
=== 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 LabelFlippingWorker
=> Add worker LabelFlippingWorker

=== Start adding graph ===
<codes.graph_utils.RandomSmallWorldGraph object at 0x7fc1e0a93400>

Train epoch 1
[E 1B0  |    384/60000 (  1%) ] Loss: 2.3054 top1= 10.0000

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



=== Log global consensus distance @ E1B0 ===
consensus_distance=0.001



=== Log average shortest path distance for small world @ E1B0 ===
2.7777777777777777


[E 1B10 |   4224/60000 (  7%) ] Loss: 0.1557 top1= 94.6875

=== Log global consensus distance @ E1B10 ===
consensus_distance=0.348


[E 1B20 |   8064/60000 ( 13%) ] Loss: 0.1069 top1= 98.1250

=== Log global consensus distance @ E1B20 ===
consensus_distance=0.222


[E 1B30 |  11904/60000 ( 20%) ] Loss: 0.0803 top1= 98.4375

=== Log global consensus distance @ E1B30 ===
consensus_distance=0.136


[E 1B40 |  15744/60000 ( 26%) ] Loss: 0.0811 top1= 98.4375

=== Log global consensus distance @ E1B40 ===
consensus_distance=0.095


[E 1B50 |  19584/60000 ( 33%) ] Loss: 0.0797 top1= 97.8125

=== Log global consensus distance @ E1B50 ===
consensus_distance=0.079



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.9161 top1= 66.9872

Train epoch 2
[E 2B0  |    384/60000 (  1%) ] Loss: 0.1001 top1= 98.1250

=== Log global consensus distance @ E2B0 ===
consensus_distance=0.065


[E 2B10 |   4224/60000 (  7%) ] Loss: 0.0485 top1= 99.6875

=== Log global consensus distance @ E2B10 ===
consensus_distance=0.059


[E 2B20 |   8064/60000 ( 13%) ] Loss: 0.0459 top1= 99.0625

=== Log global consensus distance @ E2B20 ===
consensus_distance=0.057


[E 2B30 |  11904/60000 ( 20%) ] Loss: 0.0554 top1= 98.4375

=== Log global consensus distance @ E2B30 ===
consensus_distance=0.052


[E 2B40 |  15744/60000 ( 26%) ] Loss: 0.0775 top1= 97.8125

=== Log global consensus distance @ E2B40 ===
consensus_distance=0.049


[E 2B50 |  19584/60000 ( 33%) ] Loss: 0.0550 top1= 98.4375

=== Log global consensus distance @ E2B50 ===
consensus_distance=0.053



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.7640 top1= 72.3257

Train epoch 3
[E 3B0  |    384/60000 (  1%) ] Loss: 0.0849 top1= 98.1250

=== Log global consensus distance @ E3B0 ===
consensus_distance=0.047


[E 3B10 |   4224/60000 (  7%) ] Loss: 0.0353 top1= 99.3750

=== Log global consensus distance @ E3B10 ===
consensus_distance=0.049


[E 3B20 |   8064/60000 ( 13%) ] Loss: 0.0461 top1= 98.7500

=== Log global consensus distance @ E3B20 ===
consensus_distance=0.047


[E 3B30 |  11904/60000 ( 20%) ] Loss: 0.0474 top1= 99.0625

=== Log global consensus distance @ E3B30 ===
consensus_distance=0.049


[E 3B40 |  15744/60000 ( 26%) ] Loss: 0.0678 top1= 97.8125

=== Log global consensus distance @ E3B40 ===
consensus_distance=0.048


[E 3B50 |  19584/60000 ( 33%) ] Loss: 0.0435 top1= 98.7500

=== Log global consensus distance @ E3B50 ===
consensus_distance=0.048



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.7580 top1= 71.8650

Train epoch 4
[E 4B0  |    384/60000 (  1%) ] Loss: 0.0655 top1= 97.8125

=== Log global consensus distance @ E4B0 ===
consensus_distance=0.043


[E 4B10 |   4224/60000 (  7%) ] Loss: 0.0251 top1= 99.0625

=== Log global consensus distance @ E4B10 ===
consensus_distance=0.040


[E 4B20 |   8064/60000 ( 13%) ] Loss: 0.0273 top1= 99.6875

=== Log global consensus distance @ E4B20 ===
