E7B30 ===
consensus_distance=305.190



=== Log clique consensus distance @ E7B30 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=1231.746


[E 7B40 |  31488/50000 ( 63%) ] Loss: 1.9126 top1= 34.8011

=== Log global consensus distance @ E7B40 ===
consensus_distance=323.760



=== Log clique consensus distance @ E7B40 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1306.670


[E 7B50 |  39168/50000 ( 78%) ] Loss: 1.8648 top1= 35.0852

=== Log global consensus distance @ E7B50 ===
consensus_distance=340.229



=== Log clique consensus distance @ E7B50 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1372.979


[E 7B60 |  46848/50000 ( 94%) ] Loss: 1.7981 top1= 40.7670

=== Log global consensus distance @ E7B60 ===
consensus_distance=356.471



=== Log clique consensus distance @ E7B60 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1438.538



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3762 top1= 10.6470


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.0211 top1= 21.3041


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1192 top1= 17.7083

Train epoch 8
[E 8B0  |    768/50000 (  2%) ] Loss: 2.0342 top1= 28.4091

=== Log global consensus distance @ E8B0 ===
consensus_distance=366.305



=== Log clique consensus distance @ E8B0 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=1478.269


[E 8B10 |   8448/50000 ( 17%) ] Loss: 1.8647 top1= 34.3750

=== Log global consensus distance @ E8B10 ===
consensus_distance=385.089



=== Log clique consensus distance @ E8B10 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=1554.060


[E 8B20 |  16128/50000 ( 32%) ] Loss: 1.8678 top1= 34.2330

=== Log global consensus distance @ E8B20 ===
consensus_distance=401.574



=== Log clique consensus distance @ E8B20 ===
clique1_consensus_distance=0.008
clique2_consensus_distance=1620.462


[E 8B30 |  23808/50000 ( 48%) ] Loss: 1.8748 top1= 34.5170

=== Log global consensus distance @ E8B30 ===
consensus_distance=419.761



=== Log clique consensus distance @ E8B30 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=1693.960


[E 8B40 |  31488/50000 ( 63%) ] Loss: 1.9601 top1= 32.2443

=== Log global consensus distance @ E8B40 ===
consensus_distance=439.967



=== Log clique consensus distance @ E8B40 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1775.445


[E 8B50 |  39168/50000 ( 78%) ] Loss: 1.8691 top1= 32.9545

=== Log global consensus distance @ E8B50 ===
consensus_distance=456.875



=== Log clique consensus distance @ E8B50 ===
clique1_consensus_distance=0.003
clique2_consensus_distance=1843.445


[E 8B60 |  46848/50000 ( 94%) ] Loss: 1.7981 top1= 39.4886

=== Log global consensus distance @ E8B60 ===
consensus_distance=474.335



=== Log clique consensus distance @ E8B60 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1913.961



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4431 top1= 10.1062


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=1.9562 top1= 20.8934


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1129 top1= 20.1522

Train epoch 9
[E 9B0  |    768/50000 (  2%) ] Loss: 1.9576 top1= 29.8295

=== Log global consensus distance @ E9B0 ===
consensus_distance=484.947



=== Log clique consensus distance @ E9B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1956.863


[E 9B10 |   8448/50000 ( 17%) ] Loss: 1.8134 top1= 41.3352

=== Log global consensus distance @ E9B10 ===
consensus_distance=507.028



=== Log clique consensus distance @ E9B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2045.978


[E 9B20 |  16128/50000 ( 32%) ] Loss: 1.9399 top1= 36.9318

=== Log global consensus distance @ E9B20 ===
consensus_distance=528.507



=== Log clique consensus distance @ E9B20 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=2132.614


[E 9B30 |  23808/50000 ( 48%) ] Loss: 1.8336 top1= 31.8182

=== Log global consensus distance @ E9B30 ===
consensus_distance=549.067



=== Log clique consensus distance @ E9B30 ===
clique1_consensus_distance=0.006
clique2_consensus_distance=2215.423


[E 9B40 |  31488/50000 ( 63%) ] Loss: 1.8637 top1= 30.6818

=== Log global consensus distance @ E9B40 ===
consensus_distance=566.062



=== Log clique consensus distance @ E9B40 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2283.851


[E 9B50 |  39168/50000 ( 78%) ] Loss: 1.8672 top1= 34.8011

=== Log global consensus distance @ E9B50 ===
consensus_distance=581.109



=== Log clique consensus distance @ E9B50 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2344.426


[E 9B60 |  46848/50000 ( 94%) ] Loss: 1.7489 top1= 39.9148

=== Log global consensus distance @ E9B60 ===
consensus_distance=599.765



=== Log clique consensus distance @ E9B60 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2419.872



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4963 top1= 10.2464


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=1.9600 top1= 25.5108


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.1042 top1= 19.6214

Train epoch 10
[E10B0  |    768/50000 (  2%) ] Loss: 1.9569 top1= 29.6875

=== Log global consensus distance @ E10B0 ===
consensus_distance=610.762



=== Log clique consensus distance @ E10B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=2464.200


[E10B10 |   8448/50000 ( 17%) ] Loss: 1.8209 top1= 40.0568

=== Log global consensus distance @ E10B10 ===
consensus_distance=631.367



=== Log clique consensus distance @ E10B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2547.182


[E10B20 |  16128/50000 ( 32%) ] Loss: 1.9018 top1= 36.7898

=== Log global consensus distance @ E10B20 ===
consensus_distance=650.572



=== Log clique consensus distance @ E10B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=2624.680


[E10B30 |  23808/50000 ( 48%) ] Loss: 1.7435 top1= 38.3523

=== Log global consensus distance @ E10B30 ===
consensus_distance=668.357



=== Log clique consensus distance @ E10B30 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=2696.349


[E10B40 |  31488/50000 ( 63%) ] Loss: 1.8839 top1= 32.8125

=== Log global consensus distance @ E10B40 ===
consensus_distance=687.423



=== Log clique consensus distance @ E10B40 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=2773.248


[E10B50 |  39168/50000 ( 78%) ] Loss: 1.8340 top1= 36.5057

=== Log global consensus distance @ E10B50 ===
consensus_distance=704.340



=== Log clique consensus distance @ E10B50 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=2841.385


