
=== 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 0x7f793ece4730>

{'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: 2.3036 top1=  9.3750

=== 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: 2.0292 top1= 19.3750
[E 1B20 |  13440/50000 ( 27%) ] Loss: 1.8930 top1= 18.4375
[E 1B30 |  19840/50000 ( 40%) ] Loss: 1.7033 top1= 20.9375
[E 1B40 |  26240/50000 ( 52%) ] Loss: 1.6109 top1= 26.7188
[E 1B50 |  32640/50000 ( 65%) ] Loss: 1.8126 top1= 19.0625
[E 1B60 |  39040/50000 ( 78%) ] Loss: 1.6851 top1= 24.6875
[E 1B70 |  45440/50000 ( 91%) ] Loss: 1.6192 top1= 23.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3799 top1= 10.0060


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.8298 top1= 10.7672


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=3.2318 top1= 16.0857

Train epoch 2
[E 2B0  |    640/50000 (  1%) ] Loss: 1.6062 top1= 27.9688
[E 2B10 |   7040/50000 ( 14%) ] Loss: 1.5406 top1= 30.3125
[E 2B20 |  13440/50000 ( 27%) ] Loss: 1.5372 top1= 28.9062
[E 2B30 |  19840/50000 ( 40%) ] Loss: 1.4762 top1= 33.5938
[E 2B40 |  26240/50000 ( 52%) ] Loss: 1.6039 top1= 26.2500
[E 2B50 |  32640/50000 ( 65%) ] Loss: 1.5198 top1= 27.8125
[E 2B60 |  39040/50000 ( 78%) ] Loss: 1.5798 top1= 28.5938
[E 2B70 |  45440/50000 ( 91%) ] Loss: 1.4884 top1= 30.9375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2707 top1= 15.8053


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.6777 top1= 10.5569


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.1055 top1= 23.2873

Train epoch 3
[E 3B0  |    640/50000 (  1%) ] Loss: 1.5035 top1= 30.0000
[E 3B10 |   7040/50000 ( 14%) ] Loss: 1.5840 top1= 26.2500
[E 3B20 |  13440/50000 ( 27%) ] Loss: 1.5184 top1= 28.4375
[E 3B30 |  19840/50000 ( 40%) ] Loss: 1.4543 top1= 35.6250
[E 3B40 |  26240/50000 ( 52%) ] Loss: 1.3742 top1= 37.9688
[E 3B50 |  32640/50000 ( 65%) ] Loss: 1.4001 top1= 38.1250
[E 3B60 |  39040/50000 ( 78%) ] Loss: 1.3741 top1= 41.2500
[E 3B70 |  45440/50000 ( 91%) ] Loss: 1.3758 top1= 37.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1480 top1= 22.6462


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.4265 top1= 20.7131


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9054 top1= 24.8898

Train epoch 4
[E 4B0  |    640/50000 (  1%) ] Loss: 1.3880 top1= 42.1875
[E 4B10 |   7040/50000 ( 14%) ] Loss: 1.2461 top1= 46.4062
[E 4B20 |  13440/50000 ( 27%) ] Loss: 1.2530 top1= 45.6250
[E 4B30 |  19840/50000 ( 40%) ] Loss: 1.2224 top1= 49.6875
[E 4B40 |  26240/50000 ( 52%) ] Loss: 1.2364 top1= 49.3750
[E 4B50 |  32640/50000 ( 65%) ] Loss: 1.1968 top1= 51.8750
[E 4B60 |  39040/50000 ( 78%) ] Loss: 1.1796 top1= 51.8750
[E 4B70 |  45440/50000 ( 91%) ] Loss: 1.1498 top1= 53.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0003 top1= 26.6326


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.6348 top1= 23.8882


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8164 top1= 28.8862

Train epoch 5
[E 5B0  |    640/50000 (  1%) ] Loss: 1.1544 top1= 53.4375
[E 5B10 |   7040/50000 ( 14%) ] Loss: 1.0981 top1= 57.0312
[E 5B20 |  13440/50000 ( 27%) ] Loss: 1.0682 top1= 55.6250
[E 5B30 |  19840/50000 ( 40%) ] Loss: 1.1130 top1= 57.8125
[E 5B40 |  26240/50000 ( 52%) ] Loss: 1.0120 top1= 59.6875
[E 5B50 |  32640/50000 ( 65%) ] Loss: 1.0252 top1= 59.2188
[E 5B60 |  39040/50000 ( 78%) ] Loss: 1.1016 top1= 57.8125
[E 5B70 |  45440/50000 ( 91%) ] Loss: 1.0289 top1= 59.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9691 top1= 34.1847


