
=== 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=10,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=10,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=10,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=10,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=10,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=10,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=10,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=10,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=10,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=10,rank=9,shuffle=True)'}
=> Add worker SGDMWorker(index=9, momentum=0.9)

=== Start adding graph ===
<codes.graph_utils.Dumbbell object at 0x7efd31fd6730>

{'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  |    640/50000 (  1%) ] Loss: 2.3035 top1=  9.2188

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



=== Log mixing matrix @ E1B0 ===
[[0.167 0.167 0.167 0.167 0.167 0.    0.    0.    0.    0.167]
 [0.167 0.333 0.167 0.167 0.167 0.    0.    0.    0.    0.   ]
 [0.167 0.167 0.333 0.167 0.167 0.    0.    0.    0.    0.   ]
 [0.167 0.167 0.167 0.333 0.167 0.    0.    0.    0.    0.   ]
 [0.167 0.167 0.167 0.167 0.333 0.    0.    0.    0.    0.   ]
 [0.    0.    0.    0.    0.    0.333 0.167 0.167 0.167 0.167]
 [0.    0.    0.    0.    0.    0.167 0.333 0.167 0.167 0.167]
 [0.    0.    0.    0.    0.    0.167 0.167 0.333 0.167 0.167]
 [0.    0.    0.    0.    0.    0.167 0.167 0.167 0.333 0.167]
 [0.167 0.    0.    0.    0.    0.167 0.167 0.167 0.167 0.167]]


[E 1B10 |   7040/50000 ( 14%) ] Loss: 2.0253 top1= 19.3750
[E 1B20 |  13440/50000 ( 27%) ] Loss: 1.8936 top1= 18.1250
[E 1B30 |  19840/50000 ( 40%) ] Loss: 1.7005 top1= 21.4062
[E 1B40 |  26240/50000 ( 52%) ] Loss: 1.7330 top1= 20.1562
[E 1B50 |  32640/50000 ( 65%) ] Loss: 1.5875 top1= 24.8438
[E 1B60 |  39040/50000 ( 78%) ] Loss: 1.5678 top1= 25.9375
[E 1B70 |  45440/50000 ( 91%) ] Loss: 1.6326 top1= 24.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3063 top1= 11.1378


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.5849 top1= 10.0160


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.0473 top1= 16.3862

Train epoch 2
[E 2B0  |    640/50000 (  1%) ] Loss: 1.6956 top1= 24.3750
[E 2B10 |   7040/50000 ( 14%) ] Loss: 1.5580 top1= 27.9688
[E 2B20 |  13440/50000 ( 27%) ] Loss: 1.5918 top1= 29.3750
[E 2B30 |  19840/50000 ( 40%) ] Loss: 1.4849 top1= 35.1562
[E 2B40 |  26240/50000 ( 52%) ] Loss: 1.6957 top1= 24.6875
[E 2B50 |  32640/50000 ( 65%) ] Loss: 1.6005 top1= 26.2500
[E 2B60 |  39040/50000 ( 78%) ] Loss: 1.6543 top1= 24.2188
[E 2B70 |  45440/50000 ( 91%) ] Loss: 1.6086 top1= 27.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3488 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.9743 top1= 10.0060


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.1320 top1= 10.0260

Train epoch 3
[E 3B0  |    640/50000 (  1%) ] Loss: 1.7164 top1= 21.5625
[E 3B10 |   7040/50000 ( 14%) ] Loss: 1.5814 top1= 25.7812
[E 3B20 |  13440/50000 ( 27%) ] Loss: 1.6178 top1= 26.5625
[E 3B30 |  19840/50000 ( 40%) ] Loss: 1.5616 top1= 26.2500
[E 3B40 |  26240/50000 ( 52%) ] Loss: 1.5146 top1= 30.1562
[E 3B50 |  32640/50000 ( 65%) ] Loss: 1.4878 top1= 29.2188
[E 3B60 |  39040/50000 ( 78%) ] Loss: 1.4858 top1= 33.9062
[E 3B70 |  45440/50000 ( 91%) ] Loss: 1.4384 top1= 34.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3767 top1= 18.0689


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0472 top1= 18.4395


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6957 top1= 19.7316

Train epoch 4
[E 4B0  |    640/50000 (  1%) ] Loss: 1.4657 top1= 34.3750
[E 4B10 |   7040/50000 ( 14%) ] Loss: 1.4129 top1= 37.5000
[E 4B20 |  13440/50000 ( 27%) ] Loss: 1.6164 top1= 25.6250
[E 4B30 |  19840/50000 ( 40%) ] Loss: 1.5261 top1= 30.9375
[E 4B40 |  26240/50000 ( 52%) ] Loss: 1.4263 top1= 36.4062
[E 4B50 |  32640/50000 ( 65%) ] Loss: 1.4113 top1= 37.8125
[E 4B60 |  39040/50000 ( 78%) ] Loss: 1.5765 top1= 24.3750
[E 4B70 |  45440/50000 ( 91%) ] Loss: 1.4289 top1= 35.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.8764 top1= 10.6571


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.3351 top1= 19.7716


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0506 top1= 20.5929

Train epoch 5
[E 5B0  |    640/50000 (  1%) ] Loss: 1.3912 top1= 37.9688
[E 5B10 |   7040/50000 ( 14%) ] Loss: 1.3970 top1= 36.8750
[E 5B20 |  13440/50000 ( 27%) ] Loss: 1.3165 top1= 44.5312
[E 5B30 |  19840/50000 ( 40%) ] Loss: 1.3222 top1= 40.7812
[E 5B40 |  26240/50000 ( 52%) ] Loss: 1.2545 top1= 43.5938
[E 5B50 |  32640/50000 ( 65%) ] Loss: 1.2777 top1= 40.3125
[E 5B60 |  39040/50000 ( 78%) ] Loss: 1.3173 top1= 44.8438
[E 5B70 |  45440/50000 ( 91%) ] Loss: 1.2072 top1= 46.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6024 top1= 20.9435


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.2053 top1= 18.3193


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5915 top1= 21.1639

Train epoch 6
[E 6B0  |    640/50000 (  1%) ] Loss: 1.3303 top1= 39.2188
[E 6B10 |   7040/50000 ( 14%) ] Loss: 1.2404 top1= 47.6562
[E 6B20 |  13440/50000 ( 27%) ] Loss: 1.1870 top1= 46.8750
[E 6B30 |  19840/50000 ( 40%) ] Loss: 1.2037 top1= 49.8438
[E 6B40 |  26240/50000 ( 52%) ] Loss: 1.1714 top1= 50.9375
[E 6B50 |  32640/50000 ( 65%) ] Loss: 1.0871 top1= 55.4688
[E 6B60 |  39040/50000 ( 78%) ] Loss: 1.1692 top1= 56.4062
[E 6B70 |  45440/50000 ( 91%) ] Loss: 1.0925 top1= 55.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1752 top1= 29.6975


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3863 top1= 26.6827


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.1105 top1= 29.8778

Train epoch 7
[E 7B0  |    640/50000 (  1%) ] Loss: 1.1669 top1= 50.0000
[E 7B10 |   7040/50000 ( 14%) ] Loss: 1.0368 top1= 58.9062
[E 7B20 |  13440/50000 ( 27%) ] Loss: 1.0586 top1= 56.8750
[E 7B30 |  19840/50000 ( 40%) ] Loss: 1.0422 top1= 58.1250
[E 7B40 |  26240/50000 ( 52%) ] Loss: 1.0430 top1= 57.6562
[E 7B50 |  32640/50000 ( 65%) ] Loss: 1.0155 top1= 59.2188
[E 7B60 |  39040/50000 ( 78%) ] Loss: 1.0029 top1= 60.4688
[E 7B70 |  45440/50000 ( 91%) ] Loss: 0.9932 top1= 57.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5742 top1= 34.7155


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4866 top1= 29.6875


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8858 top1= 32.5821

Train epoch 8
[E 8B0  |    640/50000 (  1%) ] Loss: 0.9919 top1= 60.6250
[E 8B10 |   7040/50000 ( 14%) ] Loss: 0.9822 top1= 61.2500
[E 8B20 |  13440/50000 ( 27%) ] Loss: 1.0057 top1= 63.2812
[E 8B30 |  19840/50000 ( 40%) ] Loss: 0.9613 top1= 60.0000
[E 8B40 |  26240/50000 ( 52%) ] Loss: 0.9008 top1= 62.9688
[E 8B50 |  32640/50000 ( 65%) ] Loss: 0.9690 top1= 63.1250
[E 8B60 |  39040/50000 ( 78%) ] Loss: 0.9153 top1= 63.4375
[E 8B70 |  45440/50000 ( 91%) ] Loss: 0.9044 top1= 63.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1769 top1= 35.1663


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5549 top1= 29.8277


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0674 top1= 35.9375

Train epoch 9
[E 9B0  |    640/50000 (  1%) ] Loss: 0.9743 top1= 62.9688
[E 9B10 |   7040/50000 ( 14%) ] Loss: 0.9060 top1= 64.6875
[E 9B20 |  13440/50000 ( 27%) ] Loss: 0.9373 top1= 65.0000
[E 9B30 |  19840/50000 ( 40%) ] Loss: 0.8712 top1= 65.0000
[E 9B40 |  26240/50000 ( 52%) ] Loss: 0.8498 top1= 66.5625
[E 9B50 |  32640/50000 ( 65%) ] Loss: 0.8676 top1= 66.8750
[E 9B60 |  39040/50000 ( 78%) ] Loss: 0.8164 top1= 67.8125
[E 9B70 |  45440/50000 ( 91%) ] Loss: 0.8562 top1= 66.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9856 top1= 40.7552


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3271 top1= 31.1198


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.9564 top1= 36.1278

Train epoch 10
[E10B0  |    640/50000 (  1%) ] Loss: 0.9512 top1= 62.9688
[E10B10 |   7040/50000 ( 14%) ] Loss: 0.8509 top1= 66.4062
[E10B20 |  13440/50000 ( 27%) ] Loss: 0.8316 top1= 67.5000
[E10B30 |  19840/50000 ( 40%) ] Loss: 0.8299 top1= 68.5938
[E10B40 |  26240/50000 ( 52%) ] Loss: 0.8018 top1= 68.9062
[E10B50 |  32640/50000 ( 65%) ] Loss: 0.8517 top1= 67.6562
[E10B60 |  39040/50000 ( 78%) ] Loss: 0.7617 top1= 72.8125
[E10B70 |  45440/50000 ( 91%) ] Loss: 0.7995 top1= 68.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0404 top1= 42.9788


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4004 top1= 33.3233


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6472 top1= 34.2648

Train epoch 11
[E11B0  |    640/50000 (  1%) ] Loss: 0.8842 top1= 66.4062
[E11B10 |   7040/50000 ( 14%) ] Loss: 0.8076 top1= 69.8438
[E11B20 |  13440/50000 ( 27%) ] Loss: 0.7087 top1= 73.7500
[E11B30 |  19840/50000 ( 40%) ] Loss: 0.7157 top1= 73.7500
[E11B40 |  26240/50000 ( 52%) ] Loss: 0.7546 top1= 72.0312
[E11B50 |  32640/50000 ( 65%) ] Loss: 0.8042 top1= 70.9375
[E11B60 |  39040/50000 ( 78%) ] Loss: 0.7565 top1= 71.2500
[E11B70 |  45440/50000 ( 91%) ] Loss: 0.7555 top1= 71.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9844 top1= 44.7416


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4998 top1= 33.2833


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7171 top1= 38.6318

Train epoch 12
[E12B0  |    640/50000 (  1%) ] Loss: 0.7745 top1= 69.0625
[E12B10 |   7040/50000 ( 14%) ] Loss: 0.7198 top1= 72.3438
[E12B20 |  13440/50000 ( 27%) ] Loss: 0.6930 top1= 74.5312
[E12B30 |  19840/50000 ( 40%) ] Loss: 0.7022 top1= 73.7500
[E12B40 |  26240/50000 ( 52%) ] Loss: 0.7442 top1= 71.0938
[E12B50 |  32640/50000 ( 65%) ] Loss: 0.7241 top1= 74.3750
[E12B60 |  39040/50000 ( 78%) ] Loss: 0.7077 top1= 73.2812
[E12B70 |  45440/50000 ( 91%) ] Loss: 0.6765 top1= 73.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8076 top1= 48.0769


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9806 top1= 32.6122


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3698 top1= 38.2812

Train epoch 13
[E13B0  |    640/50000 (  1%) ] Loss: 0.8278 top1= 68.1250
[E13B10 |   7040/50000 ( 14%) ] Loss: 0.6979 top1= 74.0625
[E13B20 |  13440/50000 ( 27%) ] Loss: 0.7443 top1= 72.3438
[E13B30 |  19840/50000 ( 40%) ] Loss: 0.6461 top1= 74.2188
[E13B40 |  26240/50000 ( 52%) ] Loss: 0.6459 top1= 75.9375
[E13B50 |  32640/50000 ( 65%) ] Loss: 0.6857 top1= 75.1562
[E13B60 |  39040/50000 ( 78%) ] Loss: 0.6517 top1= 76.4062
[E13B70 |  45440/50000 ( 91%) ] Loss: 0.6662 top1= 72.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7101 top1= 51.0717


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0005 top1= 35.2063


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1621 top1= 40.1342

Train epoch 14
[E14B0  |    640/50000 (  1%) ] Loss: 0.6798 top1= 73.5938
[E14B10 |   7040/50000 ( 14%) ] Loss: 0.6860 top1= 75.4688
[E14B20 |  13440/50000 ( 27%) ] Loss: 0.6431 top1= 76.5625
[E14B30 |  19840/50000 ( 40%) ] Loss: 0.6102 top1= 76.8750
[E14B40 |  26240/50000 ( 52%) ] Loss: 0.6549 top1= 74.5312
[E14B50 |  32640/50000 ( 65%) ] Loss: 0.6413 top1= 77.6562
[E14B60 |  39040/50000 ( 78%) ] Loss: 0.5805 top1= 79.2188
[E14B70 |  45440/50000 ( 91%) ] Loss: 0.6225 top1= 77.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6594 top1= 53.3854


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9578 top1= 36.1278


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2498 top1= 40.7352

Train epoch 15
[E15B0  |    640/50000 (  1%) ] Loss: 0.6554 top1= 74.6875
[E15B10 |   7040/50000 ( 14%) ] Loss: 0.6753 top1= 75.6250
[E15B20 |  13440/50000 ( 27%) ] Loss: 0.6010 top1= 77.1875
[E15B30 |  19840/50000 ( 40%) ] Loss: 0.6474 top1= 75.7812
[E15B40 |  26240/50000 ( 52%) ] Loss: 0.6279 top1= 76.8750
[E15B50 |  32640/50000 ( 65%) ] Loss: 0.6215 top1= 77.0312
[E15B60 |  39040/50000 ( 78%) ] Loss: 0.5796 top1= 78.4375
[E15B70 |  45440/50000 ( 91%) ] Loss: 0.5652 top1= 79.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6316 top1= 53.6558


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1284 top1= 36.4183


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1784 top1= 40.8754

Train epoch 16
[E16B0  |    640/50000 (  1%) ] Loss: 0.6160 top1= 78.1250
[E16B10 |   7040/50000 ( 14%) ] Loss: 0.6426 top1= 76.2500
[E16B20 |  13440/50000 ( 27%) ] Loss: 0.5523 top1= 80.1562
[E16B30 |  19840/50000 ( 40%) ] Loss: 0.5768 top1= 78.9062
[E16B40 |  26240/50000 ( 52%) ] Loss: 0.5610 top1= 78.4375
[E16B50 |  32640/50000 ( 65%) ] Loss: 0.6006 top1= 79.5312
[E16B60 |  39040/50000 ( 78%) ] Loss: 0.5994 top1= 77.0312
[E16B70 |  45440/50000 ( 91%) ] Loss: 0.6166 top1= 75.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6916 top1= 54.2869


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4865 top1= 37.3998


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0903 top1= 41.8369

Train epoch 17
[E17B0  |    640/50000 (  1%) ] Loss: 0.6063 top1= 77.0312
[E17B10 |   7040/50000 ( 14%) ] Loss: 0.6059 top1= 77.1875
[E17B20 |  13440/50000 ( 27%) ] Loss: 0.5403 top1= 80.9375
[E17B30 |  19840/50000 ( 40%) ] Loss: 0.5647 top1= 80.9375
[E17B40 |  26240/50000 ( 52%) ] Loss: 0.5545 top1= 80.9375
[E17B50 |  32640/50000 ( 65%) ] Loss: 0.5821 top1= 78.7500
[E17B60 |  39040/50000 ( 78%) ] Loss: 0.5396 top1= 80.3125
[E17B70 |  45440/50000 ( 91%) ] Loss: 0.5666 top1= 79.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6748 top1= 53.7861