consensus_distance=0.040


[E 4B30 |  11904/60000 ( 20%) ] Loss: 0.0239 top1= 99.3750

=== Log global consensus distance @ E4B30 ===
consensus_distance=0.040


[E 4B40 |  15744/60000 ( 26%) ] Loss: 0.0507 top1= 98.7500

=== Log global consensus distance @ E4B40 ===
consensus_distance=0.036


[E 4B50 |  19584/60000 ( 33%) ] Loss: 0.0334 top1= 99.3750

=== Log global consensus distance @ E4B50 ===
consensus_distance=0.045



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.7859 top1= 71.8850

Train epoch 5
[E 5B0  |    384/60000 (  1%) ] Loss: 0.0604 top1= 99.0625

=== Log global consensus distance @ E5B0 ===
consensus_distance=0.041


[E 5B10 |   4224/60000 (  7%) ] Loss: 0.0236 top1= 99.3750

=== Log global consensus distance @ E5B10 ===
consensus_distance=0.037


[E 5B20 |   8064/60000 ( 13%) ] Loss: 0.0195 top1= 99.3750

=== Log global consensus distance @ E5B20 ===
consensus_distance=0.037


[E 5B30 |  11904/60000 ( 20%) ] Loss: 0.0427 top1= 97.8125

=== Log global consensus distance @ E5B30 ===
consensus_distance=0.037


[E 5B40 |  15744/60000 ( 26%) ] Loss: 0.0325 top1= 99.0625

=== Log global consensus distance @ E5B40 ===
consensus_distance=0.039


[E 5B50 |  19584/60000 ( 33%) ] Loss: 0.0372 top1= 99.0625

=== Log global consensus distance @ E5B50 ===
consensus_distance=0.040



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.6656 top1= 78.9263

Train epoch 6
[E 6B0  |    384/60000 (  1%) ] Loss: 0.0538 top1= 98.7500

=== Log global consensus distance @ E6B0 ===
consensus_distance=0.043


[E 6B10 |   4224/60000 (  7%) ] Loss: 0.0173 top1= 99.6875

=== Log global consensus distance @ E6B10 ===
consensus_distance=0.046


[E 6B20 |   8064/60000 ( 13%) ] Loss: 0.0276 top1= 99.3750

=== Log global consensus distance @ E6B20 ===
consensus_distance=0.041


[E 6B30 |  11904/60000 ( 20%) ] Loss: 0.0268 top1= 99.3750

=== Log global consensus distance @ E6B30 ===
consensus_distance=0.037


[E 6B40 |  15744/60000 ( 26%) ] Loss: 0.0263 top1= 99.3750

=== Log global consensus distance @ E6B40 ===
consensus_distance=0.037


[E 6B50 |  19584/60000 ( 33%) ] Loss: 0.0255 top1= 99.3750

=== Log global consensus distance @ E6B50 ===
consensus_distance=0.041



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5924 top1= 82.8726

Train epoch 7
[E 7B0  |    384/60000 (  1%) ] Loss: 0.0628 top1= 98.1250

=== Log global consensus distance @ E7B0 ===
consensus_distance=0.038


[E 7B10 |   4224/60000 (  7%) ] Loss: 0.0222 top1= 99.0625

=== Log global consensus distance @ E7B10 ===
consensus_distance=0.045


[E 7B20 |   8064/60000 ( 13%) ] Loss: 0.0104 top1= 99.6875

=== Log global consensus distance @ E7B20 ===
consensus_distance=0.060


[E 7B30 |  11904/60000 ( 20%) ] Loss: 0.0373 top1= 98.1250

=== Log global consensus distance @ E7B30 ===
consensus_distance=0.046


[E 7B40 |  15744/60000 ( 26%) ] Loss: 0.0238 top1= 99.6875

=== Log global consensus distance @ E7B40 ===
consensus_distance=0.040


[E 7B50 |  19584/60000 ( 33%) ] Loss: 0.0313 top1= 98.7500

=== Log global consensus distance @ E7B50 ===
consensus_distance=0.042



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5471 top1= 83.9844

Train epoch 8
[E 8B0  |    384/60000 (  1%) ] Loss: 0.0247 top1= 99.3750

=== Log global consensus distance @ E8B0 ===
consensus_distance=0.042


[E 8B10 |   4224/60000 (  7%) ] Loss: 0.0321 top1= 99.3750