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.4964 top1= 29.3870


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5942 top1= 34.0244

Train epoch 6
[E 6B0  |    640/50000 (  1%) ] Loss: 0.9837 top1= 61.0938
[E 6B10 |   7040/50000 ( 14%) ] Loss: 1.0258 top1= 60.4688
[E 6B20 |  13440/50000 ( 27%) ] Loss: 0.9946 top1= 58.7500
[E 6B30 |  19840/50000 ( 40%) ] Loss: 0.9631 top1= 63.9062
[E 6B40 |  26240/50000 ( 52%) ] Loss: 0.9053 top1= 62.6562
[E 6B50 |  32640/50000 ( 65%) ] Loss: 0.9435 top1= 62.6562
[E 6B60 |  39040/50000 ( 78%) ] Loss: 0.9207 top1= 64.8438
[E 6B70 |  45440/50000 ( 91%) ] Loss: 0.8956 top1= 64.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7475 top1= 38.5417


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2036 top1= 30.2384


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6333 top1= 35.7171

Train epoch 7
[E 7B0  |    640/50000 (  1%) ] Loss: 0.9254 top1= 64.6875
[E 7B10 |   7040/50000 ( 14%) ] Loss: 0.8887 top1= 65.3125
[E 7B20 |  13440/50000 ( 27%) ] Loss: 0.8411 top1= 66.8750
[E 7B30 |  19840/50000 ( 40%) ] Loss: 0.8780 top1= 67.5000
[E 7B40 |  26240/50000 ( 52%) ] Loss: 0.8361 top1= 69.0625
[E 7B50 |  32640/50000 ( 65%) ] Loss: 0.8747 top1= 63.2812
[E 7B60 |  39040/50000 ( 78%) ] Loss: 0.7799 top1= 68.7500
[E 7B70 |  45440/50000 ( 91%) ] Loss: 0.8427 top1= 65.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7208 top1= 41.3361


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3591 top1= 31.3602


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1913 top1= 36.6987

Train epoch 8
[E 8B0  |    640/50000 (  1%) ] Loss: 0.9147 top1= 62.9688
[E 8B10 |   7040/50000 ( 14%) ] Loss: 0.8729 top1= 67.5000
[E 8B20 |  13440/50000 ( 27%) ] Loss: 0.8086 top1= 68.5938
[E 8B30 |  19840/50000 ( 40%) ] Loss: 0.8253 top1= 66.0938
[E 8B40 |  26240/50000 ( 52%) ] Loss: 0.7399 top1= 71.2500
[E 8B50 |  32640/50000 ( 65%) ] Loss: 0.8610 top1= 67.0312
[E 8B60 |  39040/50000 ( 78%) ] Loss: 0.8211 top1= 69.3750
[E 8B70 |  45440/50000 ( 91%) ] Loss: 0.7735 top1= 68.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6230 top1= 43.3594


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.3910 top1= 33.9643


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.4737 top1= 37.5100

Train epoch 9
[E 9B0  |    640/50000 (  1%) ] Loss: 0.8051 top1= 67.8125
[E 9B10 |   7040/50000 ( 14%) ] Loss: 0.8391 top1= 67.0312
[E 9B20 |  13440/50000 ( 27%) ] Loss: 0.7486 top1= 71.8750
[E 9B30 |  19840/50000 ( 40%) ] Loss: 0.7292 top1= 73.1250
[E 9B40 |  26240/50000 ( 52%) ] Loss: 0.6952 top1= 74.3750
[E 9B50 |  32640/50000 ( 65%) ] Loss: 0.7424 top1= 71.7188
[E 9B60 |  39040/50000 ( 78%) ] Loss: 0.7153 top1= 71.5625
[E 9B70 |  45440/50000 ( 91%) ] Loss: 0.7054 top1= 71.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6337 top1= 47.3157


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1685 top1= 34.6354


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8952 top1= 38.9724

Train epoch 10
[E10B0  |    640/50000 (  1%) ] Loss: 0.7511 top1= 71.2500
[E10B10 |   7040/50000 ( 14%) ] Loss: 0.7395 top1= 73.5938
[E10B20 |  13440/50000 ( 27%) ] Loss: 0.7150 top1= 72.5000
[E10B30 |  19840/50000 ( 40%) ] Loss: 0.6840 top1= 73.7500
[E10B40 |  26240/50000 ( 52%) ] Loss: 0.6704 top1= 75.4688
[E10B50 |  32640/50000 ( 65%) ] Loss: 0.7450 top1= 72.5000
[E10B60 |  39040/50000 ( 78%) ] Loss: 0.6659 top1= 74.5312
[E10B70 |  45440/50000 ( 91%) ] Loss: 0.6585 top1= 74.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7217 top1= 46.0837


