
=== 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 SGDMWorker(index=11, momentum=0.9)
=> Add worker SGDMWorker(index=12, momentum=0.9)
=> Add worker SGDMWorker(index=13, momentum=0.9)
=> Add worker SGDMWorker(index=14, momentum=0.9)
=> Add worker SGDMWorker(index=15, momentum=0.9)
=> Add worker SGDMWorker(index=16, momentum=0.9)
=> Add worker SGDMWorker(index=17, momentum=0.9)
=> Add worker SGDMWorker(index=18, momentum=0.9)
=> Add worker SGDMWorker(index=19, momentum=0.9)
=> Add worker ByzantineWorker(index=20)
=> Add worker ByzantineWorker(index=21)

=== Start adding graph ===
<codes.graph_utils.Dumbbell object at 0x7fdc022c4490>

Train epoch 1
[E 1B0  |    704/60000 (  1%) ] Loss: 2.3066 top1=  9.2188

=== Peeking data label distribution E1B0 ===
Worker 0 has targets: tensor([3, 2, 2, 4, 3], device='cuda:0')
Worker 1 has targets: tensor([1, 0, 0, 4, 0], device='cuda:0')
Worker 2 has targets: tensor([4, 1, 0, 1, 0], device='cuda:0')
Worker 3 has targets: tensor([0, 1, 4, 1, 3], device='cuda:0')
Worker 4 has targets: tensor([0, 4, 1, 2, 4], device='cuda:0')
Worker 5 has targets: tensor([2, 2, 4, 4, 4], device='cuda:0')
Worker 6 has targets: tensor([1, 1, 4, 4, 3], device='cuda:0')
Worker 7 has targets: tensor([4, 4, 1, 3, 0], device='cuda:0')
Worker 8 has targets: tensor([1, 3, 1, 0, 4], device='cuda:0')
Worker 9 has targets: tensor([1, 3, 3, 3, 1], device='cuda:0')
Worker 10 has targets: tensor([8, 9, 7, 7, 9], device='cuda:0')
Worker 11 has targets: tensor([8, 9, 6, 6, 7], device='cuda:0')
Worker 12 has targets: tensor([8, 6, 5, 7, 8], device='cuda:0')
Worker 13 has targets: tensor([7, 6, 9, 6, 5], device='cuda:0')
Worker 14 has targets: tensor([8, 5, 8, 6, 7], device='cuda:0')
Worker 15 has targets: tensor([9, 5, 6, 8, 6], device='cuda:0')
Worker 16 has targets: tensor([7, 7, 8, 5, 8], device='cuda:0')
Worker 17 has targets: tensor([9, 7, 5, 6, 6], device='cuda:0')
Worker 18 has targets: tensor([7, 7, 7, 6, 6], device='cuda:0')
Worker 19 has targets: tensor([5, 7, 9, 9, 7], device='cuda:0')
Worker 20 has targets: tensor([3, 2, 2, 4, 3], device='cuda:0')
Worker 21 has targets: tensor([8, 9, 7, 7, 9], device='cuda:0')


[E 1B10 |   7744/60000 ( 13%) ] Loss: 1.0319 top1= 62.0312
[E 1B20 |  14784/60000 ( 25%) ] Loss: 0.3980 top1= 88.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.7162 top1= 79.2969


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8859 top1= 49.2989


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.6784 top1= 44.2508

Train epoch 2
[E 2B0  |    704/60000 (  1%) ] Loss: 0.2786 top1= 91.0938
[E 2B10 |   7744/60000 ( 13%) ] Loss: 0.1889 top1= 93.7500
[E 2B20 |  14784/60000 ( 25%) ] Loss: 0.1961 top1= 94.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.6529 top1= 84.8658


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4374 top1= 49.8498


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4440 top1= 45.8233

Train epoch 3
[E 3B0  |    704/60000 (  1%) ] Loss: 0.1613 top1= 95.0000
[E 3B10 |   7744/60000 ( 13%) ] Loss: 0.1085 top1= 96.7188
[E 3B20 |  14784/60000 ( 25%) ] Loss: 0.1216 top1= 96.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5811 top1= 86.2480


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3462 top1= 50.1102


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5489 top1= 46.1939

Train epoch 4
[E 4B0  |    704/60000 (  1%) ] Loss: 0.1176 top1= 96.5625
[E 4B10 |   7744/60000 ( 13%) ] Loss: 0.0790 top1= 98.1250
[E 4B20 |  14784/60000 ( 25%) ] Loss: 0.0792 top1= 97.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.5199 top1= 87.0693