=== Log global consensus distance @ E8B10 ===
consensus_distance=0.043


[E 8B20 |   8064/60000 ( 13%) ] Loss: 0.0218 top1= 99.3750

=== Log global consensus distance @ E8B20 ===
consensus_distance=0.039


[E 8B30 |  11904/60000 ( 20%) ] Loss: 0.0347 top1= 99.0625

=== Log global consensus distance @ E8B30 ===
consensus_distance=0.037


[E 8B40 |  15744/60000 ( 26%) ] Loss: 0.0296 top1= 99.3750

=== Log global consensus distance @ E8B40 ===
consensus_distance=0.038


[E 8B50 |  19584/60000 ( 33%) ] Loss: 0.0190 top1= 99.3750

=== Log global consensus distance @ E8B50 ===
consensus_distance=0.038



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5742 top1= 81.7508

Train epoch 9
[E 9B0  |    384/60000 (  1%) ] Loss: 0.0331 top1= 98.4375

=== Log global consensus distance @ E9B0 ===
consensus_distance=0.038


[E 9B10 |   4224/60000 (  7%) ] Loss: 0.0212 top1= 99.6875

=== Log global consensus distance @ E9B10 ===
consensus_distance=0.040


[E 9B20 |   8064/60000 ( 13%) ] Loss: 0.0283 top1= 99.0625

=== Log global consensus distance @ E9B20 ===
consensus_distance=0.045


[E 9B30 |  11904/60000 ( 20%) ] Loss: 0.0249 top1= 99.3750

=== Log global consensus distance @ E9B30 ===
consensus_distance=0.046


[E 9B40 |  15744/60000 ( 26%) ] Loss: 0.0262 top1= 99.3750

=== Log global consensus distance @ E9B40 ===
consensus_distance=0.040


[E 9B50 |  19584/60000 ( 33%) ] Loss: 0.0370 top1= 99.0625

=== Log global consensus distance @ E9B50 ===
consensus_distance=0.050



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5583 top1= 81.9111

Train epoch 10
[E10B0  |    384/60000 (  1%) ] Loss: 0.0420 top1= 98.1250

=== Log global consensus distance @ E10B0 ===
consensus_distance=0.043


[E10B10 |   4224/60000 (  7%) ] Loss: 0.0156 top1= 99.6875

=== Log global consensus distance @ E10B10 ===
consensus_distance=0.042


[E10B20 |   8064/60000 ( 13%) ] Loss: 0.0173 top1= 99.6875

=== Log global consensus distance @ E10B20 ===
consensus_distance=0.035


[E10B30 |  11904/60000 ( 20%) ] Loss: 0.0199 top1= 99.6875

=== Log global consensus distance @ E10B30 ===
consensus_distance=0.035


[E10B40 |  15744/60000 ( 26%) ] Loss: 0.0186 top1= 99.6875

=== Log global consensus distance @ E10B40 ===
consensus_distance=0.033


[E10B50 |  19584/60000 ( 33%) ] Loss: 0.0154 top1= 99.3750

=== Log global consensus distance @ E10B50 ===
consensus_distance=0.037



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5459 top1= 81.1498

Train epoch 11
[E11B0  |    384/60000 (  1%) ] Loss: 0.0348 top1= 98.7500

=== Log global consensus distance @ E11B0 ===
consensus_distance=0.040


[E11B10 |   4224/60000 (  7%) ] Loss: 0.0078 top1=100.0000

=== Log global consensus distance @ E11B10 ===
consensus_distance=0.043


[E11B20 |   8064/60000 ( 13%) ] Loss: 0.0203 top1= 99.3750

=== Log global consensus distance @ E11B20 ===
consensus_distance=0.043


[E11B30 |  11904/60000 ( 20%) ] Loss: 0.0072 top1=100.0000

=== Log global consensus distance @ E11B30 ===
consensus_distance=0.033


[E11B40 |  15744/60000 ( 26%) ] Loss: 0.0173 top1= 99.0625

=== Log global consensus distance @ E11B40 ===
consensus_distance=0.033


[E11B50 |  19584/60000 ( 33%) ] Loss: 0.0141 top1= 99.6875

=== Log global consensus distance @ E11B50 ===
consensus_distance=0.048



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5674 top1= 80.9395

Train epoch 12