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1974 top1= 35.4868


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0175 top1= 38.0809

Train epoch 11
[E11B0  |    640/50000 (  1%) ] Loss: 0.6954 top1= 73.4375
[E11B10 |   7040/50000 ( 14%) ] Loss: 0.6919 top1= 73.9062
[E11B20 |  13440/50000 ( 27%) ] Loss: 0.6355 top1= 75.0000
[E11B30 |  19840/50000 ( 40%) ] Loss: 0.6220 top1= 76.7188
[E11B40 |  26240/50000 ( 52%) ] Loss: 0.6722 top1= 74.8438
[E11B50 |  32640/50000 ( 65%) ] Loss: 0.7387 top1= 73.4375
[E11B60 |  39040/50000 ( 78%) ] Loss: 0.6448 top1= 74.6875
[E11B70 |  45440/50000 ( 91%) ] Loss: 0.6326 top1= 77.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5804 top1= 49.9800


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5669 top1= 34.3049


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5174 top1= 40.4547

Train epoch 12
[E12B0  |    640/50000 (  1%) ] Loss: 0.6774 top1= 71.2500
[E12B10 |   7040/50000 ( 14%) ] Loss: 0.6463 top1= 75.0000
[E12B20 |  13440/50000 ( 27%) ] Loss: 0.6083 top1= 75.9375
[E12B30 |  19840/50000 ( 40%) ] Loss: 0.6043 top1= 75.7812
[E12B40 |  26240/50000 ( 52%) ] Loss: 0.5681 top1= 76.5625
[E12B50 |  32640/50000 ( 65%) ] Loss: 0.6293 top1= 76.8750
[E12B60 |  39040/50000 ( 78%) ] Loss: 0.5861 top1= 78.5938
[E12B70 |  45440/50000 ( 91%) ] Loss: 0.5609 top1= 78.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5177 top1= 51.0717


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1674 top1= 36.4483


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9177 top1= 40.7252

Train epoch 13
[E13B0  |    640/50000 (  1%) ] Loss: 0.5911 top1= 76.0938
[E13B10 |   7040/50000 ( 14%) ] Loss: 0.6122 top1= 76.2500
[E13B20 |  13440/50000 ( 27%) ] Loss: 0.5470 top1= 80.7812
[E13B30 |  19840/50000 ( 40%) ] Loss: 0.5835 top1= 77.8125
[E13B40 |  26240/50000 ( 52%) ] Loss: 0.6009 top1= 78.2812
[E13B50 |  32640/50000 ( 65%) ] Loss: 0.6284 top1= 79.0625
[E13B60 |  39040/50000 ( 78%) ] Loss: 0.5477 top1= 78.7500
[E13B70 |  45440/50000 ( 91%) ] Loss: 0.5827 top1= 79.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5127 top1= 52.2035


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7756 top1= 37.0593


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4708 top1= 41.8069

Train epoch 14
[E14B0  |    640/50000 (  1%) ] Loss: 0.5554 top1= 78.9062
[E14B10 |   7040/50000 ( 14%) ] Loss: 0.5319 top1= 80.0000
[E14B20 |  13440/50000 ( 27%) ] Loss: 0.5060 top1= 81.5625
[E14B30 |  19840/50000 ( 40%) ] Loss: 0.4961 top1= 80.9375
[E14B40 |  26240/50000 ( 52%) ] Loss: 0.5268 top1= 80.1562
[E14B50 |  32640/50000 ( 65%) ] Loss: 0.6029 top1= 77.6562
[E14B60 |  39040/50000 ( 78%) ] Loss: 0.5422 top1= 79.8438
[E14B70 |  45440/50000 ( 91%) ] Loss: 0.4998 top1= 81.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7375 top1= 51.5024


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7636 top1= 37.5901


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1189 top1= 41.2861

Train epoch 15
[E15B0  |    640/50000 (  1%) ] Loss: 0.5635 top1= 79.0625
[E15B10 |   7040/50000 ( 14%) ] Loss: 0.5510 top1= 80.4688
[E15B20 |  13440/50000 ( 27%) ] Loss: 0.4989 top1= 82.0312
[E15B30 |  19840/50000 ( 40%) ] Loss: 0.4859 top1= 83.2812
[E15B40 |  26240/50000 ( 52%) ] Loss: 0.5176 top1= 80.9375
[E15B50 |  32640/50000 ( 65%) ] Loss: 0.5467 top1= 79.6875
[E15B60 |  39040/50000 ( 78%) ] Loss: 0.5412 top1= 80.0000
[E15B70 |  45440/50000 ( 91%) ] Loss: 0.4744 top1= 82.3438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4425 top1= 53.5757