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1212 top1= 37.2596


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6519 top1= 41.7969

Train epoch 18
[E18B0  |    640/50000 (  1%) ] Loss: 0.5915 top1= 77.8125
[E18B10 |   7040/50000 ( 14%) ] Loss: 0.6276 top1= 76.8750
[E18B20 |  13440/50000 ( 27%) ] Loss: 0.4885 top1= 80.9375
[E18B30 |  19840/50000 ( 40%) ] Loss: 0.5323 top1= 79.0625
[E18B40 |  26240/50000 ( 52%) ] Loss: 0.5081 top1= 82.5000
[E18B50 |  32640/50000 ( 65%) ] Loss: 0.5987 top1= 78.1250
[E18B60 |  39040/50000 ( 78%) ] Loss: 0.4832 top1= 83.2812
[E18B70 |  45440/50000 ( 91%) ] Loss: 0.5298 top1= 80.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5616 top1= 57.3017


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4159 top1= 38.0709


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3831 top1= 42.4079

Train epoch 19
[E19B0  |    640/50000 (  1%) ] Loss: 0.5419 top1= 77.8125
[E19B10 |   7040/50000 ( 14%) ] Loss: 0.6131 top1= 79.8438
[E19B20 |  13440/50000 ( 27%) ] Loss: 0.4721 top1= 82.1875
[E19B30 |  19840/50000 ( 40%) ] Loss: 0.5051 top1= 81.4062
[E19B40 |  26240/50000 ( 52%) ] Loss: 0.4990 top1= 81.5625
[E19B50 |  32640/50000 ( 65%) ] Loss: 0.4658 top1= 83.1250
[E19B60 |  39040/50000 ( 78%) ] Loss: 0.4688 top1= 82.5000
[E19B70 |  45440/50000 ( 91%) ] Loss: 0.4570 top1= 84.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2245 top1= 51.9932


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5832 top1= 37.5200


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7174 top1= 40.2143

Train epoch 20
[E20B0  |    640/50000 (  1%) ] Loss: 0.5851 top1= 76.7188
[E20B10 |   7040/50000 ( 14%) ] Loss: 0.4877 top1= 82.0312
[E20B20 |  13440/50000 ( 27%) ] Loss: 0.4538 top1= 83.1250
[E20B30 |  19840/50000 ( 40%) ] Loss: 0.4825 top1= 83.2812
[E20B40 |  26240/50000 ( 52%) ] Loss: 0.4487 top1= 83.1250
[E20B50 |  32640/50000 ( 65%) ] Loss: 0.4673 top1= 82.1875
[E20B60 |  39040/50000 ( 78%) ] Loss: 0.4763 top1= 82.1875
[E20B70 |  45440/50000 ( 91%) ] Loss: 0.4649 top1= 84.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9719 top1= 54.0465


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9226 top1= 38.7320


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3172 top1= 42.9287

Train epoch 21
[E21B0  |    640/50000 (  1%) ] Loss: 0.5227 top1= 80.3125
[E21B10 |   7040/50000 ( 14%) ] Loss: 0.4894 top1= 81.5625
[E21B20 |  13440/50000 ( 27%) ] Loss: 0.4540 top1= 83.4375
[E21B30 |  19840/50000 ( 40%) ] Loss: 0.4528 top1= 83.4375
[E21B40 |  26240/50000 ( 52%) ] Loss: 0.4168 top1= 84.3750
[E21B50 |  32640/50000 ( 65%) ] Loss: 0.5084 top1= 80.6250
[E21B60 |  39040/50000 ( 78%) ] Loss: 0.4746 top1= 82.5000
[E21B70 |  45440/50000 ( 91%) ] Loss: 0.4523 top1= 84.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6299 top1= 58.2332


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5099 top1= 38.9824


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5176 top1= 41.9171

Train epoch 22
[E22B0  |    640/50000 (  1%) ] Loss: 0.5156 top1= 81.7188
[E22B10 |   7040/50000 ( 14%) ] Loss: 0.4885 top1= 81.8750
[E22B20 |  13440/50000 ( 27%) ] Loss: 0.4560 top1= 82.1875
[E22B30 |  19840/50000 ( 40%) ] Loss: 0.4593 top1= 83.2812
[E22B40 |  26240/50000 ( 52%) ] Loss: 0.4355 top1= 82.6562
[E22B50 |  32640/50000 ( 65%) ] Loss: 0.4334 top1= 84.2188
[E22B60 |  39040/50000 ( 78%) ] Loss: 0.4584 top1= 82.8125
[E22B70 |  45440/50000 ( 91%) ] Loss: 0.4019 top1= 84.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7040 top1= 57.5421


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1381 top1= 39.5833


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3489 top1= 43.3193

Train epoch 23
[E23B0  |    640/50000 (  1%) ] Loss: 0.4663 top1= 82.1875
[E23B10 |   7040/50000 ( 14%) ] Loss: 0.4274 top1= 85.9375
[E23B20 |  13440/50000 ( 27%) ] Loss: 0.4012 top1= 84.0625
[E23B30 |  19840/50000 ( 40%) ] Loss: 0.4084 top1= 85.9375
[E23B40 |  26240/50000 ( 52%) ] Loss: 0.3939 top1= 85.7812
[E23B50 |  32640/50000 ( 65%) ] Loss: 0.4307 top1= 84.0625
[E23B60 |  39040/50000 ( 78%) ] Loss: 0.4323 top1= 82.6562
[E23B70 |  45440/50000 ( 91%) ] Loss: 0.4489 top1= 82.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6186 top1= 58.7440


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7729 top1= 39.7436


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0220 top1= 42.8886

Train epoch 24
[E24B0  |    640/50000 (  1%) ] Loss: 0.4560 top1= 83.9062
[E24B10 |   7040/50000 ( 14%) ] Loss: 0.4458 top1= 84.6875
[E24B20 |  13440/50000 ( 27%) ] Loss: 0.4350 top1= 83.4375
[E24B30 |  19840/50000 ( 40%) ] Loss: 0.3880 top1= 85.6250
[E24B40 |  26240/50000 ( 52%) ] Loss: 0.3868 top1= 85.1562
[E24B50 |  32640/50000 ( 65%) ] Loss: 0.3981 top1= 85.3125
[E24B60 |  39040/50000 ( 78%) ] Loss: 0.4289 top1= 83.7500
[E24B70 |  45440/50000 ( 91%) ] Loss: 0.3957 top1= 84.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6055 top1= 59.4050


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7450 top1= 39.5633


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7478 top1= 43.4896

Train epoch 25
[E25B0  |    640/50000 (  1%) ] Loss: 0.3931 top1= 83.9062
[E25B10 |   7040/50000 ( 14%) ] Loss: 0.4233 top1= 85.3125
[E25B20 |  13440/50000 ( 27%) ] Loss: 0.3594 top1= 87.0312
[E25B30 |  19840/50000 ( 40%) ] Loss: 0.3819 top1= 86.7188
[E25B40 |  26240/50000 ( 52%) ] Loss: 0.4005 top1= 85.1562
[E25B50 |  32640/50000 ( 65%) ] Loss: 0.4301 top1= 84.2188
[E25B60 |  39040/50000 ( 78%) ] Loss: 0.4657 top1= 83.2812
[E25B70 |  45440/50000 ( 91%) ] Loss: 0.3776 top1= 86.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6632 top1= 58.2933


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4982 top1= 40.6450


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6416 top1= 43.0288

Train epoch 26
[E26B0  |    640/50000 (  1%) ] Loss: 0.3730 top1= 85.3125
[E26B10 |   7040/50000 ( 14%) ] Loss: 0.4415 top1= 83.1250
[E26B20 |  13440/50000 ( 27%) ] Loss: 0.3917 top1= 86.0938
[E26B30 |  19840/50000 ( 40%) ] Loss: 0.3857 top1= 85.4688
[E26B40 |  26240/50000 ( 52%) ] Loss: 0.3632 top1= 87.0312
[E26B50 |  32640/50000 ( 65%) ] Loss: 0.3493 top1= 87.6562
[E26B60 |  39040/50000 ( 78%) ] Loss: 0.4208 top1= 84.3750
[E26B70 |  45440/50000 ( 91%) ] Loss: 0.3353 top1= 87.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4353 top1= 56.5905


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5394 top1= 40.5950


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1874 top1= 44.0204

Train epoch 27
[E27B0  |    640/50000 (  1%) ] Loss: 0.3923 top1= 83.5938
[E27B10 |   7040/50000 ( 14%) ] Loss: 0.4191 top1= 86.0938
[E27B20 |  13440/50000 ( 27%) ] Loss: 0.3466 top1= 87.8125
[E27B30 |  19840/50000 ( 40%) ] Loss: 0.3492 top1= 86.7188
[E27B40 |  26240/50000 ( 52%) ] Loss: 0.3366 top1= 86.5625
[E27B50 |  32640/50000 ( 65%) ] Loss: 0.4059 top1= 84.0625
[E27B60 |  39040/50000 ( 78%) ] Loss: 0.3297 top1= 88.1250
[E27B70 |  45440/50000 ( 91%) ] Loss: 0.3777 top1= 86.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5534 top1= 61.1478


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4779 top1= 40.3345


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7467 top1= 43.3994

Train epoch 28
[E28B0  |    640/50000 (  1%) ] Loss: 0.3736 top1= 86.4062
[E28B10 |   7040/50000 ( 14%) ] Loss: 0.3470 top1= 87.1875
[E28B20 |  13440/50000 ( 27%) ] Loss: 0.3249 top1= 87.6562
[E28B30 |  19840/50000 ( 40%) ] Loss: 0.3555 top1= 87.1875
[E28B40 |  26240/50000 ( 52%) ] Loss: 0.3265 top1= 87.0312
[E28B50 |  32640/50000 ( 65%) ] Loss: 0.3086 top1= 88.9062
[E28B60 |  39040/50000 ( 78%) ] Loss: 0.3324 top1= 87.3438
[E28B70 |  45440/50000 ( 91%) ] Loss: 0.3474 top1= 86.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7136 top1= 60.7572


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3210 top1= 40.2244


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2110 top1= 43.8301

Train epoch 29
[E29B0  |    640/50000 (  1%) ] Loss: 0.3907 top1= 86.4062
[E29B10 |   7040/50000 ( 14%) ] Loss: 0.3667 top1= 86.8750
[E29B20 |  13440/50000 ( 27%) ] Loss: 0.3979 top1= 85.3125
[E29B30 |  19840/50000 ( 40%) ] Loss: 0.3203 top1= 89.5312
[E29B40 |  26240/50000 ( 52%) ] Loss: 0.3510 top1= 87.1875
[E29B50 |  32640/50000 ( 65%) ] Loss: 0.3828 top1= 85.4688
[E29B60 |  39040/50000 ( 78%) ] Loss: 0.3504 top1= 86.8750
[E29B70 |  45440/50000 ( 91%) ] Loss: 0.3486 top1= 87.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4152 top1= 63.3313


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2789 top1= 40.9555


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.9287 top1= 43.6899

Train epoch 30
[E30B0  |    640/50000 (  1%) ] Loss: 0.3689 top1= 86.5625
[E30B10 |   7040/50000 ( 14%) ] Loss: 0.4392 top1= 84.0625
[E30B20 |  13440/50000 ( 27%) ] Loss: 0.3766 top1= 86.5625
[E30B30 |  19840/50000 ( 40%) ] Loss: 0.3509 top1= 87.0312
[E30B40 |  26240/50000 ( 52%) ] Loss: 0.3153 top1= 89.2188
[E30B50 |  32640/50000 ( 65%) ] Loss: 0.3030 top1= 87.6562
[E30B60 |  39040/50000 ( 78%) ] Loss: 0.3369 top1= 87.5000
[E30B70 |  45440/50000 ( 91%) ] Loss: 0.2978 top1= 88.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9769 top1= 60.7171


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3856 top1= 42.5180


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1068 top1= 44.4010

Train epoch 31
[E31B0  |    640/50000 (  1%) ] Loss: 0.3594 top1= 86.4062
[E31B10 |   7040/50000 ( 14%) ] Loss: 0.3289 top1= 87.8125
[E31B20 |  13440/50000 ( 27%) ] Loss: 0.3652 top1= 85.4688
[E31B30 |  19840/50000 ( 40%) ] Loss: 0.3510 top1= 86.2500
[E31B40 |  26240/50000 ( 52%) ] Loss: 0.3501 top1= 87.9688
[E31B50 |  32640/50000 ( 65%) ] Loss: 0.3068 top1= 86.7188
[E31B60 |  39040/50000 ( 78%) ] Loss: 0.3309 top1= 87.5000
[E31B70 |  45440/50000 ( 91%) ] Loss: 0.3037 top1= 87.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3766 top1= 62.5000


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9636 top1= 41.8870


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7802 top1= 44.5713

Train epoch 32
[E32B0  |    640/50000 (  1%) ] Loss: 0.3340 top1= 87.5000
[E32B10 |   7040/50000 ( 14%) ] Loss: 0.3500 top1= 86.7188
[E32B20 |  13440/50000 ( 27%) ] Loss: 0.3148 top1= 88.7500
[E32B30 |  19840/50000 ( 40%) ] Loss: 0.3515 top1= 87.5000
[E32B40 |  26240/50000 ( 52%) ] Loss: 0.2855 top1= 89.6875
[E32B50 |  32640/50000 ( 65%) ] Loss: 0.2966 top1= 89.0625
[E32B60 |  39040/50000 ( 78%) ] Loss: 0.3080 top1= 89.0625
[E32B70 |  45440/50000 ( 91%) ] Loss: 0.3325 top1= 88.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6088 top1= 60.9475


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3610 top1= 41.5264


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4999 top1= 44.4211

Train epoch 33
[E33B0  |    640/50000 (  1%) ] Loss: 0.3100 top1= 87.5000
[E33B10 |   7040/50000 ( 14%) ] Loss: 0.3394 top1= 87.5000
[E33B20 |  13440/50000 ( 27%) ] Loss: 0.3015 top1= 89.5312
[E33B30 |  19840/50000 ( 40%) ] Loss: 0.3599 top1= 87.1875
[E33B40 |  26240/50000 ( 52%) ] Loss: 0.3349 top1= 86.5625
[E33B50 |  32640/50000 ( 65%) ] Loss: 0.2985 top1= 88.7500
[E33B60 |  39040/50000 ( 78%) ] Loss: 0.3027 top1= 88.5938
[E33B70 |  45440/50000 ( 91%) ] Loss: 0.2852 top1= 89.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7102 top1= 61.1979


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2354 top1= 40.9355


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8014 top1= 44.2808

Train epoch 34
[E34B0  |    640/50000 (  1%) ] Loss: 0.3016 top1= 88.4375
[E34B10 |   7040/50000 ( 14%) ] Loss: 0.3073 top1= 87.8125
[E34B20 |  13440/50000 ( 27%) ] Loss: 0.2773 top1= 90.7812
[E34B30 |  19840/50000 ( 40%) ] Loss: 0.2744 top1= 89.0625
[E34B40 |  26240/50000 ( 52%) ] Loss: 0.3122 top1= 89.2188
[E34B50 |  32640/50000 ( 65%) ] Loss: 0.2887 top1= 89.3750
[E34B60 |  39040/50000 ( 78%) ] Loss: 0.3034 top1= 88.7500
[E34B70 |  45440/50000 ( 91%) ] Loss: 0.2302 top1= 90.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6559 top1= 60.8774


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1978 top1= 41.7969


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0086 top1= 44.2708

Train epoch 35
[E35B0  |    640/50000 (  1%) ] Loss: 0.2867 top1= 89.0625
[E35B10 |   7040/50000 ( 14%) ] Loss: 0.3497 top1= 86.2500
[E35B20 |  13440/50000 ( 27%) ] Loss: 0.3060 top1= 88.2812
[E35B30 |  19840/50000 ( 40%) ] Loss: 0.3143 top1= 88.1250
[E35B40 |  26240/50000 ( 52%) ] Loss: 0.2794 top1= 89.6875
[E35B50 |  32640/50000 ( 65%) ] Loss: 0.2798 top1= 88.5938
[E35B60 |  39040/50000 ( 78%) ] Loss: 0.3003 top1= 89.3750
[E35B70 |  45440/50000 ( 91%) ] Loss: 0.2668 top1= 89.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4297 top1= 65.8353


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9015 top1= 41.8670


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2321 top1= 44.9920

Train epoch 36
[E36B0  |    640/50000 (  1%) ] Loss: 0.3284 top1= 90.0000
[E36B10 |   7040/50000 ( 14%) ] Loss: 0.2744 top1= 89.3750
[E36B20 |  13440/50000 ( 27%) ] Loss: 0.2770 top1= 89.5312
[E36B30 |  19840/50000 ( 40%) ] Loss: 0.3129 top1= 88.7500
[E36B40 |  26240/50000 ( 52%) ] Loss: 0.2700 top1= 89.3750
[E36B50 |  32640/50000 ( 65%) ] Loss: 0.3190 top1= 87.8125
[E36B60 |  39040/50000 ( 78%) ] Loss: 0.2452 top1= 90.4688
[E36B70 |  45440/50000 ( 91%) ] Loss: 0.2704 top1= 90.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6753 top1= 60.6771