[E12B0  |    384/60000 (  1%) ] Loss: 0.0221 top1= 99.0625

=== Log global consensus distance @ E12B0 ===
consensus_distance=0.048


[E12B10 |   4224/60000 (  7%) ] Loss: 0.0403 top1= 99.0625

=== Log global consensus distance @ E12B10 ===
consensus_distance=0.046


[E12B20 |   8064/60000 ( 13%) ] Loss: 0.0173 top1= 99.6875

=== Log global consensus distance @ E12B20 ===
consensus_distance=0.062


[E12B30 |  11904/60000 ( 20%) ] Loss: 0.0129 top1= 99.6875

=== Log global consensus distance @ E12B30 ===
consensus_distance=0.045


[E12B40 |  15744/60000 ( 26%) ] Loss: 0.0131 top1= 99.6875

=== Log global consensus distance @ E12B40 ===
consensus_distance=0.040


[E12B50 |  19584/60000 ( 33%) ] Loss: 0.0203 top1= 99.0625

=== Log global consensus distance @ E12B50 ===
consensus_distance=0.041



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5076 top1= 83.0829

Train epoch 13
[E13B0  |    384/60000 (  1%) ] Loss: 0.0215 top1= 99.3750

=== Log global consensus distance @ E13B0 ===
consensus_distance=0.039


[E13B10 |   4224/60000 (  7%) ] Loss: 0.0105 top1= 99.6875

=== Log global consensus distance @ E13B10 ===
consensus_distance=0.036


[E13B20 |   8064/60000 ( 13%) ] Loss: 0.0081 top1=100.0000

=== Log global consensus distance @ E13B20 ===
consensus_distance=0.036


[E13B30 |  11904/60000 ( 20%) ] Loss: 0.0071 top1=100.0000

=== Log global consensus distance @ E13B30 ===
consensus_distance=0.032


[E13B40 |  15744/60000 ( 26%) ] Loss: 0.0222 top1= 99.6875

=== Log global consensus distance @ E13B40 ===
consensus_distance=0.029


[E13B50 |  19584/60000 ( 33%) ] Loss: 0.0050 top1=100.0000

=== Log global consensus distance @ E13B50 ===
consensus_distance=0.034



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5685 top1= 81.2500

Train epoch 14
[E14B0  |    384/60000 (  1%) ] Loss: 0.0135 top1= 99.6875

=== Log global consensus distance @ E14B0 ===
consensus_distance=0.049


[E14B10 |   4224/60000 (  7%) ] Loss: 0.0085 top1=100.0000

=== Log global consensus distance @ E14B10 ===
consensus_distance=0.051


[E14B20 |   8064/60000 ( 13%) ] Loss: 0.0516 top1= 98.4375

=== Log global consensus distance @ E14B20 ===
consensus_distance=0.047


[E14B30 |  11904/60000 ( 20%) ] Loss: 0.0113 top1= 99.3750

=== Log global consensus distance @ E14B30 ===
consensus_distance=0.042


[E14B40 |  15744/60000 ( 26%) ] Loss: 0.0165 top1= 99.3750

=== Log global consensus distance @ E14B40 ===
consensus_distance=0.043


[E14B50 |  19584/60000 ( 33%) ] Loss: 0.0147 top1= 99.3750

=== Log global consensus distance @ E14B50 ===
consensus_distance=0.043



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5090 top1= 82.9627

Train epoch 15
[E15B0  |    384/60000 (  1%) ] Loss: 0.0203 top1= 99.3750

=== Log global consensus distance @ E15B0 ===
consensus_distance=0.035


[E15B10 |   4224/60000 (  7%) ] Loss: 0.0247 top1= 99.6875

=== Log global consensus distance @ E15B10 ===
consensus_distance=0.035


[E15B20 |   8064/60000 ( 13%) ] Loss: 0.0130 top1= 99.3750

=== Log global consensus distance @ E15B20 ===
consensus_distance=0.038


[E15B30 |  11904/60000 ( 20%) ] Loss: 0.0138 top1= 99.6875

=== Log global consensus distance @ E15B30 ===
consensus_distance=0.036


[E15B40 |  15744/60000 ( 26%) ] Loss: 0.0149 top1= 99.6875

=== Log global consensus distance @ E15B40 ===
consensus_distance=0.030


[E15B50 |  19584/60000 ( 33%) ] Loss: 0.0060 top1=100.0000