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3895 top1= 37.9307


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9142 top1= 41.5665

Train epoch 16
[E16B0  |    640/50000 (  1%) ] Loss: 0.5423 top1= 80.3125
[E16B10 |   7040/50000 ( 14%) ] Loss: 0.5374 top1= 81.0938
[E16B20 |  13440/50000 ( 27%) ] Loss: 0.4663 top1= 82.6562
[E16B30 |  19840/50000 ( 40%) ] Loss: 0.4736 top1= 82.8125
[E16B40 |  26240/50000 ( 52%) ] Loss: 0.4450 top1= 83.4375
[E16B50 |  32640/50000 ( 65%) ] Loss: 0.5429 top1= 79.5312
[E16B60 |  39040/50000 ( 78%) ] Loss: 0.4907 top1= 81.8750
[E16B70 |  45440/50000 ( 91%) ] Loss: 0.4689 top1= 82.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4521 top1= 56.9511


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1291 top1= 38.8421


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3616 top1= 42.6082

Train epoch 17
[E17B0  |    640/50000 (  1%) ] Loss: 0.5073 top1= 80.6250
[E17B10 |   7040/50000 ( 14%) ] Loss: 0.4762 top1= 81.5625
[E17B20 |  13440/50000 ( 27%) ] Loss: 0.4531 top1= 82.8125
[E17B30 |  19840/50000 ( 40%) ] Loss: 0.4334 top1= 83.7500
[E17B40 |  26240/50000 ( 52%) ] Loss: 0.4576 top1= 82.5000
[E17B50 |  32640/50000 ( 65%) ] Loss: 0.4309 top1= 85.0000
[E17B60 |  39040/50000 ( 78%) ] Loss: 0.4650 top1= 82.5000
[E17B70 |  45440/50000 ( 91%) ] Loss: 0.4377 top1= 83.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5104 top1= 54.5573


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2393 top1= 39.2628


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6868 top1= 42.5481

Train epoch 18
[E18B0  |    640/50000 (  1%) ] Loss: 0.4915 top1= 81.2500
[E18B10 |   7040/50000 ( 14%) ] Loss: 0.4486 top1= 82.8125
[E18B20 |  13440/50000 ( 27%) ] Loss: 0.3859 top1= 85.1562
[E18B30 |  19840/50000 ( 40%) ] Loss: 0.4348 top1= 83.7500
[E18B40 |  26240/50000 ( 52%) ] Loss: 0.3647 top1= 86.7188
[E18B50 |  32640/50000 ( 65%) ] Loss: 0.4927 top1= 82.3438
[E18B60 |  39040/50000 ( 78%) ] Loss: 0.4545 top1= 82.9688
[E18B70 |  45440/50000 ( 91%) ] Loss: 0.4280 top1= 84.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6225 top1= 54.8678


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0038 top1= 39.2328


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8442 top1= 42.2776

Train epoch 19
[E19B0  |    640/50000 (  1%) ] Loss: 0.4891 top1= 82.3438
[E19B10 |   7040/50000 ( 14%) ] Loss: 0.4855 top1= 81.8750
[E19B20 |  13440/50000 ( 27%) ] Loss: 0.4201 top1= 84.0625
[E19B30 |  19840/50000 ( 40%) ] Loss: 0.4124 top1= 84.3750
[E19B40 |  26240/50000 ( 52%) ] Loss: 0.3934 top1= 85.3125
[E19B50 |  32640/50000 ( 65%) ] Loss: 0.4780 top1= 80.9375
[E19B60 |  39040/50000 ( 78%) ] Loss: 0.4485 top1= 84.0625
[E19B70 |  45440/50000 ( 91%) ] Loss: 0.3786 top1= 86.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9655 top1= 55.8093