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4057 top1= 50.3205


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5378 top1= 46.4643

Train epoch 5
[E 5B0  |    704/60000 (  1%) ] Loss: 0.0782 top1= 98.4375
[E 5B10 |   7744/60000 ( 13%) ] Loss: 0.0579 top1= 98.5938
[E 5B20 |  14784/60000 ( 25%) ] Loss: 0.0555 top1= 98.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4695 top1= 87.0092


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5482 top1= 50.4507


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6205 top1= 46.5345

Train epoch 6
[E 6B0  |    704/60000 (  1%) ] Loss: 0.0593 top1= 99.0625
[E 6B10 |   7744/60000 ( 13%) ] Loss: 0.0425 top1= 98.7500
[E 6B20 |  14784/60000 ( 25%) ] Loss: 0.0399 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.4300 top1= 87.7103


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.6305 top1= 50.5509


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5765 top1= 46.7849

Train epoch 7
[E 7B0  |    704/60000 (  1%) ] Loss: 0.0393 top1= 99.3750
[E 7B10 |   7744/60000 ( 13%) ] Loss: 0.0256 top1= 99.5312
[E 7B20 |  14784/60000 ( 25%) ] Loss: 0.0313 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3951 top1= 88.8421


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7514 top1= 50.5409


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4490 top1= 46.9050

Train epoch 8
[E 8B0  |    704/60000 (  1%) ] Loss: 0.0282 top1= 99.5312
[E 8B10 |   7744/60000 ( 13%) ] Loss: 0.0213 top1= 99.6875
[E 8B20 |  14784/60000 ( 25%) ] Loss: 0.0343 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3738 top1= 89.1226


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8149 top1= 50.5409


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3113 top1= 46.7147

Train epoch 9
[E 9B0  |    704/60000 (  1%) ] Loss: 0.0253 top1= 99.2188
[E 9B10 |   7744/60000 ( 13%) ] Loss: 0.0187 top1= 99.8438
[E 9B20 |  14784/60000 ( 25%) ] Loss: 0.0209 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3659 top1= 88.4515


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7741 top1= 50.5308


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1939 top1= 46.9852

Train epoch 10
[E10B0  |    704/60000 (  1%) ] Loss: 0.0187 top1= 99.5312
[E10B10 |   7744/60000 ( 13%) ] Loss: 0.0203 top1= 99.3750
[E10B20 |  14784/60000 ( 25%) ] Loss: 0.0164 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3454 top1= 89.0425


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5256 top1= 50.6010


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3877 top1= 46.9151

Train epoch 11
[E11B0  |    704/60000 (  1%) ] Loss: 0.0176 top1= 99.5312
[E11B10 |   7744/60000 ( 13%) ] Loss: 0.0094 top1=100.0000
[E11B20 |  14784/60000 ( 25%) ] Loss: 0.0190 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3429 top1= 88.8822


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2150 top1= 50.6611


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4716 top1= 46.9852

Train epoch 12
[E12B0  |    704/60000 (  1%) ] Loss: 0.0104 top1=100.0000
[E12B10 |   7744/60000 ( 13%) ] Loss: 0.0106 top1= 99.8438
[E12B20 |  14784/60000 ( 25%) ] Loss: 0.0127 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3403 top1= 89.0625


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1290 top1= 50.6010


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3589 top1= 47.0853

Train epoch 13
[E13B0  |    704/60000 (  1%) ] Loss: 0.0093 top1=100.0000
[E13B10 |   7744/60000 ( 13%) ] Loss: 0.0115 top1= 99.8438
[E13B20 |  14784/60000 ( 25%) ] Loss: 0.0128 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3125 top1= 89.8237


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.8456 top1= 50.6811


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1686 top1= 47.1855

Train epoch 14
[E14B0  |    704/60000 (  1%) ] Loss: 0.0046 top1=100.0000
[E14B10 |   7744/60000 ( 13%) ] Loss: 0.0040 top1=100.0000
[E14B20 |  14784/60000 ( 25%) ] Loss: 0.0071 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3041 top1= 90.1743


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.6733 top1= 50.7011


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9273 top1= 47.0453

Train epoch 15
[E15B0  |    704/60000 (  1%) ] Loss: 0.0069 top1=100.0000
[E15B10 |   7744/60000 ( 13%) ] Loss: 0.0038 top1=100.0000
[E15B20 |  14784/60000 ( 25%) ] Loss: 0.0057 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.3096 top1= 89.7937