=== Log global consensus distance @ E15B50 ===
consensus_distance=0.032



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4699 top1= 84.3850

Train epoch 16
[E16B0  |    384/60000 (  1%) ] Loss: 0.0163 top1= 99.6875

=== Log global consensus distance @ E16B0 ===
consensus_distance=0.031


[E16B10 |   4224/60000 (  7%) ] Loss: 0.0100 top1= 99.6875

=== Log global consensus distance @ E16B10 ===
consensus_distance=0.031


[E16B20 |   8064/60000 ( 13%) ] Loss: 0.0076 top1= 99.6875

=== Log global consensus distance @ E16B20 ===
consensus_distance=0.032


[E16B30 |  11904/60000 ( 20%) ] Loss: 0.0653 top1= 98.4375

=== Log global consensus distance @ E16B30 ===
consensus_distance=0.034


[E16B40 |  15744/60000 ( 26%) ] Loss: 0.0095 top1=100.0000

=== Log global consensus distance @ E16B40 ===
consensus_distance=0.048


[E16B50 |  19584/60000 ( 33%) ] Loss: 0.0201 top1= 99.0625

=== Log global consensus distance @ E16B50 ===
consensus_distance=0.042



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5014 top1= 83.2632

Train epoch 17
[E17B0  |    384/60000 (  1%) ] Loss: 0.0227 top1= 99.3750

=== Log global consensus distance @ E17B0 ===
consensus_distance=0.040


[E17B10 |   4224/60000 (  7%) ] Loss: 0.0065 top1=100.0000

=== Log global consensus distance @ E17B10 ===
consensus_distance=0.054


[E17B20 |   8064/60000 ( 13%) ] Loss: 0.0077 top1=100.0000

=== Log global consensus distance @ E17B20 ===
consensus_distance=0.044


[E17B30 |  11904/60000 ( 20%) ] Loss: 0.0089 top1=100.0000

=== Log global consensus distance @ E17B30 ===
consensus_distance=0.035


[E17B40 |  15744/60000 ( 26%) ] Loss: 0.0126 top1=100.0000

=== Log global consensus distance @ E17B40 ===
consensus_distance=0.035


[E17B50 |  19584/60000 ( 33%) ] Loss: 0.0106 top1= 99.6875

=== Log global consensus distance @ E17B50 ===
consensus_distance=0.044



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4451 top1= 85.9976

Train epoch 18
[E18B0  |    384/60000 (  1%) ] Loss: 0.0229 top1= 99.0625

=== Log global consensus distance @ E18B0 ===
consensus_distance=0.040


[E18B10 |   4224/60000 (  7%) ] Loss: 0.0249 top1= 99.3750

=== Log global consensus distance @ E18B10 ===
consensus_distance=0.043


[E18B20 |   8064/60000 ( 13%) ] Loss: 0.0068 top1= 99.6875

=== Log global consensus distance @ E18B20 ===
consensus_distance=0.041


[E18B30 |  11904/60000 ( 20%) ] Loss: 0.0138 top1=100.0000

=== Log global consensus distance @ E18B30 ===
consensus_distance=0.038


[E18B40 |  15744/60000 ( 26%) ] Loss: 0.0241 top1= 99.0625

=== Log global consensus distance @ E18B40 ===
consensus_distance=0.042


[E18B50 |  19584/60000 ( 33%) ] Loss: 0.0065 top1=100.0000

=== Log global consensus distance @ E18B50 ===
consensus_distance=0.041



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5362 top1= 81.1999

Train epoch 19
[E19B0  |    384/60000 (  1%) ] Loss: 0.0294 top1= 99.3750

=== Log global consensus distance @ E19B0 ===
consensus_distance=0.032


[E19B10 |   4224/60000 (  7%) ] Loss: 0.0207 top1= 99.3750

=== Log global consensus distance @ E19B10 ===
consensus_distance=0.030


[E19B20 |   8064/60000 ( 13%) ] Loss: 0.0085 top1= 99.6875

=== Log global consensus distance @ E19B20 ===
consensus_distance=0.028


[E19B30 |  11904/60000 ( 20%) ] Loss: 0.0053 top1=100.0000

=== Log global consensus distance @ E19B30 ===
consensus_distance=0.038