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9821 top1= 39.7736


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3850 top1= 42.8586

Train epoch 20
[E20B0  |    640/50000 (  1%) ] Loss: 0.4644 top1= 81.7188
[E20B10 |   7040/50000 ( 14%) ] Loss: 0.4731 top1= 82.8125
[E20B20 |  13440/50000 ( 27%) ] Loss: 0.3924 top1= 85.6250
[E20B30 |  19840/50000 ( 40%) ] Loss: 0.3889 top1= 85.6250
[E20B40 |  26240/50000 ( 52%) ] Loss: 0.3793 top1= 86.2500
[E20B50 |  32640/50000 ( 65%) ] Loss: 0.4018 top1= 85.9375
[E20B60 |  39040/50000 ( 78%) ] Loss: 0.4192 top1= 85.0000
[E20B70 |  45440/50000 ( 91%) ] Loss: 0.3590 top1= 86.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5550 top1= 58.4034


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3074 top1= 39.8137


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0202 top1= 43.5196

Train epoch 21
[E21B0  |    640/50000 (  1%) ] Loss: 0.4290 top1= 84.6875
[E21B10 |   7040/50000 ( 14%) ] Loss: 0.3866 top1= 85.6250
[E21B20 |  13440/50000 ( 27%) ] Loss: 0.3411 top1= 87.1875
[E21B30 |  19840/50000 ( 40%) ] Loss: 0.3698 top1= 85.9375
[E21B40 |  26240/50000 ( 52%) ] Loss: 0.3768 top1= 87.5000
[E21B50 |  32640/50000 ( 65%) ] Loss: 0.4068 top1= 85.6250
[E21B60 |  39040/50000 ( 78%) ] Loss: 0.4105 top1= 85.4688
[E21B70 |  45440/50000 ( 91%) ] Loss: 0.3537 top1= 86.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3441 top1= 62.3498


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9858 top1= 39.8538


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7394 top1= 44.2608

Train epoch 22
[E22B0  |    640/50000 (  1%) ] Loss: 0.3919 top1= 85.6250
[E22B10 |   7040/50000 ( 14%) ] Loss: 0.4009 top1= 86.4062
[E22B20 |  13440/50000 ( 27%) ] Loss: 0.3458 top1= 87.9688
[E22B30 |  19840/50000 ( 40%) ] Loss: 0.3134 top1= 87.8125
[E22B40 |  26240/50000 ( 52%) ] Loss: 0.3620 top1= 86.5625
[E22B50 |  32640/50000 ( 65%) ] Loss: 0.3977 top1= 84.2188
[E22B60 |  39040/50000 ( 78%) ] Loss: 0.3542 top1= 87.5000
[E22B70 |  45440/50000 ( 91%) ] Loss: 0.3429 top1= 86.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4016 top1= 61.4083


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9864 top1= 39.9239


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4979 top1= 43.8401

Train epoch 23
[E23B0  |    640/50000 (  1%) ] Loss: 0.4309 top1= 83.2812
[E23B10 |   7040/50000 ( 14%) ] Loss: 0.4233 top1= 84.3750
[E23B20 |  13440/50000 ( 27%) ] Loss: 0.2918 top1= 89.5312
[E23B30 |  19840/50000 ( 40%) ] Loss: 0.3553 top1= 86.2500
[E23B40 |  26240/50000 ( 52%) ] Loss: 0.3541 top1= 86.8750
[E23B50 |  32640/50000 ( 65%) ] Loss: 0.3338 top1= 86.0938
[E23B60 |  39040/50000 ( 78%) ] Loss: 0.3781 top1= 85.7812
[E23B70 |  45440/50000 ( 91%) ] Loss: 0.3336 top1= 87.1875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4209 top1= 61.0076


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2104 top1= 40.5148


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4465 top1= 43.6198

Train epoch 24
[E24B0  |    640/50000 (  1%) ] Loss: 0.3822 top1= 85.1562
[E24B10 |   7040/50000 ( 14%) ] Loss: 0.4032 top1= 85.0000
[E24B20 |  13440/50000 ( 27%) ] Loss: 0.3611 top1= 85.1562
[E24B30 |  19840/50000 ( 40%) ] Loss: 0.3374 top1= 87.6562
[E24B40 |  26240/50000 ( 52%) ] Loss: 0.3482 top1= 87.3438
[E24B50 |  32640/50000 ( 65%) ] Loss: 0.3231 top1= 87.9688
[E24B60 |  39040/50000 ( 78%) ] Loss: 0.3190 top1= 88.7500
[E24B70 |  45440/50000 ( 91%) ] Loss: 0.3038 top1= 88.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4840 top1= 61.6687


