
=== Start adding workers ===
=> Add worker SGDMWorker(index=0, momentum=0.9)
=> Add worker SGDMWorker(index=1, momentum=0.9)
=> Add worker SGDMWorker(index=2, momentum=0.9)
=> Add worker SGDMWorker(index=3, momentum=0.9)
=> Add worker SGDMWorker(index=4, momentum=0.9)
=> Add worker SGDMWorker(index=5, momentum=0.9)
=> Add worker SGDMWorker(index=6, momentum=0.9)
=> Add worker SGDMWorker(index=7, momentum=0.9)
=> Add worker SGDMWorker(index=8, momentum=0.9)
=> Add worker SGDMWorker(index=9, momentum=0.9)
=> Add worker SGDMWorker(index=10, momentum=0.9)
=> Add worker ByzantineWorker(index=11)

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

Train epoch 1
[E 1B0  |    384/60000 (  1%) ] Loss: 2.3142 top1=  7.6705

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


[E 1B10 |   4224/60000 (  7%) ] Loss: 1.9616 top1= 42.6136

=== Log global consensus distance @ E1B10 ===
consensus_distance=0.030



=== Log clique consensus distance @ E1B10 ===
clique1_consensus_distance=0.001
clique2_consensus_distance=0.033


[E 1B20 |   8064/60000 ( 13%) ] Loss: 1.7733 top1= 32.6705

=== Log global consensus distance @ E1B20 ===
consensus_distance=0.098



=== Log clique consensus distance @ E1B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=0.170



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4735 top1=  9.7456


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.5896 top1= 35.5268


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.4904 top1= 37.0793

Train epoch 2
[E 2B0  |    384/60000 (  1%) ] Loss: 1.3516 top1= 57.1023

=== Log global consensus distance @ E2B0 ===
consensus_distance=0.155



=== Log clique consensus distance @ E2B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=0.417


[E 2B10 |   4224/60000 (  7%) ] Loss: 1.0671 top1= 68.1818

=== Log global consensus distance @ E2B10 ===
consensus_distance=0.203



=== Log clique consensus distance @ E2B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=0.632


[E 2B20 |   8064/60000 ( 13%) ] Loss: 0.9433 top1= 72.1591

=== Log global consensus distance @ E2B20 ===
consensus_distance=0.232



=== Log clique consensus distance @ E2B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=0.758



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.7718 top1= 14.6835


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.8671 top1= 39.2829


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.9666 top1= 37.6302

Train epoch 3
[E 3B0  |    384/60000 (  1%) ] Loss: 0.7729 top1= 80.3977

=== Log global consensus distance @ E3B0 ===
consensus_distance=0.258



=== Log clique consensus distance @ E3B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=0.874


[E 3B10 |   4224/60000 (  7%) ] Loss: 0.6849 top1= 85.5114

=== Log global consensus distance @ E3B10 ===
consensus_distance=0.276



=== Log clique consensus distance @ E3B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=0.955


[E 3B20 |   8064/60000 ( 13%) ] Loss: 0.6742 top1= 84.0909

=== Log global consensus distance @ E3B20 ===
consensus_distance=0.288



=== Log clique consensus distance @ E3B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.009



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.0768 top1= 18.6899


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.9052 top1= 43.9904


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.1877 top1= 39.4732

Train epoch 4
[E 4B0  |    384/60000 (  1%) ] Loss: 0.5427 top1= 88.9205

=== Log global consensus distance @ E4B0 ===
consensus_distance=0.298



=== Log clique consensus distance @ E4B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.056


[E 4B10 |   4224/60000 (  7%) ] Loss: 0.5471 top1= 88.6364

=== Log global consensus distance @ E4B10 ===
consensus_distance=0.302



=== Log clique consensus distance @ E4B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.073


[E 4B20 |   8064/60000 ( 13%) ] Loss: 0.5752 top1= 86.3636

=== Log global consensus distance @ E4B20 ===
consensus_distance=0.306



=== Log clique consensus distance @ E4B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.093



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.3443 top1= 19.0405


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.9551 top1= 46.9852


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.3568 top1= 40.4447

Train epoch 5
[E 5B0  |    384/60000 (  1%) ] Loss: 0.4754 top1= 91.4773

=== Log global consensus distance @ E5B0 ===
consensus_distance=0.312



=== Log clique consensus distance @ E5B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.121