[E19B40 |  15744/60000 ( 26%) ] Loss: 0.0240 top1= 99.6875

=== Log global consensus distance @ E19B40 ===
consensus_distance=0.043


[E19B50 |  19584/60000 ( 33%) ] Loss: 0.0147 top1= 99.3750

=== Log global consensus distance @ E19B50 ===
consensus_distance=0.040



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4923 top1= 84.4151

Train epoch 20
[E20B0  |    384/60000 (  1%) ] Loss: 0.0294 top1= 99.0625

=== Log global consensus distance @ E20B0 ===
consensus_distance=0.034


[E20B10 |   4224/60000 (  7%) ] Loss: 0.0049 top1= 99.6875

=== Log global consensus distance @ E20B10 ===
consensus_distance=0.042


[E20B20 |   8064/60000 ( 13%) ] Loss: 0.0055 top1=100.0000

=== Log global consensus distance @ E20B20 ===
consensus_distance=0.040


[E20B30 |  11904/60000 ( 20%) ] Loss: 0.0059 top1=100.0000

=== Log global consensus distance @ E20B30 ===
consensus_distance=0.047


[E20B40 |  15744/60000 ( 26%) ] Loss: 0.0092 top1=100.0000

=== Log global consensus distance @ E20B40 ===
consensus_distance=0.049


[E20B50 |  19584/60000 ( 33%) ] Loss: 0.0224 top1= 99.3750

=== Log global consensus distance @ E20B50 ===
consensus_distance=0.040



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4344 top1= 86.6887

Train epoch 21
[E21B0  |    384/60000 (  1%) ] Loss: 0.0295 top1= 99.6875

=== Log global consensus distance @ E21B0 ===
consensus_distance=0.039


[E21B10 |   4224/60000 (  7%) ] Loss: 0.0055 top1=100.0000

=== Log global consensus distance @ E21B10 ===
consensus_distance=0.034


[E21B20 |   8064/60000 ( 13%) ] Loss: 0.0197 top1= 99.6875

=== Log global consensus distance @ E21B20 ===
consensus_distance=0.034


[E21B30 |  11904/60000 ( 20%) ] Loss: 0.0076 top1= 99.6875

=== Log global consensus distance @ E21B30 ===
consensus_distance=0.030


[E21B40 |  15744/60000 ( 26%) ] Loss: 0.0044 top1=100.0000

=== Log global consensus distance @ E21B40 ===
consensus_distance=0.033


[E21B50 |  19584/60000 ( 33%) ] Loss: 0.0079 top1= 99.6875

=== Log global consensus distance @ E21B50 ===
consensus_distance=0.027



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4336 top1= 84.8257

Train epoch 22
[E22B0  |    384/60000 (  1%) ] Loss: 0.0057 top1=100.0000

=== Log global consensus distance @ E22B0 ===
consensus_distance=0.023


[E22B10 |   4224/60000 (  7%) ] Loss: 0.0027 top1=100.0000

=== Log global consensus distance @ E22B10 ===
consensus_distance=0.025


[E22B20 |   8064/60000 ( 13%) ] Loss: 0.0022 top1=100.0000

=== Log global consensus distance @ E22B20 ===
consensus_distance=0.021


[E22B30 |  11904/60000 ( 20%) ] Loss: 0.0053 top1=100.0000

=== Log global consensus distance @ E22B30 ===
consensus_distance=0.020


[E22B40 |  15744/60000 ( 26%) ] Loss: 0.0091 top1= 99.6875

=== Log global consensus distance @ E22B40 ===
consensus_distance=0.024


[E22B50 |  19584/60000 ( 33%) ] Loss: 0.0108 top1= 99.6875

=== Log global consensus distance @ E22B50 ===
consensus_distance=0.025



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4711 top1= 86.1378

Train epoch 23
[E23B0  |    384/60000 (  1%) ] Loss: 0.0067 top1=100.0000

=== Log global consensus distance @ E23B0 ===
consensus_distance=0.032


[E23B10 |   4224/60000 (  7%) ] Loss: 0.0118 top1=100.0000

=== Log global consensus distance @ E23B10 ===
consensus_distance=0.036


[E23B20 |   8064/60000 ( 13%) ] Loss: 0.0084 top1=100.0000

=== Log global consensus distance @ E23B20 ===