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8369 top1= 41.5365


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0596 top1= 44.9920

Train epoch 25
[E25B0  |    640/50000 (  1%) ] Loss: 0.3374 top1= 88.1250
[E25B10 |   7040/50000 ( 14%) ] Loss: 0.3496 top1= 85.4688
[E25B20 |  13440/50000 ( 27%) ] Loss: 0.3237 top1= 88.7500
[E25B30 |  19840/50000 ( 40%) ] Loss: 0.2984 top1= 89.2188
[E25B40 |  26240/50000 ( 52%) ] Loss: 0.2982 top1= 89.0625
[E25B50 |  32640/50000 ( 65%) ] Loss: 0.3136 top1= 87.6562
[E25B60 |  39040/50000 ( 78%) ] Loss: 0.2948 top1= 89.3750
[E25B70 |  45440/50000 ( 91%) ] Loss: 0.2697 top1= 90.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4581 top1= 61.7989


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6565 top1= 40.8454


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8804 top1= 44.3510

Train epoch 26
[E26B0  |    640/50000 (  1%) ] Loss: 0.4059 top1= 85.1562
[E26B10 |   7040/50000 ( 14%) ] Loss: 0.3228 top1= 87.3438
[E26B20 |  13440/50000 ( 27%) ] Loss: 0.3079 top1= 88.2812
[E26B30 |  19840/50000 ( 40%) ] Loss: 0.3086 top1= 88.2812
[E26B40 |  26240/50000 ( 52%) ] Loss: 0.3197 top1= 87.6562
[E26B50 |  32640/50000 ( 65%) ] Loss: 0.3461 top1= 86.4062
[E26B60 |  39040/50000 ( 78%) ] Loss: 0.3467 top1= 87.1875
[E26B70 |  45440/50000 ( 91%) ] Loss: 0.3077 top1= 89.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5665 top1= 60.6871


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5540 top1= 41.3361


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7785 top1= 43.4896

Train epoch 27
[E27B0  |    640/50000 (  1%) ] Loss: 0.3411 top1= 86.5625
[E27B10 |   7040/50000 ( 14%) ] Loss: 0.3275 top1= 88.7500
[E27B20 |  13440/50000 ( 27%) ] Loss: 0.2962 top1= 89.0625
[E27B30 |  19840/50000 ( 40%) ] Loss: 0.3022 top1= 88.1250
[E27B40 |  26240/50000 ( 52%) ] Loss: 0.3253 top1= 88.5938
[E27B50 |  32640/50000 ( 65%) ] Loss: 0.2980 top1= 88.5938
[E27B60 |  39040/50000 ( 78%) ] Loss: 0.3196 top1= 87.1875
[E27B70 |  45440/50000 ( 91%) ] Loss: 0.2894 top1= 88.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4468 top1= 62.4800


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4994 top1= 41.3662


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.6166 top1= 44.6114

Train epoch 28
[E28B0  |    640/50000 (  1%) ] Loss: 0.3490 top1= 87.9688
[E28B10 |   7040/50000 ( 14%) ] Loss: 0.3183 top1= 88.7500
[E28B20 |  13440/50000 ( 27%) ] Loss: 0.2649 top1= 90.0000
[E28B30 |  19840/50000 ( 40%) ] Loss: 0.2975 top1= 89.6875
[E28B40 |  26240/50000 ( 52%) ] Loss: 0.2709 top1= 90.3125
[E28B50 |  32640/50000 ( 65%) ] Loss: 0.3006 top1= 90.1562
[E28B60 |  39040/50000 ( 78%) ] Loss: 0.2490 top1= 90.4688
[E28B70 |  45440/50000 ( 91%) ] Loss: 0.2753 top1= 91.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3557 top1= 65.3846


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1184 top1= 41.8269


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8610 top1= 45.1022

Train epoch 29
[E29B0  |    640/50000 (  1%) ] Loss: 0.3014 top1= 89.6875
[E29B10 |   7040/50000 ( 14%) ] Loss: 0.2948 top1= 90.0000
[E29B20 |  13440/50000 ( 27%) ] Loss: 0.2243 top1= 92.0312
[E29B30 |  19840/50000 ( 40%) ] Loss: 0.3018 top1= 89.0625
[E29B40 |  26240/50000 ( 52%) ] Loss: 0.2655 top1= 90.1562
[E29B50 |  32640/50000 ( 65%) ] Loss: 0.2543 top1= 90.1562
[E29B60 |  39040/50000 ( 78%) ] Loss: 0.2763 top1= 89.8438
[E29B70 |  45440/50000 ( 91%) ] Loss: 0.2763 top1= 90.9375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1588 top1= 60.0761