[E 5B10 |   4224/60000 (  7%) ] Loss: 0.5003 top1= 88.0682

=== Log global consensus distance @ E5B10 ===
consensus_distance=0.313



=== Log clique consensus distance @ E5B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.126


[E 5B20 |   8064/60000 ( 13%) ] Loss: 0.5398 top1= 86.9318

=== Log global consensus distance @ E5B20 ===
consensus_distance=0.316



=== Log clique consensus distance @ E5B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.141



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.5659 top1= 19.1506


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.9773 top1= 47.6562


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.4557 top1= 40.7953

Train epoch 6
[E 6B0  |    384/60000 (  1%) ] Loss: 0.4517 top1= 91.1932

=== Log global consensus distance @ E6B0 ===
consensus_distance=0.322



=== Log clique consensus distance @ E6B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.165


[E 6B10 |   4224/60000 (  7%) ] Loss: 0.4819 top1= 88.0682

=== Log global consensus distance @ E6B10 ===
consensus_distance=0.322



=== Log clique consensus distance @ E6B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.168


[E 6B20 |   8064/60000 ( 13%) ] Loss: 0.5240 top1= 87.2159

=== Log global consensus distance @ E6B20 ===
consensus_distance=0.326



=== Log clique consensus distance @ E6B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.182



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.7173 top1= 19.1406


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.9282 top1= 47.8666


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.4805 top1= 40.8454

Train epoch 7
[E 7B0  |    384/60000 (  1%) ] Loss: 0.4414 top1= 90.9091

=== Log global consensus distance @ E7B0 ===
consensus_distance=0.331



=== Log clique consensus distance @ E7B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.206


[E 7B10 |   4224/60000 (  7%) ] Loss: 0.4723 top1= 87.5000

=== Log global consensus distance @ E7B10 ===
consensus_distance=0.331



=== Log clique consensus distance @ E7B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.207


[E 7B20 |   8064/60000 ( 13%) ] Loss: 0.5184 top1= 87.5000

=== Log global consensus distance @ E7B20 ===
consensus_distance=0.334



=== Log clique consensus distance @ E7B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.219



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.7989 top1= 19.1707


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.8117 top1= 47.9868


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.4395 top1= 40.7953

Train epoch 8
[E 8B0  |    384/60000 (  1%) ] Loss: 0.4375 top1= 91.4773

=== Log global consensus distance @ E8B0 ===
consensus_distance=0.338



=== Log clique consensus distance @ E8B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.237


[E 8B10 |   4224/60000 (  7%) ] Loss: 0.4676 top1= 86.9318

=== Log global consensus distance @ E8B10 ===
consensus_distance=0.337



=== Log clique consensus distance @ E8B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.233


[E 8B20 |   8064/60000 ( 13%) ] Loss: 0.5187 top1= 87.2159

=== Log global consensus distance @ E8B20 ===
consensus_distance=0.339



=== Log clique consensus distance @ E8B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.242



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.7968 top1= 19.2107


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.6556 top1= 48.3173


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.3465 top1= 40.8454

Train epoch 9
[E 9B0  |    384/60000 (  1%) ] Loss: 0.4385 top1= 91.4773

=== Log global consensus distance @ E9B0 ===
consensus_distance=0.341



=== Log clique consensus distance @ E9B0 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.253


[E 9B10 |   4224/60000 (  7%) ] Loss: 0.4662 top1= 88.0682

=== Log global consensus distance @ E9B10 ===
consensus_distance=0.339



=== Log clique consensus distance @ E9B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.243


[E 9B20 |   8064/60000 ( 13%) ] Loss: 0.5220 top1= 87.2159

=== Log global consensus distance @ E9B20 ===
consensus_distance=0.340



=== Log clique consensus distance @ E9B20 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.245



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.7302 top1= 19.2408


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.5196 top1= 48.5777


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.2136 top1= 40.7853

Train epoch 10
[E10B0  |    384/60000 (  1%) ] Loss: 0.4442 top1= 91.4773

=== Log global consensus distance @ E10B0 ===
consensus_distance=0.340



=== Log clique consensus distance @ E10B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.248


[E10B10 |   4224/60000 (  7%) ] Loss: 0.4675 top1= 88.0682

=== Log global consensus distance @ E10B10 ===
consensus_distance=0.336



=== Log clique consensus distance @ E10B10 ===
clique1_consensus_distance=0.005
clique2_consensus_distance=1.231