consensus_distance=0.032


[E23B30 |  11904/60000 ( 20%) ] Loss: 0.0056 top1=100.0000

=== Log global consensus distance @ E23B30 ===
consensus_distance=0.026


[E23B40 |  15744/60000 ( 26%) ] Loss: 0.0119 top1= 99.6875

=== Log global consensus distance @ E23B40 ===
consensus_distance=0.034


[E23B50 |  19584/60000 ( 33%) ] Loss: 0.0151 top1= 99.6875

=== Log global consensus distance @ E23B50 ===
consensus_distance=0.032



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4625 top1= 85.6370

Train epoch 24
[E24B0  |    384/60000 (  1%) ] Loss: 0.0280 top1= 99.0625

=== Log global consensus distance @ E24B0 ===
consensus_distance=0.027


[E24B10 |   4224/60000 (  7%) ] Loss: 0.0027 top1=100.0000

=== Log global consensus distance @ E24B10 ===
consensus_distance=0.029


[E24B20 |   8064/60000 ( 13%) ] Loss: 0.0156 top1= 99.3750

=== Log global consensus distance @ E24B20 ===
consensus_distance=0.024


[E24B30 |  11904/60000 ( 20%) ] Loss: 0.0149 top1= 99.6875

=== Log global consensus distance @ E24B30 ===
consensus_distance=0.028


[E24B40 |  15744/60000 ( 26%) ] Loss: 0.0078 top1= 99.6875

=== Log global consensus distance @ E24B40 ===
consensus_distance=0.034


[E24B50 |  19584/60000 ( 33%) ] Loss: 0.0125 top1= 99.6875

=== Log global consensus distance @ E24B50 ===
consensus_distance=0.041



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4457 top1= 85.7672

Train epoch 25
[E25B0  |    384/60000 (  1%) ] Loss: 0.0168 top1= 99.0625

=== Log global consensus distance @ E25B0 ===
consensus_distance=0.034


[E25B10 |   4224/60000 (  7%) ] Loss: 0.0151 top1= 99.3750

=== Log global consensus distance @ E25B10 ===
consensus_distance=0.038


[E25B20 |   8064/60000 ( 13%) ] Loss: 0.0030 top1=100.0000

=== Log global consensus distance @ E25B20 ===
consensus_distance=0.040


[E25B30 |  11904/60000 ( 20%) ] Loss: 0.0035 top1=100.0000

=== Log global consensus distance @ E25B30 ===
consensus_distance=0.036


[E25B40 |  15744/60000 ( 26%) ] Loss: 0.0060 top1=100.0000

=== Log global consensus distance @ E25B40 ===
consensus_distance=0.033


[E25B50 |  19584/60000 ( 33%) ] Loss: 0.0042 top1=100.0000

=== Log global consensus distance @ E25B50 ===
consensus_distance=0.026



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3508 top1= 89.9239

Train epoch 26
[E26B0  |    384/60000 (  1%) ] Loss: 0.0047 top1=100.0000

=== Log global consensus distance @ E26B0 ===
consensus_distance=0.020


[E26B10 |   4224/60000 (  7%) ] Loss: 0.0016 top1=100.0000

=== Log global consensus distance @ E26B10 ===
consensus_distance=0.021


[E26B20 |   8064/60000 ( 13%) ] Loss: 0.0098 top1= 99.3750

=== Log global consensus distance @ E26B20 ===
consensus_distance=0.025


[E26B30 |  11904/60000 ( 20%) ] Loss: 0.0042 top1=100.0000

=== Log global consensus distance @ E26B30 ===
consensus_distance=0.031


[E26B40 |  15744/60000 ( 26%) ] Loss: 0.0037 top1=100.0000

=== Log global consensus distance @ E26B40 ===
consensus_distance=0.032


[E26B50 |  19584/60000 ( 33%) ] Loss: 0.0043 top1=100.0000

=== Log global consensus distance @ E26B50 ===
consensus_distance=0.027



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3616 top1= 89.4231

Train epoch 27
[E27B0  |    384/60000 (  1%) ] Loss: 0.0228 top1= 99.0625

=== Log global consensus distance @ E27B0 ===
consensus_distance=0.024


[E27B10 |   4224/60000 (  7%) ] Loss: 0.0075 top1=100.0000