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6859 top1= 41.3562


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.9209 top1= 43.9704

Train epoch 37
[E37B0  |    640/50000 (  1%) ] Loss: 0.3246 top1= 87.8125
[E37B10 |   7040/50000 ( 14%) ] Loss: 0.3117 top1= 89.6875
[E37B20 |  13440/50000 ( 27%) ] Loss: 0.2565 top1= 91.4062
[E37B30 |  19840/50000 ( 40%) ] Loss: 0.2717 top1= 89.6875
[E37B40 |  26240/50000 ( 52%) ] Loss: 0.2860 top1= 89.8438
[E37B50 |  32640/50000 ( 65%) ] Loss: 0.3040 top1= 87.9688
[E37B60 |  39040/50000 ( 78%) ] Loss: 0.2842 top1= 89.0625
[E37B70 |  45440/50000 ( 91%) ] Loss: 0.2399 top1= 91.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9233 top1= 61.4984


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7816 top1= 42.0172


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1971 top1= 44.9419

Train epoch 38
[E38B0  |    640/50000 (  1%) ] Loss: 0.2709 top1= 89.5312
[E38B10 |   7040/50000 ( 14%) ] Loss: 0.3050 top1= 89.8438
[E38B20 |  13440/50000 ( 27%) ] Loss: 0.2515 top1= 91.4062
[E38B30 |  19840/50000 ( 40%) ] Loss: 0.2591 top1= 90.3125
[E38B40 |  26240/50000 ( 52%) ] Loss: 0.2319 top1= 91.7188
[E38B50 |  32640/50000 ( 65%) ] Loss: 0.2622 top1= 90.0000
[E38B60 |  39040/50000 ( 78%) ] Loss: 0.2522 top1= 91.4062
[E38B70 |  45440/50000 ( 91%) ] Loss: 0.2329 top1= 91.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0003 top1= 59.9659


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.1659 top1= 42.2376


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7740 top1= 44.7817

Train epoch 39
[E39B0  |    640/50000 (  1%) ] Loss: 0.2788 top1= 90.3125
[E39B10 |   7040/50000 ( 14%) ] Loss: 0.3385 top1= 86.8750
[E39B20 |  13440/50000 ( 27%) ] Loss: 0.2145 top1= 92.8125
[E39B30 |  19840/50000 ( 40%) ] Loss: 0.2700 top1= 90.3125
[E39B40 |  26240/50000 ( 52%) ] Loss: 0.2578 top1= 90.3125
[E39B50 |  32640/50000 ( 65%) ] Loss: 0.2328 top1= 91.2500
[E39B60 |  39040/50000 ( 78%) ] Loss: 0.2174 top1= 91.2500
[E39B70 |  45440/50000 ( 91%) ] Loss: 0.2200 top1= 91.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8058 top1= 63.9824


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6364 top1= 42.0773


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2266 top1= 44.7216

Train epoch 40
[E40B0  |    640/50000 (  1%) ] Loss: 0.2653 top1= 88.4375
[E40B10 |   7040/50000 ( 14%) ] Loss: 0.2826 top1= 89.3750
[E40B20 |  13440/50000 ( 27%) ] Loss: 0.2489 top1= 91.4062
[E40B30 |  19840/50000 ( 40%) ] Loss: 0.2277 top1= 91.7188
[E40B40 |  26240/50000 ( 52%) ] Loss: 0.2760 top1= 89.3750
[E40B50 |  32640/50000 ( 65%) ] Loss: 0.3395 top1= 88.2812
[E40B60 |  39040/50000 ( 78%) ] Loss: 0.2543 top1= 92.1875
[E40B70 |  45440/50000 ( 91%) ] Loss: 0.2583 top1= 90.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7648 top1= 64.3429


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1211 top1= 42.4679


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2582 top1= 44.2808

Train epoch 41
[E41B0  |    640/50000 (  1%) ] Loss: 0.2927 top1= 89.3750
[E41B10 |   7040/50000 ( 14%) ] Loss: 0.2762 top1= 89.8438
[E41B20 |  13440/50000 ( 27%) ] Loss: 0.2440 top1= 91.5625
[E41B30 |  19840/50000 ( 40%) ] Loss: 0.2441 top1= 91.0938
[E41B40 |  26240/50000 ( 52%) ] Loss: 0.2268 top1= 92.1875
[E41B50 |  32640/50000 ( 65%) ] Loss: 0.2649 top1= 90.1562
[E41B60 |  39040/50000 ( 78%) ] Loss: 0.2010 top1= 92.8125
[E41B70 |  45440/50000 ( 91%) ] Loss: 0.2390 top1= 90.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5651 top1= 66.1058


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8030 top1= 41.8870


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3923 top1= 44.7516

Train epoch 42
[E42B0  |    640/50000 (  1%) ] Loss: 0.2485 top1= 91.5625
[E42B10 |   7040/50000 ( 14%) ] Loss: 0.2913 top1= 89.5312
[E42B20 |  13440/50000 ( 27%) ] Loss: 0.2420 top1= 91.4062
[E42B30 |  19840/50000 ( 40%) ] Loss: 0.2456 top1= 91.8750
[E42B40 |  26240/50000 ( 52%) ] Loss: 0.2409 top1= 92.0312
[E42B50 |  32640/50000 ( 65%) ] Loss: 0.2230 top1= 91.4062
[E42B60 |  39040/50000 ( 78%) ] Loss: 0.1875 top1= 92.8125
[E42B70 |  45440/50000 ( 91%) ] Loss: 0.2185 top1= 91.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7921 top1= 62.8405


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6875 top1= 42.3077


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4580 top1= 45.6731

Train epoch 43
[E43B0  |    640/50000 (  1%) ] Loss: 0.2738 top1= 90.7812
[E43B10 |   7040/50000 ( 14%) ] Loss: 0.3351 top1= 88.7500
[E43B20 |  13440/50000 ( 27%) ] Loss: 0.2827 top1= 91.0938
[E43B30 |  19840/50000 ( 40%) ] Loss: 0.2640 top1= 90.0000
[E43B40 |  26240/50000 ( 52%) ] Loss: 0.2613 top1= 90.7812
[E43B50 |  32640/50000 ( 65%) ] Loss: 0.2553 top1= 90.0000
[E43B60 |  39040/50000 ( 78%) ] Loss: 0.2196 top1= 91.4062
[E43B70 |  45440/50000 ( 91%) ] Loss: 0.2473 top1= 91.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2766 top1= 67.2877


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.0958 top1= 42.4479


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4291 top1= 45.6631

Train epoch 44
[E44B0  |    640/50000 (  1%) ] Loss: 0.2378 top1= 91.4062
[E44B10 |   7040/50000 ( 14%) ] Loss: 0.2583 top1= 90.7812
[E44B20 |  13440/50000 ( 27%) ] Loss: 0.2444 top1= 91.2500
[E44B30 |  19840/50000 ( 40%) ] Loss: 0.2170 top1= 91.4062
[E44B40 |  26240/50000 ( 52%) ] Loss: 0.2265 top1= 91.5625
[E44B50 |  32640/50000 ( 65%) ] Loss: 0.2314 top1= 91.4062
[E44B60 |  39040/50000 ( 78%) ] Loss: 0.2109 top1= 91.2500
[E44B70 |  45440/50000 ( 91%) ] Loss: 0.2256 top1= 91.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5752 top1= 65.2945


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4162 top1= 42.2276


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6135 top1= 45.3626

Train epoch 45
[E45B0  |    640/50000 (  1%) ] Loss: 0.2085 top1= 92.0312
[E45B10 |   7040/50000 ( 14%) ] Loss: 0.2699 top1= 90.0000
[E45B20 |  13440/50000 ( 27%) ] Loss: 0.2164 top1= 92.0312
[E45B30 |  19840/50000 ( 40%) ] Loss: 0.2259 top1= 90.4688
[E45B40 |  26240/50000 ( 52%) ] Loss: 0.1934 top1= 92.3438
[E45B50 |  32640/50000 ( 65%) ] Loss: 0.2034 top1= 92.0312
[E45B60 |  39040/50000 ( 78%) ] Loss: 0.2108 top1= 92.9688
[E45B70 |  45440/50000 ( 91%) ] Loss: 0.2648 top1= 89.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4062 top1= 65.4046


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6432 top1= 43.0088


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4670 top1= 46.1138

Train epoch 46
[E46B0  |    640/50000 (  1%) ] Loss: 0.2166 top1= 92.1875
[E46B10 |   7040/50000 ( 14%) ] Loss: 0.2038 top1= 92.0312
[E46B20 |  13440/50000 ( 27%) ] Loss: 0.2120 top1= 92.8125
[E46B30 |  19840/50000 ( 40%) ] Loss: 0.2233 top1= 92.3438
[E46B40 |  26240/50000 ( 52%) ] Loss: 0.2926 top1= 90.6250
[E46B50 |  32640/50000 ( 65%) ] Loss: 0.2667 top1= 89.5312
[E46B60 |  39040/50000 ( 78%) ] Loss: 0.2136 top1= 93.5938
[E46B70 |  45440/50000 ( 91%) ] Loss: 0.2308 top1= 92.3438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4702 top1= 64.3429


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.9924 top1= 42.7684


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0987 top1= 45.4928

Train epoch 47
[E47B0  |    640/50000 (  1%) ] Loss: 0.2088 top1= 92.8125
[E47B10 |   7040/50000 ( 14%) ] Loss: 0.2923 top1= 91.2500
[E47B20 |  13440/50000 ( 27%) ] Loss: 0.2610 top1= 89.3750
[E47B30 |  19840/50000 ( 40%) ] Loss: 0.2612 top1= 90.7812
[E47B40 |  26240/50000 ( 52%) ] Loss: 0.2339 top1= 90.9375
[E47B50 |  32640/50000 ( 65%) ] Loss: 0.2036 top1= 92.0312
[E47B60 |  39040/50000 ( 78%) ] Loss: 0.2297 top1= 91.5625
[E47B70 |  45440/50000 ( 91%) ] Loss: 0.2355 top1= 92.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4223 top1= 66.7768


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8826 top1= 41.3562


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.5771 top1= 46.0036

Train epoch 48
[E48B0  |    640/50000 (  1%) ] Loss: 0.2604 top1= 90.4688
[E48B10 |   7040/50000 ( 14%) ] Loss: 0.2393 top1= 90.7812
[E48B20 |  13440/50000 ( 27%) ] Loss: 0.2130 top1= 91.2500
[E48B30 |  19840/50000 ( 40%) ] Loss: 0.2285 top1= 91.4062
[E48B40 |  26240/50000 ( 52%) ] Loss: 0.2065 top1= 92.6562
[E48B50 |  32640/50000 ( 65%) ] Loss: 0.2535 top1= 92.0312
[E48B60 |  39040/50000 ( 78%) ] Loss: 0.1845 top1= 92.9688
[E48B70 |  45440/50000 ( 91%) ] Loss: 0.2466 top1= 90.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3496 top1= 68.8902


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2094 top1= 42.3578


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7697 top1= 46.2540

Train epoch 49
[E49B0  |    640/50000 (  1%) ] Loss: 0.2149 top1= 92.8125
[E49B10 |   7040/50000 ( 14%) ] Loss: 0.1999 top1= 92.6562
[E49B20 |  13440/50000 ( 27%) ] Loss: 0.1912 top1= 92.0312
[E49B30 |  19840/50000 ( 40%) ] Loss: 0.2275 top1= 91.7188
[E49B40 |  26240/50000 ( 52%) ] Loss: 0.1885 top1= 92.8125
[E49B50 |  32640/50000 ( 65%) ] Loss: 0.2107 top1= 92.0312
[E49B60 |  39040/50000 ( 78%) ] Loss: 0.2116 top1= 92.1875
[E49B70 |  45440/50000 ( 91%) ] Loss: 0.2537 top1= 90.9375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2177 top1= 60.7372


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2077 top1= 43.9103


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.8571 top1= 45.2424

Train epoch 50
[E50B0  |    640/50000 (  1%) ] Loss: 0.2128 top1= 91.5625
[E50B10 |   7040/50000 ( 14%) ] Loss: 0.1846 top1= 93.4375
[E50B20 |  13440/50000 ( 27%) ] Loss: 0.2234 top1= 90.7812
[E50B30 |  19840/50000 ( 40%) ] Loss: 0.2372 top1= 91.0938
[E50B40 |  26240/50000 ( 52%) ] Loss: 0.1934 top1= 92.6562
[E50B50 |  32640/50000 ( 65%) ] Loss: 0.1822 top1= 92.9688
[E50B60 |  39040/50000 ( 78%) ] Loss: 0.1795 top1= 94.6875
[E50B70 |  45440/50000 ( 91%) ] Loss: 0.2284 top1= 91.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6228 top1= 65.5749


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7691 top1= 42.7784


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0580 top1= 45.3025

Train epoch 51
[E51B0  |    640/50000 (  1%) ] Loss: 0.2209 top1= 92.1875
[E51B10 |   7040/50000 ( 14%) ] Loss: 0.1927 top1= 93.1250
[E51B20 |  13440/50000 ( 27%) ] Loss: 0.2036 top1= 92.3438
[E51B30 |  19840/50000 ( 40%) ] Loss: 0.2368 top1= 91.5625
[E51B40 |  26240/50000 ( 52%) ] Loss: 0.1951 top1= 93.2812
[E51B50 |  32640/50000 ( 65%) ] Loss: 0.2269 top1= 92.1875
[E51B60 |  39040/50000 ( 78%) ] Loss: 0.1908 top1= 93.2812
[E51B70 |  45440/50000 ( 91%) ] Loss: 0.2214 top1= 91.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=3.1900 top1= 60.4567


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5151 top1= 42.5381


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.2634 top1= 46.7748

Train epoch 52
[E52B0  |    640/50000 (  1%) ] Loss: 0.2061 top1= 93.1250
[E52B10 |   7040/50000 ( 14%) ] Loss: 0.2899 top1= 89.8438
[E52B20 |  13440/50000 ( 27%) ] Loss: 0.2532 top1= 90.6250
[E52B30 |  19840/50000 ( 40%) ] Loss: 0.2033 top1= 93.2812
[E52B40 |  26240/50000 ( 52%) ] Loss: 0.1477 top1= 93.9062
[E52B50 |  32640/50000 ( 65%) ] Loss: 0.1753 top1= 92.8125
[E52B60 |  39040/50000 ( 78%) ] Loss: 0.2263 top1= 92.5000
[E52B70 |  45440/50000 ( 91%) ] Loss: 0.2299 top1= 92.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5852 top1= 65.2644


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8701 top1= 42.1675


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9562 top1= 45.0521

Train epoch 53
[E53B0  |    640/50000 (  1%) ] Loss: 0.2348 top1= 92.0312
[E53B10 |   7040/50000 ( 14%) ] Loss: 0.2424 top1= 91.5625
[E53B20 |  13440/50000 ( 27%) ] Loss: 0.1704 top1= 93.7500
[E53B30 |  19840/50000 ( 40%) ] Loss: 0.2596 top1= 90.0000
[E53B40 |  26240/50000 ( 52%) ] Loss: 0.1877 top1= 92.6562
[E53B50 |  32640/50000 ( 65%) ] Loss: 0.2192 top1= 92.9688
[E53B60 |  39040/50000 ( 78%) ] Loss: 0.2165 top1= 92.9688
[E53B70 |  45440/50000 ( 91%) ] Loss: 0.1767 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6540 top1= 63.7620


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5468 top1= 41.8269


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0840 top1= 45.5228

Train epoch 54
[E54B0  |    640/50000 (  1%) ] Loss: 0.2407 top1= 91.7188
[E54B10 |   7040/50000 ( 14%) ] Loss: 0.2682 top1= 90.6250
[E54B20 |  13440/50000 ( 27%) ] Loss: 0.1931 top1= 92.3438
[E54B30 |  19840/50000 ( 40%) ] Loss: 0.1947 top1= 93.7500
[E54B40 |  26240/50000 ( 52%) ] Loss: 0.1817 top1= 92.6562
[E54B50 |  32640/50000 ( 65%) ] Loss: 0.2064 top1= 92.1875
[E54B60 |  39040/50000 ( 78%) ] Loss: 0.1685 top1= 94.2188
[E54B70 |  45440/50000 ( 91%) ] Loss: 0.1929 top1= 91.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4998 top1= 66.1959


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4275 top1= 42.4679


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2596 top1= 46.7748

Train epoch 55
[E55B0  |    640/50000 (  1%) ] Loss: 0.2133 top1= 91.4062
[E55B10 |   7040/50000 ( 14%) ] Loss: 0.2071 top1= 92.5000
[E55B20 |  13440/50000 ( 27%) ] Loss: 0.2102 top1= 91.8750
[E55B30 |  19840/50000 ( 40%) ] Loss: 0.2132 top1= 93.5938
[E55B40 |  26240/50000 ( 52%) ] Loss: 0.2031 top1= 93.2812
[E55B50 |  32640/50000 ( 65%) ] Loss: 0.1888 top1= 93.9062
[E55B60 |  39040/50000 ( 78%) ] Loss: 0.1338 top1= 94.6875
[E55B70 |  45440/50000 ( 91%) ] Loss: 0.1742 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7339 top1= 64.3530