[E10B20 |   8064/60000 ( 13%) ] Loss: 0.5238 top1= 87.7841

=== Log global consensus distance @ E10B20 ===
consensus_distance=0.335



=== Log clique consensus distance @ E10B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.225



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.5902 top1= 19.2408


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.4350 top1= 48.5377


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.0626 top1= 40.8854

Train epoch 11
[E11B0  |    384/60000 (  1%) ] Loss: 0.4481 top1= 92.0455

=== Log global consensus distance @ E11B0 ===
consensus_distance=0.333



=== Log clique consensus distance @ E11B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.218


[E11B10 |   4224/60000 (  7%) ] Loss: 0.4665 top1= 88.9205

=== Log global consensus distance @ E11B10 ===
consensus_distance=0.328



=== Log clique consensus distance @ E11B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.194


[E11B20 |   8064/60000 ( 13%) ] Loss: 0.5192 top1= 88.3523

=== Log global consensus distance @ E11B20 ===
consensus_distance=0.324



=== Log clique consensus distance @ E11B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.179



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.3878 top1= 19.2608


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.3788 top1= 48.3574


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.9073 top1= 40.9155

Train epoch 12
[E12B0  |    384/60000 (  1%) ] Loss: 0.4476 top1= 92.6136

=== Log global consensus distance @ E12B0 ===
consensus_distance=0.320



=== Log clique consensus distance @ E12B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.164


[E12B10 |   4224/60000 (  7%) ] Loss: 0.4632 top1= 89.4886

=== Log global consensus distance @ E12B10 ===
consensus_distance=0.315



=== Log clique consensus distance @ E12B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.137


[E12B20 |   8064/60000 ( 13%) ] Loss: 0.5138 top1= 88.6364

=== Log global consensus distance @ E12B20 ===
consensus_distance=0.311



=== Log clique consensus distance @ E12B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.121



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.2254 top1= 19.2508


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.3396 top1= 48.0569


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.8167 top1= 40.9255

Train epoch 13
[E13B0  |    384/60000 (  1%) ] Loss: 0.4475 top1= 92.3295

=== Log global consensus distance @ E13B0 ===
consensus_distance=0.307



=== Log clique consensus distance @ E13B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.108


[E13B10 |   4224/60000 (  7%) ] Loss: 0.4631 top1= 91.4773

=== Log global consensus distance @ E13B10 ===
consensus_distance=0.302



=== Log clique consensus distance @ E13B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.085


[E13B20 |   8064/60000 ( 13%) ] Loss: 0.5132 top1= 87.7841

=== Log global consensus distance @ E13B20 ===
consensus_distance=0.300



=== Log clique consensus distance @ E13B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.075



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.1290 top1= 19.2808


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.3221 top1= 47.7865


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.7770 top1= 40.8754

Train epoch 14
[E14B0  |    384/60000 (  1%) ] Loss: 0.4459 top1= 92.3295

=== Log global consensus distance @ E14B0 ===
consensus_distance=0.297



=== Log clique consensus distance @ E14B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.066


[E14B10 |   4224/60000 (  7%) ] Loss: 0.4621 top1= 91.7614

=== Log global consensus distance @ E14B10 ===
consensus_distance=0.294



=== Log clique consensus distance @ E14B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.049


[E14B20 |   8064/60000 ( 13%) ] Loss: 0.5112 top1= 88.9205

=== Log global consensus distance @ E14B20 ===
consensus_distance=0.292



=== Log clique consensus distance @ E14B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.042



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.0723 top1= 19.2808


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.3046 top1= 47.6162


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.7668 top1= 40.8554

Train epoch 15
[E15B0  |    384/60000 (  1%) ] Loss: 0.4434 top1= 92.0455

=== Log global consensus distance @ E15B0 ===
consensus_distance=0.290



=== Log clique consensus distance @ E15B0 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.036


[E15B10 |   4224/60000 (  7%) ] Loss: 0.4617 top1= 91.1932

=== Log global consensus distance @ E15B10 ===
consensus_distance=0.287



=== Log clique consensus distance @ E15B10 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.021


[E15B20 |   8064/60000 ( 13%) ] Loss: 0.5089 top1= 89.4886

=== Log global consensus distance @ E15B20 ===
consensus_distance=0.286



=== Log clique consensus distance @ E15B20 ===
clique1_consensus_distance=0.004
clique2_consensus_distance=1.017



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.0492 top1= 19.2808


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=2.2907 top1= 47.5861


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.7616 top1= 40.8954