=== Log global consensus distance @ E27B10 ===
consensus_distance=0.029


[E27B20 |   8064/60000 ( 13%) ] Loss: 0.0039 top1=100.0000

=== Log global consensus distance @ E27B20 ===
consensus_distance=0.027


[E27B30 |  11904/60000 ( 20%) ] Loss: 0.0102 top1=100.0000

=== Log global consensus distance @ E27B30 ===
consensus_distance=0.022


[E27B40 |  15744/60000 ( 26%) ] Loss: 0.0021 top1=100.0000

=== Log global consensus distance @ E27B40 ===
consensus_distance=0.021


[E27B50 |  19584/60000 ( 33%) ] Loss: 0.0030 top1=100.0000

=== Log global consensus distance @ E27B50 ===
consensus_distance=0.024



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4256 top1= 87.1094

Train epoch 28
[E28B0  |    384/60000 (  1%) ] Loss: 0.0032 top1=100.0000

=== Log global consensus distance @ E28B0 ===
consensus_distance=0.022


[E28B10 |   4224/60000 (  7%) ] Loss: 0.0042 top1=100.0000

=== Log global consensus distance @ E28B10 ===
consensus_distance=0.022


[E28B20 |   8064/60000 ( 13%) ] Loss: 0.0868 top1= 97.8125

=== Log global consensus distance @ E28B20 ===
consensus_distance=0.040


[E28B30 |  11904/60000 ( 20%) ] Loss: 0.0077 top1= 99.6875

=== Log global consensus distance @ E28B30 ===
consensus_distance=0.221


[E28B40 |  15744/60000 ( 26%) ] Loss: 0.0217 top1= 99.0625

=== Log global consensus distance @ E28B40 ===
consensus_distance=0.092


[E28B50 |  19584/60000 ( 33%) ] Loss: 0.0350 top1= 99.0625

=== Log global consensus distance @ E28B50 ===
consensus_distance=0.085



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5614 top1= 81.7308

Train epoch 29
[E29B0  |    384/60000 (  1%) ] Loss: 0.0291 top1= 99.0625

=== Log global consensus distance @ E29B0 ===
consensus_distance=0.068


[E29B10 |   4224/60000 (  7%) ] Loss: 0.0105 top1= 99.6875

=== Log global consensus distance @ E29B10 ===
consensus_distance=0.051


[E29B20 |   8064/60000 ( 13%) ] Loss: 0.0120 top1=100.0000

=== Log global consensus distance @ E29B20 ===
consensus_distance=0.039


[E29B30 |  11904/60000 ( 20%) ] Loss: 0.0068 top1=100.0000

=== Log global consensus distance @ E29B30 ===
consensus_distance=0.035


[E29B40 |  15744/60000 ( 26%) ] Loss: 0.0085 top1= 99.3750

=== Log global consensus distance @ E29B40 ===
consensus_distance=0.026


[E29B50 |  19584/60000 ( 33%) ] Loss: 0.0063 top1=100.0000

=== Log global consensus distance @ E29B50 ===
consensus_distance=0.029



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4390 top1= 85.0561

Train epoch 30
[E30B0  |    384/60000 (  1%) ] Loss: 0.0336 top1= 99.0625

=== Log global consensus distance @ E30B0 ===
consensus_distance=0.026


[E30B10 |   4224/60000 (  7%) ] Loss: 0.0091 top1= 99.6875

=== Log global consensus distance @ E30B10 ===
consensus_distance=0.030


[E30B20 |   8064/60000 ( 13%) ] Loss: 0.0063 top1=100.0000

=== Log global consensus distance @ E30B20 ===
consensus_distance=0.030


[E30B30 |  11904/60000 ( 20%) ] Loss: 0.0059 top1=100.0000

=== Log global consensus distance @ E30B30 ===
consensus_distance=0.024


[E30B40 |  15744/60000 ( 26%) ] Loss: 0.0053 top1=100.0000

=== Log global consensus distance @ E30B40 ===
consensus_distance=0.029


[E30B50 |  19584/60000 ( 33%) ] Loss: 0.0026 top1=100.0000

=== Log global consensus distance @ E30B50 ===
consensus_distance=0.025



=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4205 top1= 85.9575