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0339 top1= 40.9355


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6492 top1= 44.0905

Train epoch 30
[E30B0  |    640/50000 (  1%) ] Loss: 0.3087 top1= 89.0625
[E30B10 |   7040/50000 ( 14%) ] Loss: 0.3354 top1= 87.5000
[E30B20 |  13440/50000 ( 27%) ] Loss: 0.2578 top1= 90.0000
[E30B30 |  19840/50000 ( 40%) ] Loss: 0.2766 top1= 89.0625
[E30B40 |  26240/50000 ( 52%) ] Loss: 0.3153 top1= 89.5312
[E30B50 |  32640/50000 ( 65%) ] Loss: 0.2831 top1= 89.0625
[E30B60 |  39040/50000 ( 78%) ] Loss: 0.2403 top1= 91.0938
[E30B70 |  45440/50000 ( 91%) ] Loss: 0.2392 top1= 90.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4287 top1= 61.6086


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1313 top1= 40.7953


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7492 top1= 44.6214

Train epoch 31
[E31B0  |    640/50000 (  1%) ] Loss: 0.3101 top1= 88.4375
[E31B10 |   7040/50000 ( 14%) ] Loss: 0.2783 top1= 89.8438
[E31B20 |  13440/50000 ( 27%) ] Loss: 0.2474 top1= 90.6250
[E31B30 |  19840/50000 ( 40%) ] Loss: 0.2845 top1= 87.5000
[E31B40 |  26240/50000 ( 52%) ] Loss: 0.2105 top1= 91.8750
[E31B50 |  32640/50000 ( 65%) ] Loss: 0.3249 top1= 87.9688
[E31B60 |  39040/50000 ( 78%) ] Loss: 0.2503 top1= 90.0000
[E31B70 |  45440/50000 ( 91%) ] Loss: 0.2155 top1= 92.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6970 top1= 61.0877


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5129 top1= 41.7668


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.0632 top1= 44.6014

Train epoch 32
[E32B0  |    640/50000 (  1%) ] Loss: 0.2985 top1= 89.6875
[E32B10 |   7040/50000 ( 14%) ] Loss: 0.2896 top1= 89.3750
[E32B20 |  13440/50000 ( 27%) ] Loss: 0.2740 top1= 89.3750
[E32B30 |  19840/50000 ( 40%) ] Loss: 0.2705 top1= 89.2188
[E32B40 |  26240/50000 ( 52%) ] Loss: 0.2692 top1= 90.7812
[E32B50 |  32640/50000 ( 65%) ] Loss: 0.2529 top1= 90.7812
[E32B60 |  39040/50000 ( 78%) ] Loss: 0.2893 top1= 89.6875
[E32B70 |  45440/50000 ( 91%) ] Loss: 0.2122 top1= 91.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7834 top1= 61.6887


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4043 top1= 41.6266


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1291 top1= 45.1923

Train epoch 33
[E33B0  |    640/50000 (  1%) ] Loss: 0.3003 top1= 88.1250
[E33B10 |   7040/50000 ( 14%) ] Loss: 0.2683 top1= 88.9062
[E33B20 |  13440/50000 ( 27%) ] Loss: 0.2609 top1= 89.3750
[E33B30 |  19840/50000 ( 40%) ] Loss: 0.2586 top1= 89.5312
[E33B40 |  26240/50000 ( 52%) ] Loss: 0.2402 top1= 91.4062
[E33B50 |  32640/50000 ( 65%) ] Loss: 0.3061 top1= 88.2812
[E33B60 |  39040/50000 ( 78%) ] Loss: 0.2549 top1= 90.4688
[E33B70 |  45440/50000 ( 91%) ] Loss: 0.2302 top1= 92.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5293 top1= 63.7821


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2300 top1= 41.9772


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7853 top1= 44.7516

Train epoch 34
[E34B0  |    640/50000 (  1%) ] Loss: 0.2420 top1= 90.6250
[E34B10 |   7040/50000 ( 14%) ] Loss: 0.2473 top1= 90.0000
[E34B20 |  13440/50000 ( 27%) ] Loss: 0.2496 top1= 90.3125
[E34B30 |  19840/50000 ( 40%) ] Loss: 0.2866 top1= 89.0625
[E34B40 |  26240/50000 ( 52%) ] Loss: 0.2527 top1= 90.1562
[E34B50 |  32640/50000 ( 65%) ] Loss: 0.2575 top1= 89.6875
[E34B60 |  39040/50000 ( 78%) ] Loss: 0.2210 top1= 92.1875
[E34B70 |  45440/50000 ( 91%) ] Loss: 0.2536 top1= 90.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4869 top1= 63.1010