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.2346 top1= 42.2676


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6452 top1= 45.6130

Train epoch 56
[E56B0  |    640/50000 (  1%) ] Loss: 0.1836 top1= 93.1250
[E56B10 |   7040/50000 ( 14%) ] Loss: 0.1960 top1= 92.0312
[E56B20 |  13440/50000 ( 27%) ] Loss: 0.1987 top1= 92.8125
[E56B30 |  19840/50000 ( 40%) ] Loss: 0.1753 top1= 92.9688
[E56B40 |  26240/50000 ( 52%) ] Loss: 0.1743 top1= 94.6875
[E56B50 |  32640/50000 ( 65%) ] Loss: 0.2027 top1= 92.3438
[E56B60 |  39040/50000 ( 78%) ] Loss: 0.1816 top1= 93.5938
[E56B70 |  45440/50000 ( 91%) ] Loss: 0.2010 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7748 top1= 66.5064


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.3181 top1= 42.4379


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.5662 top1= 45.7131

Train epoch 57
[E57B0  |    640/50000 (  1%) ] Loss: 0.1504 top1= 94.5312
[E57B10 |   7040/50000 ( 14%) ] Loss: 0.1613 top1= 93.1250
[E57B20 |  13440/50000 ( 27%) ] Loss: 0.2128 top1= 92.9688
[E57B30 |  19840/50000 ( 40%) ] Loss: 0.1997 top1= 93.7500
[E57B40 |  26240/50000 ( 52%) ] Loss: 0.1846 top1= 93.4375
[E57B50 |  32640/50000 ( 65%) ] Loss: 0.2502 top1= 91.7188
[E57B60 |  39040/50000 ( 78%) ] Loss: 0.1755 top1= 94.0625
[E57B70 |  45440/50000 ( 91%) ] Loss: 0.1901 top1= 92.3438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6309 top1= 65.5749


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.0801 top1= 41.9772


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.8498 top1= 45.4427

Train epoch 58
[E58B0  |    640/50000 (  1%) ] Loss: 0.1815 top1= 94.0625
[E58B10 |   7040/50000 ( 14%) ] Loss: 0.1791 top1= 92.1875
[E58B20 |  13440/50000 ( 27%) ] Loss: 0.2295 top1= 92.3438
[E58B30 |  19840/50000 ( 40%) ] Loss: 0.2481 top1= 92.0312
[E58B40 |  26240/50000 ( 52%) ] Loss: 0.1807 top1= 93.9062
[E58B50 |  32640/50000 ( 65%) ] Loss: 0.2222 top1= 91.2500
[E58B60 |  39040/50000 ( 78%) ] Loss: 0.1613 top1= 94.6875
[E58B70 |  45440/50000 ( 91%) ] Loss: 0.2276 top1= 92.1875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9569 top1= 66.8570


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.9462 top1= 42.8085


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.2942 top1= 45.8734

Train epoch 59
[E59B0  |    640/50000 (  1%) ] Loss: 0.1778 top1= 93.7500
[E59B10 |   7040/50000 ( 14%) ] Loss: 0.2347 top1= 92.3438
[E59B20 |  13440/50000 ( 27%) ] Loss: 0.1864 top1= 93.9062
[E59B30 |  19840/50000 ( 40%) ] Loss: 0.2415 top1= 91.4062
[E59B40 |  26240/50000 ( 52%) ] Loss: 0.1756 top1= 93.5938
[E59B50 |  32640/50000 ( 65%) ] Loss: 0.2167 top1= 92.6562
[E59B60 |  39040/50000 ( 78%) ] Loss: 0.1846 top1= 92.9688
[E59B70 |  45440/50000 ( 91%) ] Loss: 0.2656 top1= 91.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6564 top1= 66.3161


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.2281 top1= 42.0573


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8350 top1= 45.3425

Train epoch 60
[E60B0  |    640/50000 (  1%) ] Loss: 0.2169 top1= 92.0312
[E60B10 |   7040/50000 ( 14%) ] Loss: 0.1826 top1= 94.0625
[E60B20 |  13440/50000 ( 27%) ] Loss: 0.1912 top1= 94.2188
[E60B30 |  19840/50000 ( 40%) ] Loss: 0.1833 top1= 92.3438
[E60B40 |  26240/50000 ( 52%) ] Loss: 0.1626 top1= 94.3750
[E60B50 |  32640/50000 ( 65%) ] Loss: 0.1941 top1= 93.1250
[E60B60 |  39040/50000 ( 78%) ] Loss: 0.1692 top1= 93.9062
[E60B70 |  45440/50000 ( 91%) ] Loss: 0.1796 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7827 top1= 67.1875


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.7991 top1= 41.2760


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4295 top1= 46.3241

Train epoch 61
[E61B0  |    640/50000 (  1%) ] Loss: 0.1831 top1= 94.3750
[E61B10 |   7040/50000 ( 14%) ] Loss: 0.1632 top1= 92.9688
[E61B20 |  13440/50000 ( 27%) ] Loss: 0.1410 top1= 94.6875
[E61B30 |  19840/50000 ( 40%) ] Loss: 0.2105 top1= 91.7188
[E61B40 |  26240/50000 ( 52%) ] Loss: 0.1655 top1= 93.5938
[E61B50 |  32640/50000 ( 65%) ] Loss: 0.1759 top1= 93.7500
[E61B60 |  39040/50000 ( 78%) ] Loss: 0.1701 top1= 92.9688
[E61B70 |  45440/50000 ( 91%) ] Loss: 0.1949 top1= 94.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4972 top1= 68.6498


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3840 top1= 42.2977


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.9068 top1= 46.0437

Train epoch 62
[E62B0  |    640/50000 (  1%) ] Loss: 0.2086 top1= 91.7188
[E62B10 |   7040/50000 ( 14%) ] Loss: 0.1843 top1= 93.1250
[E62B20 |  13440/50000 ( 27%) ] Loss: 0.2196 top1= 92.8125
[E62B30 |  19840/50000 ( 40%) ] Loss: 0.1974 top1= 92.8125
[E62B40 |  26240/50000 ( 52%) ] Loss: 0.1631 top1= 93.7500
[E62B50 |  32640/50000 ( 65%) ] Loss: 0.1774 top1= 92.9688
[E62B60 |  39040/50000 ( 78%) ] Loss: 0.1322 top1= 95.7812
[E62B70 |  45440/50000 ( 91%) ] Loss: 0.1789 top1= 94.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4684 top1= 67.9788


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.0705 top1= 42.5581


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8941 top1= 45.5829

Train epoch 63
[E63B0  |    640/50000 (  1%) ] Loss: 0.1650 top1= 93.7500
[E63B10 |   7040/50000 ( 14%) ] Loss: 0.2157 top1= 93.4375
[E63B20 |  13440/50000 ( 27%) ] Loss: 0.1651 top1= 93.5938
[E63B30 |  19840/50000 ( 40%) ] Loss: 0.2458 top1= 90.7812
[E63B40 |  26240/50000 ( 52%) ] Loss: 0.1826 top1= 94.0625
[E63B50 |  32640/50000 ( 65%) ] Loss: 0.2150 top1= 92.9688
[E63B60 |  39040/50000 ( 78%) ] Loss: 0.1425 top1= 94.3750
[E63B70 |  45440/50000 ( 91%) ] Loss: 0.1469 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4024 top1= 69.6615


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8697 top1= 42.2877


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3698 top1= 46.2740

Train epoch 64
[E64B0  |    640/50000 (  1%) ] Loss: 0.1917 top1= 92.6562
[E64B10 |   7040/50000 ( 14%) ] Loss: 0.1792 top1= 93.2812
[E64B20 |  13440/50000 ( 27%) ] Loss: 0.2001 top1= 91.7188
[E64B30 |  19840/50000 ( 40%) ] Loss: 0.1787 top1= 93.4375
[E64B40 |  26240/50000 ( 52%) ] Loss: 0.2063 top1= 92.6562
[E64B50 |  32640/50000 ( 65%) ] Loss: 0.1581 top1= 94.5312
[E64B60 |  39040/50000 ( 78%) ] Loss: 0.1970 top1= 93.9062
[E64B70 |  45440/50000 ( 91%) ] Loss: 0.1582 top1= 94.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2773 top1= 69.5813


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1667 top1= 42.5381


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4394 top1= 45.5329

Train epoch 65
[E65B0  |    640/50000 (  1%) ] Loss: 0.1856 top1= 93.9062
[E65B10 |   7040/50000 ( 14%) ] Loss: 0.2018 top1= 93.7500
[E65B20 |  13440/50000 ( 27%) ] Loss: 0.1674 top1= 94.5312
[E65B30 |  19840/50000 ( 40%) ] Loss: 0.2057 top1= 91.8750
[E65B40 |  26240/50000 ( 52%) ] Loss: 0.1420 top1= 95.6250
[E65B50 |  32640/50000 ( 65%) ] Loss: 0.1621 top1= 94.0625
[E65B60 |  39040/50000 ( 78%) ] Loss: 0.1801 top1= 93.4375
[E65B70 |  45440/50000 ( 91%) ] Loss: 0.1675 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3682 top1= 68.6298


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4873 top1= 42.5982


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.5328 top1= 45.3125

Train epoch 66
[E66B0  |    640/50000 (  1%) ] Loss: 0.1735 top1= 93.5938
[E66B10 |   7040/50000 ( 14%) ] Loss: 0.1559 top1= 94.8438
[E66B20 |  13440/50000 ( 27%) ] Loss: 0.1754 top1= 94.0625
[E66B30 |  19840/50000 ( 40%) ] Loss: 0.1919 top1= 92.6562
[E66B40 |  26240/50000 ( 52%) ] Loss: 0.1383 top1= 95.9375
[E66B50 |  32640/50000 ( 65%) ] Loss: 0.1771 top1= 93.4375
[E66B60 |  39040/50000 ( 78%) ] Loss: 0.1894 top1= 94.6875
[E66B70 |  45440/50000 ( 91%) ] Loss: 0.1543 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6390 top1= 66.7167


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1697 top1= 43.5897


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1088 top1= 46.0938

Train epoch 67
[E67B0  |    640/50000 (  1%) ] Loss: 0.1305 top1= 95.4688
[E67B10 |   7040/50000 ( 14%) ] Loss: 0.1776 top1= 94.0625
[E67B20 |  13440/50000 ( 27%) ] Loss: 0.1687 top1= 95.3125
[E67B30 |  19840/50000 ( 40%) ] Loss: 0.1875 top1= 92.9688
[E67B40 |  26240/50000 ( 52%) ] Loss: 0.1589 top1= 94.6875
[E67B50 |  32640/50000 ( 65%) ] Loss: 0.1610 top1= 92.9688
[E67B60 |  39040/50000 ( 78%) ] Loss: 0.1521 top1= 95.1562
[E67B70 |  45440/50000 ( 91%) ] Loss: 0.2074 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5478 top1= 67.7885


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7914 top1= 44.6514


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.5413 top1= 46.1739

Train epoch 68
[E68B0  |    640/50000 (  1%) ] Loss: 0.1576 top1= 93.5938
[E68B10 |   7040/50000 ( 14%) ] Loss: 0.1404 top1= 94.3750
[E68B20 |  13440/50000 ( 27%) ] Loss: 0.1280 top1= 95.9375
[E68B30 |  19840/50000 ( 40%) ] Loss: 0.1961 top1= 93.5938
[E68B40 |  26240/50000 ( 52%) ] Loss: 0.1684 top1= 93.1250
[E68B50 |  32640/50000 ( 65%) ] Loss: 0.1994 top1= 94.0625
[E68B60 |  39040/50000 ( 78%) ] Loss: 0.1385 top1= 94.0625
[E68B70 |  45440/50000 ( 91%) ] Loss: 0.1504 top1= 94.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4630 top1= 66.8770


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9086 top1= 44.3109


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3464 top1= 45.2424

Train epoch 69
[E69B0  |    640/50000 (  1%) ] Loss: 0.2135 top1= 92.8125
[E69B10 |   7040/50000 ( 14%) ] Loss: 0.1548 top1= 94.3750
[E69B20 |  13440/50000 ( 27%) ] Loss: 0.1724 top1= 94.6875
[E69B30 |  19840/50000 ( 40%) ] Loss: 0.1573 top1= 94.0625
[E69B40 |  26240/50000 ( 52%) ] Loss: 0.2164 top1= 92.3438
[E69B50 |  32640/50000 ( 65%) ] Loss: 0.1695 top1= 93.4375
[E69B60 |  39040/50000 ( 78%) ] Loss: 0.1792 top1= 93.5938
[E69B70 |  45440/50000 ( 91%) ] Loss: 0.1718 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5860 top1= 63.5917


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.2909 top1= 42.7885


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2950 top1= 46.3141

Train epoch 70
[E70B0  |    640/50000 (  1%) ] Loss: 0.1868 top1= 92.8125
[E70B10 |   7040/50000 ( 14%) ] Loss: 0.1821 top1= 93.2812
[E70B20 |  13440/50000 ( 27%) ] Loss: 0.1279 top1= 95.6250
[E70B30 |  19840/50000 ( 40%) ] Loss: 0.1895 top1= 93.1250
[E70B40 |  26240/50000 ( 52%) ] Loss: 0.2107 top1= 93.4375
[E70B50 |  32640/50000 ( 65%) ] Loss: 0.1452 top1= 94.6875
[E70B60 |  39040/50000 ( 78%) ] Loss: 0.1421 top1= 95.9375
[E70B70 |  45440/50000 ( 91%) ] Loss: 0.1639 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5079 top1= 66.4663


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5186 top1= 42.7985


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.5416 top1= 44.8017

Train epoch 71
[E71B0  |    640/50000 (  1%) ] Loss: 0.2177 top1= 93.2812
[E71B10 |   7040/50000 ( 14%) ] Loss: 0.2001 top1= 91.8750
[E71B20 |  13440/50000 ( 27%) ] Loss: 0.1901 top1= 94.2188
[E71B30 |  19840/50000 ( 40%) ] Loss: 0.1565 top1= 94.2188
[E71B40 |  26240/50000 ( 52%) ] Loss: 0.1901 top1= 93.7500
[E71B50 |  32640/50000 ( 65%) ] Loss: 0.1784 top1= 94.2188
[E71B60 |  39040/50000 ( 78%) ] Loss: 0.1577 top1= 95.0000
[E71B70 |  45440/50000 ( 91%) ] Loss: 0.1700 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5267 top1= 66.3862


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.9819 top1= 43.0288


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8187 top1= 46.1839

Train epoch 72
[E72B0  |    640/50000 (  1%) ] Loss: 0.1547 top1= 93.5938
[E72B10 |   7040/50000 ( 14%) ] Loss: 0.1739 top1= 94.2188
[E72B20 |  13440/50000 ( 27%) ] Loss: 0.1370 top1= 95.4688
[E72B30 |  19840/50000 ( 40%) ] Loss: 0.1607 top1= 94.0625
[E72B40 |  26240/50000 ( 52%) ] Loss: 0.1571 top1= 95.6250
[E72B50 |  32640/50000 ( 65%) ] Loss: 0.1815 top1= 93.5938
[E72B60 |  39040/50000 ( 78%) ] Loss: 0.1611 top1= 94.2188
[E72B70 |  45440/50000 ( 91%) ] Loss: 0.1787 top1= 94.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7937 top1= 66.1058


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4768 top1= 43.5296


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.8280 top1= 45.8433

Train epoch 73
[E73B0  |    640/50000 (  1%) ] Loss: 0.1824 top1= 93.5938
[E73B10 |   7040/50000 ( 14%) ] Loss: 0.1686 top1= 94.5312
[E73B20 |  13440/50000 ( 27%) ] Loss: 0.1477 top1= 93.7500
[E73B30 |  19840/50000 ( 40%) ] Loss: 0.2263 top1= 93.4375
[E73B40 |  26240/50000 ( 52%) ] Loss: 0.1930 top1= 93.4375
[E73B50 |  32640/50000 ( 65%) ] Loss: 0.2426 top1= 92.6562
[E73B60 |  39040/50000 ( 78%) ] Loss: 0.2082 top1= 93.4375
[E73B70 |  45440/50000 ( 91%) ] Loss: 0.1675 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4094 top1= 68.8001


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8499 top1= 42.8786


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3330 top1= 45.9836

Train epoch 74
[E74B0  |    640/50000 (  1%) ] Loss: 0.1557 top1= 94.3750
[E74B10 |   7040/50000 ( 14%) ] Loss: 0.1744 top1= 94.3750
[E74B20 |  13440/50000 ( 27%) ] Loss: 0.1747 top1= 95.4688
[E74B30 |  19840/50000 ( 40%) ] Loss: 0.1544 top1= 94.6875
[E74B40 |  26240/50000 ( 52%) ] Loss: 0.1744 top1= 92.6562
[E74B50 |  32640/50000 ( 65%) ] Loss: 0.1503 top1= 95.0000
[E74B60 |  39040/50000 ( 78%) ] Loss: 0.1341 top1= 95.3125
[E74B70 |  45440/50000 ( 91%) ] Loss: 0.1604 top1= 94.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4706 top1= 68.6599