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5602 top1= 50.7011


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9425 top1= 47.1655

Train epoch 16
[E16B0  |    704/60000 (  1%) ] Loss: 0.0033 top1=100.0000
[E16B10 |   7744/60000 ( 13%) ] Loss: 0.0026 top1=100.0000
[E16B20 |  14784/60000 ( 25%) ] Loss: 0.0040 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2974 top1= 90.2544


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5095 top1= 50.7111


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8490 top1= 47.1855

Train epoch 17
[E17B0  |    704/60000 (  1%) ] Loss: 0.0030 top1=100.0000
[E17B10 |   7744/60000 ( 13%) ] Loss: 0.0025 top1=100.0000
[E17B20 |  14784/60000 ( 25%) ] Loss: 0.0035 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2892 top1= 90.5549


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.4425 top1= 50.7612


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7181 top1= 47.1354

Train epoch 18
[E18B0  |    704/60000 (  1%) ] Loss: 0.0027 top1=100.0000
[E18B10 |   7744/60000 ( 13%) ] Loss: 0.0023 top1=100.0000
[E18B20 |  14784/60000 ( 25%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2865 top1= 90.5349


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.3131 top1= 50.8514


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5447 top1= 47.1755

Train epoch 19
[E19B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E19B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E19B20 |  14784/60000 ( 25%) ] Loss: 0.0032 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2831 top1= 90.6651


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.1476 top1= 51.0917


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.3361 top1= 47.2155

Train epoch 20
[E20B0  |    704/60000 (  1%) ] Loss: 0.0025 top1=100.0000
[E20B10 |   7744/60000 ( 13%) ] Loss: 0.0023 top1=100.0000
[E20B20 |  14784/60000 ( 25%) ] Loss: 0.0032 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2794 top1= 90.8754


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.9882 top1= 51.3922


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.1283 top1= 47.3057

Train epoch 21
[E21B0  |    704/60000 (  1%) ] Loss: 0.0025 top1=100.0000
[E21B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E21B20 |  14784/60000 ( 25%) ] Loss: 0.0032 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2755 top1= 90.9655


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.8279 top1= 51.7027


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.9411 top1= 47.5561

Train epoch 22
[E22B0  |    704/60000 (  1%) ] Loss: 0.0025 top1=100.0000
[E22B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E22B20 |  14784/60000 ( 25%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2720 top1= 91.0657


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.6878 top1= 52.1134


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.7658 top1= 47.9467

Train epoch 23
[E23B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E23B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E23B20 |  14784/60000 ( 25%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2684 top1= 91.1659


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.5570 top1= 52.5942


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.6132 top1= 48.4575

Train epoch 24
[E24B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E24B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E24B20 |  14784/60000 ( 25%) ] Loss: 0.0034 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2657 top1= 91.2760


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.4458 top1= 53.1851


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.4802 top1= 48.9083

Train epoch 25
[E25B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E25B10 |   7744/60000 ( 13%) ] Loss: 0.0025 top1=100.0000
[E25B20 |  14784/60000 ( 25%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2638 top1= 91.4363


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.3491 top1= 53.6558


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.3726 top1= 49.5292

Train epoch 26
[E26B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E26B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E26B20 |  14784/60000 ( 25%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2623 top1= 91.5164


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.2672 top1= 54.3570


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.2846 top1= 50.0401

Train epoch 27
[E27B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E27B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E27B20 |  14784/60000 ( 25%) ] Loss: 0.0032 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2614 top1= 91.5465


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.1938 top1= 55.0280


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.1971 top1= 50.6711

Train epoch 28
[E28B0  |    704/60000 (  1%) ] Loss: 0.0026 top1=100.0000
[E28B10 |   7744/60000 ( 13%) ] Loss: 0.0024 top1=100.0000
[E28B20 |  14784/60000 ( 25%) ] Loss: 0.0032 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2601 top1= 91.6166


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.1287 top1= 55.6591


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.1178 top1= 51.2520

Train epoch 29
[E29B0  |    704/60000 (  1%) ] Loss: 0.0025 top1=100.0000
[E29B10 |   7744/60000 ( 13%) ] Loss: 0.0023 top1=100.0000
[E29B20 |  14784/60000 ( 25%) ] Loss: 0.0031 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2597 top1= 91.6567


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.0636 top1= 56.1198


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.0366 top1= 51.9732

Train epoch 30
[E30B0  |    704/60000 (  1%) ] Loss: 0.0025 top1=100.0000
[E30B10 |   7744/60000 ( 13%) ] Loss: 0.0023 top1=100.0000
[E30B20 |  14784/60000 ( 25%) ] Loss: 0.0030 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=0.2590 top1= 91.7668


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.0059 top1= 56.7408


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=2.9657 top1= 52.6943