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0712 top1= 43.4696


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7282 top1= 45.1322

Train epoch 75
[E75B0  |    640/50000 (  1%) ] Loss: 0.1678 top1= 92.0312
[E75B10 |   7040/50000 ( 14%) ] Loss: 0.2041 top1= 93.4375
[E75B20 |  13440/50000 ( 27%) ] Loss: 0.2114 top1= 93.5938
[E75B30 |  19840/50000 ( 40%) ] Loss: 0.1917 top1= 92.3438
[E75B40 |  26240/50000 ( 52%) ] Loss: 0.1568 top1= 94.3750
[E75B50 |  32640/50000 ( 65%) ] Loss: 0.1782 top1= 94.5312
[E75B60 |  39040/50000 ( 78%) ] Loss: 0.1663 top1= 94.3750
[E75B70 |  45440/50000 ( 91%) ] Loss: 0.1109 top1= 96.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7511 top1= 67.0673


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7536 top1= 43.5697


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0484 top1= 46.1639

Train epoch 76
[E76B0  |    640/50000 (  1%) ] Loss: 0.1714 top1= 95.1562
[E76B10 |   7040/50000 ( 14%) ] Loss: 0.1196 top1= 95.7812
[E76B20 |  13440/50000 ( 27%) ] Loss: 0.1428 top1= 94.6875
[E76B30 |  19840/50000 ( 40%) ] Loss: 0.1786 top1= 94.5312
[E76B40 |  26240/50000 ( 52%) ] Loss: 0.1518 top1= 95.0000
[E76B50 |  32640/50000 ( 65%) ] Loss: 0.1987 top1= 91.7188
[E76B60 |  39040/50000 ( 78%) ] Loss: 0.1964 top1= 94.6875
[E76B70 |  45440/50000 ( 91%) ] Loss: 0.1839 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5574 top1= 68.5797


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.1920 top1= 42.4079


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0787 top1= 46.5946

Train epoch 77
[E77B0  |    640/50000 (  1%) ] Loss: 0.1930 top1= 93.7500
[E77B10 |   7040/50000 ( 14%) ] Loss: 0.1832 top1= 94.2188
[E77B20 |  13440/50000 ( 27%) ] Loss: 0.1562 top1= 94.2188
[E77B30 |  19840/50000 ( 40%) ] Loss: 0.1703 top1= 93.9062
[E77B40 |  26240/50000 ( 52%) ] Loss: 0.1470 top1= 95.3125
[E77B50 |  32640/50000 ( 65%) ] Loss: 0.1427 top1= 94.2188
[E77B60 |  39040/50000 ( 78%) ] Loss: 0.1547 top1= 95.4688
[E77B70 |  45440/50000 ( 91%) ] Loss: 0.1151 top1= 96.5625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6011 top1= 68.7700


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3360 top1= 43.8902


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4320 top1= 45.9235

Train epoch 78
[E78B0  |    640/50000 (  1%) ] Loss: 0.1436 top1= 95.1562
[E78B10 |   7040/50000 ( 14%) ] Loss: 0.2028 top1= 93.4375
[E78B20 |  13440/50000 ( 27%) ] Loss: 0.1547 top1= 94.5312
[E78B30 |  19840/50000 ( 40%) ] Loss: 0.1736 top1= 92.8125
[E78B40 |  26240/50000 ( 52%) ] Loss: 0.1852 top1= 93.9062
[E78B50 |  32640/50000 ( 65%) ] Loss: 0.1270 top1= 96.2500
[E78B60 |  39040/50000 ( 78%) ] Loss: 0.1793 top1= 94.2188
[E78B70 |  45440/50000 ( 91%) ] Loss: 0.1714 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9800 top1= 67.2175


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.5529 top1= 43.0188


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.0866 top1= 46.4042

Train epoch 79
[E79B0  |    640/50000 (  1%) ] Loss: 0.2301 top1= 94.0625
[E79B10 |   7040/50000 ( 14%) ] Loss: 0.2088 top1= 93.9062
[E79B20 |  13440/50000 ( 27%) ] Loss: 0.1425 top1= 95.6250
[E79B30 |  19840/50000 ( 40%) ] Loss: 0.1621 top1= 93.7500
[E79B40 |  26240/50000 ( 52%) ] Loss: 0.1581 top1= 93.4375
[E79B50 |  32640/50000 ( 65%) ] Loss: 0.2039 top1= 94.5312
[E79B60 |  39040/50000 ( 78%) ] Loss: 0.1563 top1= 94.8438
[E79B70 |  45440/50000 ( 91%) ] Loss: 0.1910 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4463 top1= 67.5781


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2609 top1= 44.8718


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0759 top1= 46.3642

Train epoch 80
[E80B0  |    640/50000 (  1%) ] Loss: 0.1895 top1= 93.9062
[E80B10 |   7040/50000 ( 14%) ] Loss: 0.1895 top1= 93.7500
[E80B20 |  13440/50000 ( 27%) ] Loss: 0.1740 top1= 94.5312
[E80B30 |  19840/50000 ( 40%) ] Loss: 0.1329 top1= 95.4688
[E80B40 |  26240/50000 ( 52%) ] Loss: 0.1730 top1= 92.8125
[E80B50 |  32640/50000 ( 65%) ] Loss: 0.1685 top1= 94.6875
[E80B60 |  39040/50000 ( 78%) ] Loss: 0.1753 top1= 94.5312
[E80B70 |  45440/50000 ( 91%) ] Loss: 0.1538 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6986 top1= 67.9788


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7160 top1= 43.2492


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3383 top1= 45.9635

Train epoch 81
[E81B0  |    640/50000 (  1%) ] Loss: 0.1794 top1= 94.0625
[E81B10 |   7040/50000 ( 14%) ] Loss: 0.1322 top1= 95.7812
[E81B20 |  13440/50000 ( 27%) ] Loss: 0.1169 top1= 95.3125
[E81B30 |  19840/50000 ( 40%) ] Loss: 0.1392 top1= 94.8438
[E81B40 |  26240/50000 ( 52%) ] Loss: 0.1171 top1= 96.5625
[E81B50 |  32640/50000 ( 65%) ] Loss: 0.1092 top1= 96.7188
[E81B60 |  39040/50000 ( 78%) ] Loss: 0.0905 top1= 97.1875
[E81B70 |  45440/50000 ( 91%) ] Loss: 0.1042 top1= 97.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3144 top1= 72.3958


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3090 top1= 49.3590


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7822 top1= 50.1502

Train epoch 82
[E82B0  |    640/50000 (  1%) ] Loss: 0.1441 top1= 95.9375
[E82B10 |   7040/50000 ( 14%) ] Loss: 0.1050 top1= 96.2500
[E82B20 |  13440/50000 ( 27%) ] Loss: 0.0711 top1= 97.3438
[E82B30 |  19840/50000 ( 40%) ] Loss: 0.1079 top1= 95.9375
[E82B40 |  26240/50000 ( 52%) ] Loss: 0.0692 top1= 97.9688
[E82B50 |  32640/50000 ( 65%) ] Loss: 0.0807 top1= 97.6562
[E82B60 |  39040/50000 ( 78%) ] Loss: 0.0701 top1= 98.1250
[E82B70 |  45440/50000 ( 91%) ] Loss: 0.0626 top1= 98.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3936 top1= 72.3458


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3783 top1= 50.0501


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0397 top1= 50.3005

Train epoch 83
[E83B0  |    640/50000 (  1%) ] Loss: 0.0737 top1= 97.8125
[E83B10 |   7040/50000 ( 14%) ] Loss: 0.0731 top1= 97.6562
[E83B20 |  13440/50000 ( 27%) ] Loss: 0.0726 top1= 97.0312
[E83B30 |  19840/50000 ( 40%) ] Loss: 0.1460 top1= 96.7188
[E83B40 |  26240/50000 ( 52%) ] Loss: 0.0541 top1= 97.8125
[E83B50 |  32640/50000 ( 65%) ] Loss: 0.1041 top1= 97.6562
[E83B60 |  39040/50000 ( 78%) ] Loss: 0.0541 top1= 98.4375
[E83B70 |  45440/50000 ( 91%) ] Loss: 0.0832 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5471 top1= 71.6446


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9901 top1= 49.4091


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0765 top1= 48.7981

Train epoch 84
[E84B0  |    640/50000 (  1%) ] Loss: 0.0724 top1= 97.1875
[E84B10 |   7040/50000 ( 14%) ] Loss: 0.0636 top1= 98.1250
[E84B20 |  13440/50000 ( 27%) ] Loss: 0.0699 top1= 97.0312
[E84B30 |  19840/50000 ( 40%) ] Loss: 0.0707 top1= 97.5000
[E84B40 |  26240/50000 ( 52%) ] Loss: 0.0677 top1= 98.1250
[E84B50 |  32640/50000 ( 65%) ] Loss: 0.0433 top1= 98.2812
[E84B60 |  39040/50000 ( 78%) ] Loss: 0.0460 top1= 98.7500
[E84B70 |  45440/50000 ( 91%) ] Loss: 0.0653 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4525 top1= 72.7264


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8838 top1= 48.1871


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0288 top1= 51.0116

Train epoch 85
[E85B0  |    640/50000 (  1%) ] Loss: 0.0713 top1= 97.3438
[E85B10 |   7040/50000 ( 14%) ] Loss: 0.0630 top1= 97.6562
[E85B20 |  13440/50000 ( 27%) ] Loss: 0.0603 top1= 97.5000
[E85B30 |  19840/50000 ( 40%) ] Loss: 0.0645 top1= 97.8125
[E85B40 |  26240/50000 ( 52%) ] Loss: 0.0420 top1= 98.9062
[E85B50 |  32640/50000 ( 65%) ] Loss: 0.0798 top1= 97.5000
[E85B60 |  39040/50000 ( 78%) ] Loss: 0.0411 top1= 98.9062
[E85B70 |  45440/50000 ( 91%) ] Loss: 0.0570 top1= 98.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4993 top1= 72.6462


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6027 top1= 49.3189


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6424 top1= 52.5140

Train epoch 86
[E86B0  |    640/50000 (  1%) ] Loss: 0.0781 top1= 97.1875
[E86B10 |   7040/50000 ( 14%) ] Loss: 0.0706 top1= 97.5000
[E86B20 |  13440/50000 ( 27%) ] Loss: 0.0687 top1= 97.3438
[E86B30 |  19840/50000 ( 40%) ] Loss: 0.0549 top1= 97.9688
[E86B40 |  26240/50000 ( 52%) ] Loss: 0.0456 top1= 98.4375
[E86B50 |  32640/50000 ( 65%) ] Loss: 0.0627 top1= 97.3438
[E86B60 |  39040/50000 ( 78%) ] Loss: 0.0451 top1= 98.2812
[E86B70 |  45440/50000 ( 91%) ] Loss: 0.0323 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5795 top1= 72.0753


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6825 top1= 47.2356


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.5459 top1= 50.3405

Train epoch 87
[E87B0  |    640/50000 (  1%) ] Loss: 0.0855 top1= 98.2812
[E87B10 |   7040/50000 ( 14%) ] Loss: 0.0577 top1= 97.8125
[E87B20 |  13440/50000 ( 27%) ] Loss: 0.0535 top1= 97.9688
[E87B30 |  19840/50000 ( 40%) ] Loss: 0.1043 top1= 97.8125
[E87B40 |  26240/50000 ( 52%) ] Loss: 0.0343 top1= 98.9062
[E87B50 |  32640/50000 ( 65%) ] Loss: 0.0519 top1= 98.1250
[E87B60 |  39040/50000 ( 78%) ] Loss: 0.0417 top1= 98.4375
[E87B70 |  45440/50000 ( 91%) ] Loss: 0.0586 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7192 top1= 71.4443


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7314 top1= 49.7796


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3586 top1= 49.6995

Train epoch 88
[E88B0  |    640/50000 (  1%) ] Loss: 0.0619 top1= 98.1250
[E88B10 |   7040/50000 ( 14%) ] Loss: 0.0560 top1= 98.4375
[E88B20 |  13440/50000 ( 27%) ] Loss: 0.0394 top1= 98.4375
[E88B30 |  19840/50000 ( 40%) ] Loss: 0.0594 top1= 97.5000
[E88B40 |  26240/50000 ( 52%) ] Loss: 0.0327 top1= 98.7500
[E88B50 |  32640/50000 ( 65%) ] Loss: 0.0591 top1= 97.6562
[E88B60 |  39040/50000 ( 78%) ] Loss: 0.0167 top1= 99.5312
[E88B70 |  45440/50000 ( 91%) ] Loss: 0.0190 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9413 top1= 71.1639


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1087 top1= 49.5793


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.3444 top1= 48.2772

Train epoch 89
[E89B0  |    640/50000 (  1%) ] Loss: 0.0551 top1= 97.8125
[E89B10 |   7040/50000 ( 14%) ] Loss: 0.0655 top1= 97.1875
[E89B20 |  13440/50000 ( 27%) ] Loss: 0.0432 top1= 98.4375
[E89B30 |  19840/50000 ( 40%) ] Loss: 0.0797 top1= 97.0312
[E89B40 |  26240/50000 ( 52%) ] Loss: 0.0269 top1= 99.3750
[E89B50 |  32640/50000 ( 65%) ] Loss: 0.0414 top1= 98.7500
[E89B60 |  39040/50000 ( 78%) ] Loss: 0.0653 top1= 97.3438
[E89B70 |  45440/50000 ( 91%) ] Loss: 0.0313 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7482 top1= 72.0853


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4410 top1= 48.8081


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7125 top1= 52.3838

Train epoch 90
[E90B0  |    640/50000 (  1%) ] Loss: 0.0536 top1= 98.2812
[E90B10 |   7040/50000 ( 14%) ] Loss: 0.0605 top1= 98.4375
[E90B20 |  13440/50000 ( 27%) ] Loss: 0.0447 top1= 98.2812
[E90B30 |  19840/50000 ( 40%) ] Loss: 0.0459 top1= 98.5938
[E90B40 |  26240/50000 ( 52%) ] Loss: 0.0306 top1= 98.4375
[E90B50 |  32640/50000 ( 65%) ] Loss: 0.0350 top1= 98.9062
[E90B60 |  39040/50000 ( 78%) ] Loss: 0.0469 top1= 98.5938
[E90B70 |  45440/50000 ( 91%) ] Loss: 0.0418 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3582 top1= 68.2091


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7183 top1= 49.4491


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.8810 top1= 47.6262

Train epoch 91
[E91B0  |    640/50000 (  1%) ] Loss: 0.0749 top1= 97.3438
[E91B10 |   7040/50000 ( 14%) ] Loss: 0.0532 top1= 97.8125
[E91B20 |  13440/50000 ( 27%) ] Loss: 0.0252 top1= 98.9062
[E91B30 |  19840/50000 ( 40%) ] Loss: 0.0321 top1= 98.7500
[E91B40 |  26240/50000 ( 52%) ] Loss: 0.0217 top1= 99.2188
[E91B50 |  32640/50000 ( 65%) ] Loss: 0.0495 top1= 98.2812
[E91B60 |  39040/50000 ( 78%) ] Loss: 0.0276 top1= 98.7500
[E91B70 |  45440/50000 ( 91%) ] Loss: 0.0451 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7800 top1= 72.5260


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6019 top1= 49.3189


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.6126 top1= 50.4307

Train epoch 92
[E92B0  |    640/50000 (  1%) ] Loss: 0.0695 top1= 96.8750
[E92B10 |   7040/50000 ( 14%) ] Loss: 0.0488 top1= 98.4375
[E92B20 |  13440/50000 ( 27%) ] Loss: 0.0360 top1= 99.0625
[E92B30 |  19840/50000 ( 40%) ] Loss: 0.0345 top1= 98.2812
[E92B40 |  26240/50000 ( 52%) ] Loss: 0.0306 top1= 99.0625
[E92B50 |  32640/50000 ( 65%) ] Loss: 0.0314 top1= 98.7500
[E92B60 |  39040/50000 ( 78%) ] Loss: 0.0584 top1= 98.2812
[E92B70 |  45440/50000 ( 91%) ] Loss: 0.0254 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9859 top1= 71.0938


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7409 top1= 48.8181


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.7851 top1= 49.0184

Train epoch 93
[E93B0  |    640/50000 (  1%) ] Loss: 0.0532 top1= 98.4375
[E93B10 |   7040/50000 ( 14%) ] Loss: 0.0622 top1= 97.6562
[E93B20 |  13440/50000 ( 27%) ] Loss: 0.0587 top1= 98.9062
[E93B30 |  19840/50000 ( 40%) ] Loss: 0.0569 top1= 97.8125
[E93B40 |  26240/50000 ( 52%) ] Loss: 0.0168 top1= 99.6875
[E93B50 |  32640/50000 ( 65%) ] Loss: 0.0260 top1= 98.7500
[E93B60 |  39040/50000 ( 78%) ] Loss: 0.0188 top1= 99.6875
[E93B70 |  45440/50000 ( 91%) ] Loss: 0.0404 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0624 top1= 70.7432


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6523 top1= 49.4191


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.8963 top1= 48.9784

Train epoch 94
[E94B0  |    640/50000 (  1%) ] Loss: 0.0430 top1= 98.5938
[E94B10 |   7040/50000 ( 14%) ] Loss: 0.0347 top1= 98.5938
[E94B20 |  13440/50000 ( 27%) ] Loss: 0.0535 top1= 98.7500
[E94B30 |  19840/50000 ( 40%) ] Loss: 0.0657 top1= 97.8125
[E94B40 |  26240/50000 ( 52%) ] Loss: 0.0231 top1= 99.0625
[E94B50 |  32640/50000 ( 65%) ] Loss: 0.0187 top1= 99.3750
[E94B60 |  39040/50000 ( 78%) ] Loss: 0.0236 top1= 99.0625
[E94B70 |  45440/50000 ( 91%) ] Loss: 0.0157 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9206 top1= 71.5745


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.2287 top1= 47.0252


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.6369 top1= 51.4323

Train epoch 95
[E95B0  |    640/50000 (  1%) ] Loss: 0.0473 top1= 97.8125
[E95B10 |   7040/50000 ( 14%) ] Loss: 0.0361 top1= 98.9062
[E95B20 |  13440/50000 ( 27%) ] Loss: 0.0421 top1= 98.7500
[E95B30 |  19840/50000 ( 40%) ] Loss: 0.0459 top1= 98.7500
[E95B40 |  26240/50000 ( 52%) ] Loss: 0.0126 top1= 99.6875
[E95B50 |  32640/50000 ( 65%) ] Loss: 0.0322 top1= 99.2188
[E95B60 |  39040/50000 ( 78%) ] Loss: 0.0263 top1= 99.2188
[E95B70 |  45440/50000 ( 91%) ] Loss: 0.0244 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1345 top1= 70.0120


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.1652 top1= 46.5345


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.1989 top1= 47.5461

Train epoch 96
[E96B0  |    640/50000 (  1%) ] Loss: 0.0433 top1= 98.9062
[E96B10 |   7040/50000 ( 14%) ] Loss: 0.0395 top1= 98.2812
[E96B20 |  13440/50000 ( 27%) ] Loss: 0.0339 top1= 99.2188
[E96B30 |  19840/50000 ( 40%) ] Loss: 0.0404 top1= 98.5938
[E96B40 |  26240/50000 ( 52%) ] Loss: 0.0381 top1= 98.7500
[E96B50 |  32640/50000 ( 65%) ] Loss: 0.0333 top1= 98.9062
[E96B60 |  39040/50000 ( 78%) ] Loss: 0.0326 top1= 99.3750
[E96B70 |  45440/50000 ( 91%) ] Loss: 0.0160 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1012 top1= 70.7833


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8954 top1= 45.0120


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.6957 top1= 48.2272

Train epoch 97
[E97B0  |    640/50000 (  1%) ] Loss: 0.0424 top1= 98.5938
[E97B10 |   7040/50000 ( 14%) ] Loss: 0.0717 top1= 97.9688
[E97B20 |  13440/50000 ( 27%) ] Loss: 0.0392 top1= 98.9062
[E97B30 |  19840/50000 ( 40%) ] Loss: 0.0282 top1= 98.9062
[E97B40 |  26240/50000 ( 52%) ] Loss: 0.0350 top1= 99.0625
[E97B50 |  32640/50000 ( 65%) ] Loss: 0.0368 top1= 98.4375
[E97B60 |  39040/50000 ( 78%) ] Loss: 0.0180 top1= 99.3750
[E97B70 |  45440/50000 ( 91%) ] Loss: 0.0216 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0291 top1= 70.7432


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.1766 top1= 45.2724


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.6105 top1= 49.1587

Train epoch 98
[E98B0  |    640/50000 (  1%) ] Loss: 0.0417 top1= 98.2812
[E98B10 |   7040/50000 ( 14%) ] Loss: 0.0496 top1= 98.2812
[E98B20 |  13440/50000 ( 27%) ] Loss: 0.0194 top1= 99.3750
[E98B30 |  19840/50000 ( 40%) ] Loss: 0.0371 top1= 98.5938
[E98B40 |  26240/50000 ( 52%) ] Loss: 0.0275 top1= 99.3750
[E98B50 |  32640/50000 ( 65%) ] Loss: 0.0382 top1= 98.5938
[E98B60 |  39040/50000 ( 78%) ] Loss: 0.0231 top1= 99.3750
[E98B70 |  45440/50000 ( 91%) ] Loss: 0.0238 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2232 top1= 70.4627


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.5828 top1= 47.0954


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.6302 top1= 48.7881

Train epoch 99
[E99B0  |    640/50000 (  1%) ] Loss: 0.0344 top1= 98.9062
[E99B10 |   7040/50000 ( 14%) ] Loss: 0.0426 top1= 98.1250
[E99B20 |  13440/50000 ( 27%) ] Loss: 0.0294 top1= 98.5938
[E99B30 |  19840/50000 ( 40%) ] Loss: 0.0277 top1= 99.2188
[E99B40 |  26240/50000 ( 52%) ] Loss: 0.0394 top1= 98.4375
[E99B50 |  32640/50000 ( 65%) ] Loss: 0.0316 top1= 99.2188
[E99B60 |  39040/50000 ( 78%) ] Loss: 0.0229 top1= 99.3750
[E99B70 |  45440/50000 ( 91%) ] Loss: 0.0144 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0679 top1= 71.7648


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.7717 top1= 47.7364


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9650 top1= 50.7712

Train epoch 100
[E100B0  |    640/50000 (  1%) ] Loss: 0.0447 top1= 99.0625
[E100B10 |   7040/50000 ( 14%) ] Loss: 0.0368 top1= 98.9062
[E100B20 |  13440/50000 ( 27%) ] Loss: 0.0257 top1= 99.3750
[E100B30 |  19840/50000 ( 40%) ] Loss: 0.0154 top1= 99.3750
[E100B40 |  26240/50000 ( 52%) ] Loss: 0.0163 top1= 99.5312
[E100B50 |  32640/50000 ( 65%) ] Loss: 0.0188 top1= 99.5312
[E100B60 |  39040/50000 ( 78%) ] Loss: 0.0135 top1= 99.5312
[E100B70 |  45440/50000 ( 91%) ] Loss: 0.0326 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4098 top1= 69.3910


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.0597 top1= 49.1987


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.8317 top1= 48.0970

Train epoch 101
[E101B0  |    640/50000 (  1%) ] Loss: 0.0275 top1= 99.0625
[E101B10 |   7040/50000 ( 14%) ] Loss: 0.0351 top1= 98.7500
[E101B20 |  13440/50000 ( 27%) ] Loss: 0.0166 top1= 99.5312
[E101B30 |  19840/50000 ( 40%) ] Loss: 0.0388 top1= 98.5938
[E101B40 |  26240/50000 ( 52%) ] Loss: 0.0246 top1= 98.9062
[E101B50 |  32640/50000 ( 65%) ] Loss: 0.0284 top1= 99.2188
[E101B60 |  39040/50000 ( 78%) ] Loss: 0.0391 top1= 98.5938
[E101B70 |  45440/50000 ( 91%) ] Loss: 0.0300 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0603 top1= 70.8834


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.6947 top1= 49.0084


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.5283 top1= 49.0385

Train epoch 102
[E102B0  |    640/50000 (  1%) ] Loss: 0.0258 top1= 98.7500
[E102B10 |   7040/50000 ( 14%) ] Loss: 0.0219 top1= 99.2188
[E102B20 |  13440/50000 ( 27%) ] Loss: 0.0292 top1= 98.4375
[E102B30 |  19840/50000 ( 40%) ] Loss: 0.0368 top1= 99.0625
[E102B40 |  26240/50000 ( 52%) ] Loss: 0.0382 top1= 98.9062
[E102B50 |  32640/50000 ( 65%) ] Loss: 0.0247 top1= 98.9062
[E102B60 |  39040/50000 ( 78%) ] Loss: 0.0232 top1= 99.0625
[E102B70 |  45440/50000 ( 91%) ] Loss: 0.0113 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9018 top1= 72.2857


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.8773 top1= 45.8834


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.9542 top1= 48.9283

Train epoch 103
[E103B0  |    640/50000 (  1%) ] Loss: 0.0417 top1= 99.2188
[E103B10 |   7040/50000 ( 14%) ] Loss: 0.0332 top1= 99.0625
[E103B20 |  13440/50000 ( 27%) ] Loss: 0.0277 top1= 99.2188
[E103B30 |  19840/50000 ( 40%) ] Loss: 0.0253 top1= 99.3750
[E103B40 |  26240/50000 ( 52%) ] Loss: 0.0139 top1= 99.5312
[E103B50 |  32640/50000 ( 65%) ] Loss: 0.0365 top1= 99.2188
[E103B60 |  39040/50000 ( 78%) ] Loss: 0.0155 top1= 99.6875
[E103B70 |  45440/50000 ( 91%) ] Loss: 0.0208 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=5.2285 top1= 57.8225


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.0430 top1= 48.6478


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=13.6879 top1= 46.5445

Train epoch 104
[E104B0  |    640/50000 (  1%) ] Loss: 0.0478 top1= 98.4375
[E104B10 |   7040/50000 ( 14%) ] Loss: 0.0332 top1= 98.5938
[E104B20 |  13440/50000 ( 27%) ] Loss: 0.0206 top1= 99.2188
[E104B30 |  19840/50000 ( 40%) ] Loss: 0.0216 top1= 99.3750
[E104B40 |  26240/50000 ( 52%) ] Loss: 0.0105 top1= 99.6875
[E104B50 |  32640/50000 ( 65%) ] Loss: 0.0315 top1= 98.9062
[E104B60 |  39040/50000 ( 78%) ] Loss: 0.0212 top1= 99.3750
[E104B70 |  45440/50000 ( 91%) ] Loss: 0.0196 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1256 top1= 71.5445


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.6110 top1= 46.4643


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9263 top1= 51.6326

Train epoch 105
[E105B0  |    640/50000 (  1%) ] Loss: 0.0096 top1= 99.8438
[E105B10 |   7040/50000 ( 14%) ] Loss: 0.0178 top1= 99.3750
[E105B20 |  13440/50000 ( 27%) ] Loss: 0.0208 top1= 99.3750
[E105B30 |  19840/50000 ( 40%) ] Loss: 0.0158 top1= 99.3750
[E105B40 |  26240/50000 ( 52%) ] Loss: 0.0245 top1= 99.3750
[E105B50 |  32640/50000 ( 65%) ] Loss: 0.0346 top1= 99.2188
[E105B60 |  39040/50000 ( 78%) ] Loss: 0.0200 top1= 99.3750
[E105B70 |  45440/50000 ( 91%) ] Loss: 0.0358 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2032 top1= 71.7348


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.0183 top1= 45.9034


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.6943 top1= 49.0385

Train epoch 106
[E106B0  |    640/50000 (  1%) ] Loss: 0.0238 top1= 99.2188
[E106B10 |   7040/50000 ( 14%) ] Loss: 0.0388 top1= 98.2812
[E106B20 |  13440/50000 ( 27%) ] Loss: 0.0143 top1= 99.8438
[E106B30 |  19840/50000 ( 40%) ] Loss: 0.0355 top1= 98.7500
[E106B40 |  26240/50000 ( 52%) ] Loss: 0.0331 top1= 99.0625
[E106B50 |  32640/50000 ( 65%) ] Loss: 0.0252 top1= 99.0625
[E106B60 |  39040/50000 ( 78%) ] Loss: 0.0112 top1= 99.6875
[E106B70 |  45440/50000 ( 91%) ] Loss: 0.0166 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2393 top1= 71.4543


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.5918 top1= 45.9535


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.6063 top1= 47.3758

Train epoch 107
[E107B0  |    640/50000 (  1%) ] Loss: 0.0272 top1= 98.9062
[E107B10 |   7040/50000 ( 14%) ] Loss: 0.0195 top1= 99.3750
[E107B20 |  13440/50000 ( 27%) ] Loss: 0.0233 top1= 99.2188
[E107B30 |  19840/50000 ( 40%) ] Loss: 0.0301 top1= 99.0625
[E107B40 |  26240/50000 ( 52%) ] Loss: 0.0312 top1= 99.0625
[E107B50 |  32640/50000 ( 65%) ] Loss: 0.0366 top1= 98.9062
[E107B60 |  39040/50000 ( 78%) ] Loss: 0.0319 top1= 98.5938
[E107B70 |  45440/50000 ( 91%) ] Loss: 0.0357 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0839 top1= 72.1955


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.7527 top1= 45.3325


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.8231 top1= 47.8966

Train epoch 108
[E108B0  |    640/50000 (  1%) ] Loss: 0.0179 top1= 99.3750
[E108B10 |   7040/50000 ( 14%) ] Loss: 0.0274 top1= 99.0625
[E108B20 |  13440/50000 ( 27%) ] Loss: 0.0136 top1= 99.5312
[E108B30 |  19840/50000 ( 40%) ] Loss: 0.0240 top1= 98.7500
[E108B40 |  26240/50000 ( 52%) ] Loss: 0.0175 top1= 99.3750
[E108B50 |  32640/50000 ( 65%) ] Loss: 0.0211 top1= 99.3750
[E108B60 |  39040/50000 ( 78%) ] Loss: 0.0072 top1=100.0000
[E108B70 |  45440/50000 ( 91%) ] Loss: 0.0155 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2204 top1= 70.9836


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.0635 top1= 46.5745


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3904 top1= 50.2304

Train epoch 109
[E109B0  |    640/50000 (  1%) ] Loss: 0.0158 top1= 99.3750
[E109B10 |   7040/50000 ( 14%) ] Loss: 0.0351 top1= 98.1250
[E109B20 |  13440/50000 ( 27%) ] Loss: 0.0109 top1= 99.5312
[E109B30 |  19840/50000 ( 40%) ] Loss: 0.0202 top1= 99.3750
[E109B40 |  26240/50000 ( 52%) ] Loss: 0.0189 top1= 99.3750
[E109B50 |  32640/50000 ( 65%) ] Loss: 0.0187 top1= 99.2188
[E109B60 |  39040/50000 ( 78%) ] Loss: 0.0104 top1= 99.5312
[E109B70 |  45440/50000 ( 91%) ] Loss: 0.0192 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0685 top1= 71.6647


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.0643 top1= 46.4744


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.0609 top1= 49.6995

Train epoch 110
[E110B0  |    640/50000 (  1%) ] Loss: 0.0198 top1= 99.5312
[E110B10 |   7040/50000 ( 14%) ] Loss: 0.0304 top1= 99.0625
[E110B20 |  13440/50000 ( 27%) ] Loss: 0.0171 top1= 99.3750
[E110B30 |  19840/50000 ( 40%) ] Loss: 0.0207 top1= 99.2188
[E110B40 |  26240/50000 ( 52%) ] Loss: 0.0180 top1= 99.2188
[E110B50 |  32640/50000 ( 65%) ] Loss: 0.0144 top1= 99.5312
[E110B60 |  39040/50000 ( 78%) ] Loss: 0.0152 top1= 99.5312
[E110B70 |  45440/50000 ( 91%) ] Loss: 0.0314 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2896 top1= 70.8634


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.1808 top1= 47.4259


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1135 top1= 51.9030

Train epoch 111
[E111B0  |    640/50000 (  1%) ] Loss: 0.0319 top1= 98.7500
[E111B10 |   7040/50000 ( 14%) ] Loss: 0.0516 top1= 98.5938
[E111B20 |  13440/50000 ( 27%) ] Loss: 0.0207 top1= 99.3750
[E111B30 |  19840/50000 ( 40%) ] Loss: 0.0124 top1= 99.5312
[E111B40 |  26240/50000 ( 52%) ] Loss: 0.0119 top1= 99.5312
[E111B50 |  32640/50000 ( 65%) ] Loss: 0.0221 top1= 99.0625
[E111B60 |  39040/50000 ( 78%) ] Loss: 0.0197 top1= 99.6875
[E111B70 |  45440/50000 ( 91%) ] Loss: 0.0103 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5775 top1= 69.9820


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.4674 top1= 45.4427


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.4581 top1= 47.1554

Train epoch 112
[E112B0  |    640/50000 (  1%) ] Loss: 0.0308 top1= 99.0625
[E112B10 |   7040/50000 ( 14%) ] Loss: 0.0350 top1= 99.0625
[E112B20 |  13440/50000 ( 27%) ] Loss: 0.0177 top1= 99.2188
[E112B30 |  19840/50000 ( 40%) ] Loss: 0.0400 top1= 98.4375
[E112B40 |  26240/50000 ( 52%) ] Loss: 0.0242 top1= 99.2188
[E112B50 |  32640/50000 ( 65%) ] Loss: 0.0211 top1= 99.2188
[E112B60 |  39040/50000 ( 78%) ] Loss: 0.0383 top1= 98.7500
[E112B70 |  45440/50000 ( 91%) ] Loss: 0.0154 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6568 top1= 70.1623


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.8283 top1= 45.1022


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.0721 top1= 48.0268

Train epoch 113
[E113B0  |    640/50000 (  1%) ] Loss: 0.0247 top1= 99.2188
[E113B10 |   7040/50000 ( 14%) ] Loss: 0.0249 top1= 98.7500
[E113B20 |  13440/50000 ( 27%) ] Loss: 0.0511 top1= 99.2188
[E113B30 |  19840/50000 ( 40%) ] Loss: 0.0427 top1= 98.7500
[E113B40 |  26240/50000 ( 52%) ] Loss: 0.0145 top1= 99.3750
[E113B50 |  32640/50000 ( 65%) ] Loss: 0.0178 top1= 99.3750
[E113B60 |  39040/50000 ( 78%) ] Loss: 0.0071 top1= 99.8438
[E113B70 |  45440/50000 ( 91%) ] Loss: 0.0053 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0497 top1= 72.5861


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.5764 top1= 47.3958


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.3767 top1= 49.8297

Train epoch 114
[E114B0  |    640/50000 (  1%) ] Loss: 0.0299 top1= 99.0625
[E114B10 |   7040/50000 ( 14%) ] Loss: 0.0372 top1= 98.7500
[E114B20 |  13440/50000 ( 27%) ] Loss: 0.0153 top1= 99.5312
[E114B30 |  19840/50000 ( 40%) ] Loss: 0.0118 top1= 99.6875
[E114B40 |  26240/50000 ( 52%) ] Loss: 0.0038 top1= 99.8438
[E114B50 |  32640/50000 ( 65%) ] Loss: 0.0233 top1= 99.0625
[E114B60 |  39040/50000 ( 78%) ] Loss: 0.0088 top1= 99.6875
[E114B70 |  45440/50000 ( 91%) ] Loss: 0.0277 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3886 top1= 70.6731


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=14.2149 top1= 44.6114


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.5583 top1= 48.4475

Train epoch 115
[E115B0  |    640/50000 (  1%) ] Loss: 0.0104 top1= 99.6875
[E115B10 |   7040/50000 ( 14%) ] Loss: 0.0228 top1= 99.2188
[E115B20 |  13440/50000 ( 27%) ] Loss: 0.0082 top1= 99.6875
[E115B30 |  19840/50000 ( 40%) ] Loss: 0.0309 top1= 99.0625
[E115B40 |  26240/50000 ( 52%) ] Loss: 0.0220 top1= 99.6875
[E115B50 |  32640/50000 ( 65%) ] Loss: 0.0121 top1= 99.5312
[E115B60 |  39040/50000 ( 78%) ] Loss: 0.0142 top1= 99.5312
[E115B70 |  45440/50000 ( 91%) ] Loss: 0.0062 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4621 top1= 70.4828


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.7272 top1= 45.7532


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2969 top1= 53.7660

Train epoch 116
[E116B0  |    640/50000 (  1%) ] Loss: 0.0221 top1= 99.2188
[E116B10 |   7040/50000 ( 14%) ] Loss: 0.0205 top1= 99.2188
[E116B20 |  13440/50000 ( 27%) ] Loss: 0.0349 top1= 98.9062
[E116B30 |  19840/50000 ( 40%) ] Loss: 0.0211 top1= 99.0625
[E116B40 |  26240/50000 ( 52%) ] Loss: 0.0049 top1=100.0000
[E116B50 |  32640/50000 ( 65%) ] Loss: 0.0214 top1= 99.3750
[E116B60 |  39040/50000 ( 78%) ] Loss: 0.0095 top1= 99.5312
[E116B70 |  45440/50000 ( 91%) ] Loss: 0.0108 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.8123 top1= 68.7099


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=13.4807 top1= 44.9920


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0366 top1= 51.9832

Train epoch 117
[E117B0  |    640/50000 (  1%) ] Loss: 0.0218 top1= 99.0625
[E117B10 |   7040/50000 ( 14%) ] Loss: 0.0121 top1= 99.8438
[E117B20 |  13440/50000 ( 27%) ] Loss: 0.0197 top1= 99.2188
[E117B30 |  19840/50000 ( 40%) ] Loss: 0.0087 top1= 99.8438
[E117B40 |  26240/50000 ( 52%) ] Loss: 0.0265 top1= 98.9062
[E117B50 |  32640/50000 ( 65%) ] Loss: 0.0327 top1= 98.7500
[E117B60 |  39040/50000 ( 78%) ] Loss: 0.0487 top1= 99.0625
[E117B70 |  45440/50000 ( 91%) ] Loss: 0.0072 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5410 top1= 70.5829


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.0108 top1= 46.2340


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0788 top1= 52.5541

Train epoch 118
[E118B0  |    640/50000 (  1%) ] Loss: 0.0167 top1= 99.2188
[E118B10 |   7040/50000 ( 14%) ] Loss: 0.0170 top1= 99.3750
[E118B20 |  13440/50000 ( 27%) ] Loss: 0.0115 top1= 99.6875
[E118B30 |  19840/50000 ( 40%) ] Loss: 0.0121 top1= 99.5312
[E118B40 |  26240/50000 ( 52%) ] Loss: 0.0076 top1= 99.6875
[E118B50 |  32640/50000 ( 65%) ] Loss: 0.0075 top1= 99.6875
[E118B60 |  39040/50000 ( 78%) ] Loss: 0.0165 top1= 99.5312
[E118B70 |  45440/50000 ( 91%) ] Loss: 0.0076 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5689 top1= 70.3626


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.8721 top1= 45.4627


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0287 top1= 53.1751

Train epoch 119
[E119B0  |    640/50000 (  1%) ] Loss: 0.0263 top1= 99.0625
[E119B10 |   7040/50000 ( 14%) ] Loss: 0.0065 top1= 99.8438
[E119B20 |  13440/50000 ( 27%) ] Loss: 0.0090 top1= 99.6875
[E119B30 |  19840/50000 ( 40%) ] Loss: 0.0092 top1= 99.8438
[E119B40 |  26240/50000 ( 52%) ] Loss: 0.0024 top1=100.0000
[E119B50 |  32640/50000 ( 65%) ] Loss: 0.0109 top1= 99.8438
[E119B60 |  39040/50000 ( 78%) ] Loss: 0.0080 top1= 99.8438
[E119B70 |  45440/50000 ( 91%) ] Loss: 0.0297 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6294 top1= 70.4928


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.9637 top1= 45.6631


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0069 top1= 54.7776

Train epoch 120
[E120B0  |    640/50000 (  1%) ] Loss: 0.0200 top1= 99.3750
[E120B10 |   7040/50000 ( 14%) ] Loss: 0.0302 top1= 99.2188
[E120B20 |  13440/50000 ( 27%) ] Loss: 0.0084 top1= 99.6875
[E120B30 |  19840/50000 ( 40%) ] Loss: 0.0181 top1= 99.5312
[E120B40 |  26240/50000 ( 52%) ] Loss: 0.0035 top1=100.0000
[E120B50 |  32640/50000 ( 65%) ] Loss: 0.0221 top1= 99.0625
[E120B60 |  39040/50000 ( 78%) ] Loss: 0.0111 top1= 99.6875
[E120B70 |  45440/50000 ( 91%) ] Loss: 0.0130 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6365 top1= 70.7332


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.0632 top1= 49.2488


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.4902 top1= 48.3774

Train epoch 121
[E121B0  |    640/50000 (  1%) ] Loss: 0.0405 top1= 99.2188
[E121B10 |   7040/50000 ( 14%) ] Loss: 0.0315 top1= 99.0625
[E121B20 |  13440/50000 ( 27%) ] Loss: 0.0142 top1= 99.3750
[E121B30 |  19840/50000 ( 40%) ] Loss: 0.0138 top1= 99.5312
[E121B40 |  26240/50000 ( 52%) ] Loss: 0.0085 top1= 99.6875
[E121B50 |  32640/50000 ( 65%) ] Loss: 0.0330 top1= 99.5312
[E121B60 |  39040/50000 ( 78%) ] Loss: 0.0111 top1= 99.5312
[E121B70 |  45440/50000 ( 91%) ] Loss: 0.0222 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1210 top1= 72.8466


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4579 top1= 51.3221


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8011 top1= 53.3554

Train epoch 122
[E122B0  |    640/50000 (  1%) ] Loss: 0.0075 top1=100.0000
[E122B10 |   7040/50000 ( 14%) ] Loss: 0.0350 top1= 99.0625
[E122B20 |  13440/50000 ( 27%) ] Loss: 0.0158 top1= 99.5312
[E122B30 |  19840/50000 ( 40%) ] Loss: 0.0142 top1= 99.3750
[E122B40 |  26240/50000 ( 52%) ] Loss: 0.0214 top1= 99.3750
[E122B50 |  32640/50000 ( 65%) ] Loss: 0.0157 top1= 99.3750
[E122B60 |  39040/50000 ( 78%) ] Loss: 0.0228 top1= 99.6875
[E122B70 |  45440/50000 ( 91%) ] Loss: 0.0103 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2481 top1= 72.2857


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.6786 top1= 49.1987


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4103 top1= 56.6306

Train epoch 123
[E123B0  |    640/50000 (  1%) ] Loss: 0.0391 top1= 98.9062
[E123B10 |   7040/50000 ( 14%) ] Loss: 0.0137 top1= 99.6875
[E123B20 |  13440/50000 ( 27%) ] Loss: 0.0059 top1= 99.8438
[E123B30 |  19840/50000 ( 40%) ] Loss: 0.0166 top1= 99.6875
[E123B40 |  26240/50000 ( 52%) ] Loss: 0.0509 top1= 98.7500
[E123B50 |  32640/50000 ( 65%) ] Loss: 0.0117 top1= 99.3750
[E123B60 |  39040/50000 ( 78%) ] Loss: 0.0079 top1= 99.6875
[E123B70 |  45440/50000 ( 91%) ] Loss: 0.0035 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1352 top1= 72.5861


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.8956 top1= 49.0685


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9231 top1= 52.9046

Train epoch 124
[E124B0  |    640/50000 (  1%) ] Loss: 0.0071 top1= 99.8438
[E124B10 |   7040/50000 ( 14%) ] Loss: 0.0391 top1= 99.0625
[E124B20 |  13440/50000 ( 27%) ] Loss: 0.0097 top1= 99.5312
[E124B30 |  19840/50000 ( 40%) ] Loss: 0.0075 top1= 99.8438
[E124B40 |  26240/50000 ( 52%) ] Loss: 0.0142 top1= 99.2188
[E124B50 |  32640/50000 ( 65%) ] Loss: 0.0121 top1= 99.3750
[E124B60 |  39040/50000 ( 78%) ] Loss: 0.0103 top1= 99.8438
[E124B70 |  45440/50000 ( 91%) ] Loss: 0.0127 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3162 top1= 71.8950


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.8891 top1= 47.5962


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0622 top1= 54.8678

Train epoch 125
[E125B0  |    640/50000 (  1%) ] Loss: 0.0182 top1= 99.3750
[E125B10 |   7040/50000 ( 14%) ] Loss: 0.0098 top1= 99.6875
[E125B20 |  13440/50000 ( 27%) ] Loss: 0.0072 top1= 99.6875
[E125B30 |  19840/50000 ( 40%) ] Loss: 0.0219 top1= 99.3750
[E125B40 |  26240/50000 ( 52%) ] Loss: 0.0025 top1=100.0000
[E125B50 |  32640/50000 ( 65%) ] Loss: 0.0241 top1= 99.2188
[E125B60 |  39040/50000 ( 78%) ] Loss: 0.0028 top1=100.0000
[E125B70 |  45440/50000 ( 91%) ] Loss: 0.0124 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.8513 top1= 69.4211


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.2476 top1= 46.0737


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3919 top1= 56.7308

Train epoch 126
[E126B0  |    640/50000 (  1%) ] Loss: 0.0320 top1= 98.7500
[E126B10 |   7040/50000 ( 14%) ] Loss: 0.0366 top1= 99.2188
[E126B20 |  13440/50000 ( 27%) ] Loss: 0.0282 top1= 99.5312
[E126B30 |  19840/50000 ( 40%) ] Loss: 0.0211 top1= 99.3750
[E126B40 |  26240/50000 ( 52%) ] Loss: 0.0086 top1= 99.8438
[E126B50 |  32640/50000 ( 65%) ] Loss: 0.0325 top1= 98.9062
[E126B60 |  39040/50000 ( 78%) ] Loss: 0.0217 top1= 99.2188
[E126B70 |  45440/50000 ( 91%) ] Loss: 0.0111 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1312 top1= 72.9667


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5273 top1= 51.1719


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3351 top1= 54.2167

Train epoch 127
[E127B0  |    640/50000 (  1%) ] Loss: 0.0156 top1= 99.6875
[E127B10 |   7040/50000 ( 14%) ] Loss: 0.0224 top1= 98.9062
[E127B20 |  13440/50000 ( 27%) ] Loss: 0.0145 top1= 99.5312
[E127B30 |  19840/50000 ( 40%) ] Loss: 0.0222 top1= 99.5312
[E127B40 |  26240/50000 ( 52%) ] Loss: 0.0215 top1= 99.5312
[E127B50 |  32640/50000 ( 65%) ] Loss: 0.0050 top1= 99.8438
[E127B60 |  39040/50000 ( 78%) ] Loss: 0.0072 top1= 99.6875
[E127B70 |  45440/50000 ( 91%) ] Loss: 0.0037 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1127 top1= 73.1270


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5314 top1= 50.9916


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.6112 top1= 53.7760

Train epoch 128
[E128B0  |    640/50000 (  1%) ] Loss: 0.0096 top1= 99.8438
[E128B10 |   7040/50000 ( 14%) ] Loss: 0.0146 top1= 99.6875
[E128B20 |  13440/50000 ( 27%) ] Loss: 0.0124 top1= 99.5312
[E128B30 |  19840/50000 ( 40%) ] Loss: 0.0406 top1= 99.6875
[E128B40 |  26240/50000 ( 52%) ] Loss: 0.0147 top1= 99.5312
[E128B50 |  32640/50000 ( 65%) ] Loss: 0.0107 top1= 99.6875
[E128B60 |  39040/50000 ( 78%) ] Loss: 0.0085 top1= 99.6875
[E128B70 |  45440/50000 ( 91%) ] Loss: 0.0105 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1345 top1= 72.8466


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0017 top1= 52.0933


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6289 top1= 55.9095

Train epoch 129
[E129B0  |    640/50000 (  1%) ] Loss: 0.0205 top1= 99.3750
[E129B10 |   7040/50000 ( 14%) ] Loss: 0.0113 top1= 99.5312
[E129B20 |  13440/50000 ( 27%) ] Loss: 0.0191 top1= 99.5312
[E129B30 |  19840/50000 ( 40%) ] Loss: 0.0154 top1= 99.3750
[E129B40 |  26240/50000 ( 52%) ] Loss: 0.0035 top1= 99.8438
[E129B50 |  32640/50000 ( 65%) ] Loss: 0.0303 top1= 99.5312
[E129B60 |  39040/50000 ( 78%) ] Loss: 0.0248 top1= 99.5312
[E129B70 |  45440/50000 ( 91%) ] Loss: 0.0084 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1838 top1= 72.6963


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.3450 top1= 49.7496


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.1256 top1= 50.6010

Train epoch 130
[E130B0  |    640/50000 (  1%) ] Loss: 0.0160 top1= 99.5312
[E130B10 |   7040/50000 ( 14%) ] Loss: 0.0128 top1= 99.6875
[E130B20 |  13440/50000 ( 27%) ] Loss: 0.0139 top1= 99.3750
[E130B30 |  19840/50000 ( 40%) ] Loss: 0.0169 top1= 99.6875
[E130B40 |  26240/50000 ( 52%) ] Loss: 0.0137 top1= 99.6875
[E130B50 |  32640/50000 ( 65%) ] Loss: 0.0250 top1= 99.5312
[E130B60 |  39040/50000 ( 78%) ] Loss: 0.0136 top1= 99.6875
[E130B70 |  45440/50000 ( 91%) ] Loss: 0.0127 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1363 top1= 73.0869


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3446 top1= 53.3053


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1185 top1= 54.8878

Train epoch 131
[E131B0  |    640/50000 (  1%) ] Loss: 0.0104 top1= 99.8438
[E131B10 |   7040/50000 ( 14%) ] Loss: 0.0130 top1= 99.5312
[E131B20 |  13440/50000 ( 27%) ] Loss: 0.0091 top1= 99.6875
[E131B30 |  19840/50000 ( 40%) ] Loss: 0.0108 top1= 99.6875
[E131B40 |  26240/50000 ( 52%) ] Loss: 0.0119 top1= 99.8438
[E131B50 |  32640/50000 ( 65%) ] Loss: 0.0102 top1= 99.6875
[E131B60 |  39040/50000 ( 78%) ] Loss: 0.0031 top1=100.0000
[E131B70 |  45440/50000 ( 91%) ] Loss: 0.0031 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1093 top1= 72.7764


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.9669 top1= 50.2504


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0190 top1= 54.7877

Train epoch 132
[E132B0  |    640/50000 (  1%) ] Loss: 0.0088 top1= 99.6875
[E132B10 |   7040/50000 ( 14%) ] Loss: 0.0724 top1= 99.0625
[E132B20 |  13440/50000 ( 27%) ] Loss: 0.0051 top1= 99.8438
[E132B30 |  19840/50000 ( 40%) ] Loss: 0.0127 top1= 99.3750
[E132B40 |  26240/50000 ( 52%) ] Loss: 0.0091 top1= 99.6875
[E132B50 |  32640/50000 ( 65%) ] Loss: 0.0123 top1= 99.8438
[E132B60 |  39040/50000 ( 78%) ] Loss: 0.0024 top1=100.0000
[E132B70 |  45440/50000 ( 91%) ] Loss: 0.0055 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2106 top1= 72.5260


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.6375 top1= 49.2188


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8009 top1= 55.3385

Train epoch 133
[E133B0  |    640/50000 (  1%) ] Loss: 0.0137 top1= 99.6875
[E133B10 |   7040/50000 ( 14%) ] Loss: 0.0276 top1= 98.9062
[E133B20 |  13440/50000 ( 27%) ] Loss: 0.0123 top1= 99.6875
[E133B30 |  19840/50000 ( 40%) ] Loss: 0.0095 top1= 99.6875
[E133B40 |  26240/50000 ( 52%) ] Loss: 0.0107 top1= 99.6875
[E133B50 |  32640/50000 ( 65%) ] Loss: 0.0143 top1= 99.3750
[E133B60 |  39040/50000 ( 78%) ] Loss: 0.0094 top1= 99.6875
[E133B70 |  45440/50000 ( 91%) ] Loss: 0.0101 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1522 top1= 72.7163


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.1303 top1= 49.8898


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4261 top1= 51.8429

Train epoch 134
[E134B0  |    640/50000 (  1%) ] Loss: 0.0241 top1= 99.2188
[E134B10 |   7040/50000 ( 14%) ] Loss: 0.0255 top1= 99.0625
[E134B20 |  13440/50000 ( 27%) ] Loss: 0.0103 top1= 99.8438
[E134B30 |  19840/50000 ( 40%) ] Loss: 0.0091 top1= 99.6875
[E134B40 |  26240/50000 ( 52%) ] Loss: 0.0033 top1=100.0000
[E134B50 |  32640/50000 ( 65%) ] Loss: 0.0238 top1= 99.5312
[E134B60 |  39040/50000 ( 78%) ] Loss: 0.0023 top1=100.0000
[E134B70 |  45440/50000 ( 91%) ] Loss: 0.0092 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1790 top1= 73.0369


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0014 top1= 52.0933


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8249 top1= 53.1050

Train epoch 135
[E135B0  |    640/50000 (  1%) ] Loss: 0.0279 top1= 99.2188
[E135B10 |   7040/50000 ( 14%) ] Loss: 0.0132 top1= 99.3750
[E135B20 |  13440/50000 ( 27%) ] Loss: 0.0266 top1= 99.3750
[E135B30 |  19840/50000 ( 40%) ] Loss: 0.0110 top1= 99.5312
[E135B40 |  26240/50000 ( 52%) ] Loss: 0.0109 top1= 99.5312
[E135B50 |  32640/50000 ( 65%) ] Loss: 0.0105 top1= 99.5312
[E135B60 |  39040/50000 ( 78%) ] Loss: 0.0057 top1= 99.8438
[E135B70 |  45440/50000 ( 91%) ] Loss: 0.0167 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2486 top1= 72.0853


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.7922 top1= 48.8882


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7320 top1= 55.4187

Train epoch 136
[E136B0  |    640/50000 (  1%) ] Loss: 0.0118 top1= 99.6875
[E136B10 |   7040/50000 ( 14%) ] Loss: 0.0178 top1= 99.3750
[E136B20 |  13440/50000 ( 27%) ] Loss: 0.0128 top1= 99.6875
[E136B30 |  19840/50000 ( 40%) ] Loss: 0.0095 top1= 99.5312
[E136B40 |  26240/50000 ( 52%) ] Loss: 0.0130 top1= 99.6875
[E136B50 |  32640/50000 ( 65%) ] Loss: 0.0068 top1= 99.6875
[E136B60 |  39040/50000 ( 78%) ] Loss: 0.0341 top1= 99.3750
[E136B70 |  45440/50000 ( 91%) ] Loss: 0.0160 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6344 top1= 70.1723


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8634 top1= 46.1939


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7911 top1= 55.7692

Train epoch 137
[E137B0  |    640/50000 (  1%) ] Loss: 0.0133 top1= 99.6875
[E137B10 |   7040/50000 ( 14%) ] Loss: 0.0191 top1= 99.2188
[E137B20 |  13440/50000 ( 27%) ] Loss: 0.0075 top1= 99.8438
[E137B30 |  19840/50000 ( 40%) ] Loss: 0.0190 top1= 99.5312
[E137B40 |  26240/50000 ( 52%) ] Loss: 0.0113 top1= 99.6875
[E137B50 |  32640/50000 ( 65%) ] Loss: 0.0220 top1= 99.8438
[E137B60 |  39040/50000 ( 78%) ] Loss: 0.0139 top1= 99.6875
[E137B70 |  45440/50000 ( 91%) ] Loss: 0.0046 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1921 top1= 72.7664


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.9350 top1= 48.8482


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0183 top1= 53.0248

Train epoch 138
[E138B0  |    640/50000 (  1%) ] Loss: 0.0135 top1= 99.5312
[E138B10 |   7040/50000 ( 14%) ] Loss: 0.0092 top1= 99.6875
[E138B20 |  13440/50000 ( 27%) ] Loss: 0.0208 top1= 99.3750
[E138B30 |  19840/50000 ( 40%) ] Loss: 0.0098 top1= 99.5312
[E138B40 |  26240/50000 ( 52%) ] Loss: 0.0213 top1= 99.2188
[E138B50 |  32640/50000 ( 65%) ] Loss: 0.0140 top1= 99.5312
[E138B60 |  39040/50000 ( 78%) ] Loss: 0.0359 top1= 99.2188
[E138B70 |  45440/50000 ( 91%) ] Loss: 0.0031 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3316 top1= 71.6947


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.6492 top1= 47.7664


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1189 top1= 55.1382

Train epoch 139
[E139B0  |    640/50000 (  1%) ] Loss: 0.0088 top1= 99.6875
[E139B10 |   7040/50000 ( 14%) ] Loss: 0.0205 top1= 99.2188
[E139B20 |  13440/50000 ( 27%) ] Loss: 0.0094 top1= 99.6875
[E139B30 |  19840/50000 ( 40%) ] Loss: 0.0496 top1= 99.2188
[E139B40 |  26240/50000 ( 52%) ] Loss: 0.0158 top1= 99.5312
[E139B50 |  32640/50000 ( 65%) ] Loss: 0.0067 top1= 99.8438
[E139B60 |  39040/50000 ( 78%) ] Loss: 0.0173 top1= 99.8438
[E139B70 |  45440/50000 ( 91%) ] Loss: 0.0043 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1779 top1= 72.7764


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6778 top1= 53.1250


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2759 top1= 54.3369

Train epoch 140
[E140B0  |    640/50000 (  1%) ] Loss: 0.0211 top1= 98.9062
[E140B10 |   7040/50000 ( 14%) ] Loss: 0.0465 top1= 99.0625
[E140B20 |  13440/50000 ( 27%) ] Loss: 0.0098 top1= 99.6875
[E140B30 |  19840/50000 ( 40%) ] Loss: 0.0257 top1= 99.3750
[E140B40 |  26240/50000 ( 52%) ] Loss: 0.0069 top1= 99.8438
[E140B50 |  32640/50000 ( 65%) ] Loss: 0.0101 top1= 99.6875
[E140B60 |  39040/50000 ( 78%) ] Loss: 0.0066 top1= 99.8438
[E140B70 |  45440/50000 ( 91%) ] Loss: 0.0109 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1783 top1= 72.7564


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.3962 top1= 49.7796


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8487 top1= 52.9848

Train epoch 141
[E141B0  |    640/50000 (  1%) ] Loss: 0.0071 top1= 99.8438
[E141B10 |   7040/50000 ( 14%) ] Loss: 0.0154 top1= 99.6875
[E141B20 |  13440/50000 ( 27%) ] Loss: 0.0128 top1= 99.5312
[E141B30 |  19840/50000 ( 40%) ] Loss: 0.0214 top1= 99.5312
[E141B40 |  26240/50000 ( 52%) ] Loss: 0.0069 top1= 99.6875
[E141B50 |  32640/50000 ( 65%) ] Loss: 0.0112 top1= 99.8438
[E141B60 |  39040/50000 ( 78%) ] Loss: 0.0152 top1= 99.3750
[E141B70 |  45440/50000 ( 91%) ] Loss: 0.0049 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1657 top1= 72.5060


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4259 top1= 49.7396


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8285 top1= 52.8946

Train epoch 142
[E142B0  |    640/50000 (  1%) ] Loss: 0.0171 top1= 99.2188
[E142B10 |   7040/50000 ( 14%) ] Loss: 0.0242 top1= 99.5312
[E142B20 |  13440/50000 ( 27%) ] Loss: 0.0052 top1= 99.8438
[E142B30 |  19840/50000 ( 40%) ] Loss: 0.0095 top1= 99.8438
[E142B40 |  26240/50000 ( 52%) ] Loss: 0.0183 top1= 99.5312
[E142B50 |  32640/50000 ( 65%) ] Loss: 0.0144 top1= 99.3750
[E142B60 |  39040/50000 ( 78%) ] Loss: 0.0175 top1= 99.8438
[E142B70 |  45440/50000 ( 91%) ] Loss: 0.0106 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3116 top1= 72.3558


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0642 top1= 52.3538


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.9470 top1= 50.9014

Train epoch 143
[E143B0  |    640/50000 (  1%) ] Loss: 0.0055 top1=100.0000
[E143B10 |   7040/50000 ( 14%) ] Loss: 0.0127 top1= 99.6875
[E143B20 |  13440/50000 ( 27%) ] Loss: 0.0048 top1= 99.8438
[E143B30 |  19840/50000 ( 40%) ] Loss: 0.0081 top1= 99.6875
[E143B40 |  26240/50000 ( 52%) ] Loss: 0.0103 top1= 99.5312
[E143B50 |  32640/50000 ( 65%) ] Loss: 0.0235 top1= 99.8438
[E143B60 |  39040/50000 ( 78%) ] Loss: 0.0122 top1= 99.3750
[E143B70 |  45440/50000 ( 91%) ] Loss: 0.0267 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6609 top1= 70.5329


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.8082 top1= 45.8033


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7542 top1= 53.0248

Train epoch 144
[E144B0  |    640/50000 (  1%) ] Loss: 0.0209 top1= 99.3750
[E144B10 |   7040/50000 ( 14%) ] Loss: 0.0244 top1= 99.5312
[E144B20 |  13440/50000 ( 27%) ] Loss: 0.0074 top1= 99.6875
[E144B30 |  19840/50000 ( 40%) ] Loss: 0.0292 top1= 99.3750
[E144B40 |  26240/50000 ( 52%) ] Loss: 0.0192 top1= 99.2188
[E144B50 |  32640/50000 ( 65%) ] Loss: 0.0173 top1= 99.2188
[E144B60 |  39040/50000 ( 78%) ] Loss: 0.0054 top1= 99.8438
[E144B70 |  45440/50000 ( 91%) ] Loss: 0.0072 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1709 top1= 73.1070


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.9923 top1= 52.4840


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4343 top1= 53.9663

Train epoch 145
[E145B0  |    640/50000 (  1%) ] Loss: 0.0292 top1= 99.3750
[E145B10 |   7040/50000 ( 14%) ] Loss: 0.0418 top1= 98.9062
[E145B20 |  13440/50000 ( 27%) ] Loss: 0.0130 top1= 99.5312
[E145B30 |  19840/50000 ( 40%) ] Loss: 0.0149 top1= 99.5312
[E145B40 |  26240/50000 ( 52%) ] Loss: 0.0115 top1= 99.8438
[E145B50 |  32640/50000 ( 65%) ] Loss: 0.0195 top1= 99.2188
[E145B60 |  39040/50000 ( 78%) ] Loss: 0.0045 top1= 99.8438
[E145B70 |  45440/50000 ( 91%) ] Loss: 0.0049 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1780 top1= 72.6963


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.1222 top1= 52.3137


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4500 top1= 53.6058

Train epoch 146
[E146B0  |    640/50000 (  1%) ] Loss: 0.0169 top1= 99.3750
[E146B10 |   7040/50000 ( 14%) ] Loss: 0.0104 top1= 99.6875
[E146B20 |  13440/50000 ( 27%) ] Loss: 0.0087 top1= 99.6875
[E146B30 |  19840/50000 ( 40%) ] Loss: 0.0086 top1= 99.8438
[E146B40 |  26240/50000 ( 52%) ] Loss: 0.0246 top1= 99.3750
[E146B50 |  32640/50000 ( 65%) ] Loss: 0.0116 top1= 99.6875
[E146B60 |  39040/50000 ( 78%) ] Loss: 0.0205 top1= 99.2188
[E146B70 |  45440/50000 ( 91%) ] Loss: 0.0062 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2753 top1= 71.8049


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.0820 top1= 48.7280


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7763 top1= 53.4355

Train epoch 147
[E147B0  |    640/50000 (  1%) ] Loss: 0.0221 top1= 99.6875
[E147B10 |   7040/50000 ( 14%) ] Loss: 0.0113 top1= 99.6875
[E147B20 |  13440/50000 ( 27%) ] Loss: 0.0066 top1= 99.6875
[E147B30 |  19840/50000 ( 40%) ] Loss: 0.0092 top1= 99.6875
[E147B40 |  26240/50000 ( 52%) ] Loss: 0.0212 top1= 99.0625
[E147B50 |  32640/50000 ( 65%) ] Loss: 0.0115 top1= 99.5312
[E147B60 |  39040/50000 ( 78%) ] Loss: 0.0182 top1= 99.6875
[E147B70 |  45440/50000 ( 91%) ] Loss: 0.0074 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2046 top1= 72.7364


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8604 top1= 50.7312


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4318 top1= 51.7528

Train epoch 148
[E148B0  |    640/50000 (  1%) ] Loss: 0.0192 top1= 99.3750
[E148B10 |   7040/50000 ( 14%) ] Loss: 0.0125 top1= 99.6875
[E148B20 |  13440/50000 ( 27%) ] Loss: 0.0124 top1= 99.6875
[E148B30 |  19840/50000 ( 40%) ] Loss: 0.0152 top1= 99.6875
[E148B40 |  26240/50000 ( 52%) ] Loss: 0.0106 top1= 99.6875
[E148B50 |  32640/50000 ( 65%) ] Loss: 0.0142 top1= 99.5312
[E148B60 |  39040/50000 ( 78%) ] Loss: 0.0099 top1= 99.5312
[E148B70 |  45440/50000 ( 91%) ] Loss: 0.0157 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2786 top1= 72.1254


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2287 top1= 52.0533


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3113 top1= 52.0433

Train epoch 149
[E149B0  |    640/50000 (  1%) ] Loss: 0.0168 top1= 99.5312
[E149B10 |   7040/50000 ( 14%) ] Loss: 0.0167 top1= 99.8438
[E149B20 |  13440/50000 ( 27%) ] Loss: 0.0084 top1= 99.5312
[E149B30 |  19840/50000 ( 40%) ] Loss: 0.0171 top1= 99.3750
[E149B40 |  26240/50000 ( 52%) ] Loss: 0.0031 top1=100.0000
[E149B50 |  32640/50000 ( 65%) ] Loss: 0.0050 top1=100.0000
[E149B60 |  39040/50000 ( 78%) ] Loss: 0.0198 top1= 99.3750
[E149B70 |  45440/50000 ( 91%) ] Loss: 0.0050 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1710 top1= 72.6763


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.8794 top1= 49.0986


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0139 top1= 52.3438

Train epoch 150
[E150B0  |    640/50000 (  1%) ] Loss: 0.0138 top1= 99.5312
[E150B10 |   7040/50000 ( 14%) ] Loss: 0.0125 top1= 99.3750
[E150B20 |  13440/50000 ( 27%) ] Loss: 0.0116 top1= 99.6875
[E150B30 |  19840/50000 ( 40%) ] Loss: 0.0108 top1= 99.5312
[E150B40 |  26240/50000 ( 52%) ] Loss: 0.0081 top1= 99.6875
[E150B50 |  32640/50000 ( 65%) ] Loss: 0.0150 top1= 99.5312
[E150B60 |  39040/50000 ( 78%) ] Loss: 0.0129 top1= 99.6875
[E150B70 |  45440/50000 ( 91%) ] Loss: 0.0096 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2542 top1= 72.1955


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.5210 top1= 49.7296


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7781 top1= 53.6058