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5357 top1= 40.7352


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8330 top1= 44.5613

Train epoch 35
[E35B0  |    640/50000 (  1%) ] Loss: 0.3253 top1= 88.7500
[E35B10 |   7040/50000 ( 14%) ] Loss: 0.2892 top1= 90.6250
[E35B20 |  13440/50000 ( 27%) ] Loss: 0.2738 top1= 90.3125
[E35B30 |  19840/50000 ( 40%) ] Loss: 0.2445 top1= 90.6250
[E35B40 |  26240/50000 ( 52%) ] Loss: 0.2737 top1= 90.3125
[E35B50 |  32640/50000 ( 65%) ] Loss: 0.2969 top1= 88.7500
[E35B60 |  39040/50000 ( 78%) ] Loss: 0.2159 top1= 92.3438
[E35B70 |  45440/50000 ( 91%) ] Loss: 0.2427 top1= 90.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4605 top1= 65.4547


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3167 top1= 42.4679


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0752 top1= 45.2825

Train epoch 36
[E36B0  |    640/50000 (  1%) ] Loss: 0.2136 top1= 92.6562
[E36B10 |   7040/50000 ( 14%) ] Loss: 0.2644 top1= 88.9062
[E36B20 |  13440/50000 ( 27%) ] Loss: 0.2644 top1= 91.2500
[E36B30 |  19840/50000 ( 40%) ] Loss: 0.2396 top1= 91.7188
[E36B40 |  26240/50000 ( 52%) ] Loss: 0.2426 top1= 91.5625
[E36B50 |  32640/50000 ( 65%) ] Loss: 0.2814 top1= 87.9688
[E36B60 |  39040/50000 ( 78%) ] Loss: 0.2369 top1= 90.9375
[E36B70 |  45440/50000 ( 91%) ] Loss: 0.2202 top1= 92.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3058 top1= 64.4431


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4577 top1= 41.4363


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4168 top1= 44.7917

Train epoch 37
[E37B0  |    640/50000 (  1%) ] Loss: 0.2871 top1= 89.0625
[E37B10 |   7040/50000 ( 14%) ] Loss: 0.2739 top1= 90.1562
[E37B20 |  13440/50000 ( 27%) ] Loss: 0.2749 top1= 90.4688
[E37B30 |  19840/50000 ( 40%) ] Loss: 0.2524 top1= 90.4688
[E37B40 |  26240/50000 ( 52%) ] Loss: 0.2207 top1= 91.8750
[E37B50 |  32640/50000 ( 65%) ] Loss: 0.2255 top1= 91.2500
[E37B60 |  39040/50000 ( 78%) ] Loss: 0.2153 top1= 92.8125
[E37B70 |  45440/50000 ( 91%) ] Loss: 0.2154 top1= 91.5625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4568 top1= 63.5717


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.0035 top1= 41.8870


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7227 top1= 44.8618

Train epoch 38
[E38B0  |    640/50000 (  1%) ] Loss: 0.2673 top1= 90.1562
[E38B10 |   7040/50000 ( 14%) ] Loss: 0.2472 top1= 90.9375
[E38B20 |  13440/50000 ( 27%) ] Loss: 0.2048 top1= 92.3438
[E38B30 |  19840/50000 ( 40%) ] Loss: 0.2722 top1= 89.6875
[E38B40 |  26240/50000 ( 52%) ] Loss: 0.3073 top1= 89.3750
[E38B50 |  32640/50000 ( 65%) ] Loss: 0.2689 top1= 90.1562
[E38B60 |  39040/50000 ( 78%) ] Loss: 0.2799 top1= 89.0625
[E38B70 |  45440/50000 ( 91%) ] Loss: 0.2042 top1= 91.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2155 top1= 67.5481


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1543 top1= 42.3578


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3667 top1= 45.7933

Train epoch 39
[E39B0  |    640/50000 (  1%) ] Loss: 0.2705 top1= 89.5312
[E39B10 |   7040/50000 ( 14%) ] Loss: 0.2628 top1= 90.0000
[E39B20 |  13440/50000 ( 27%) ] Loss: 0.2287 top1= 92.3438
[E39B30 |  19840/50000 ( 40%) ] Loss: 0.2691 top1= 90.9375
[E39B40 |  26240/50000 ( 52%) ] Loss: 0.1876 top1= 93.4375
[E39B50 |  32640/50000 ( 65%) ] Loss: 0.2253 top1= 91.8750
[E39B60 |  39040/50000 ( 78%) ] Loss: 0.1904 top1= 93.2812
[E39B70 |  45440/50000 ( 91%) ] Loss: 0.2177 top1= 92.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6105 top1= 64.4431


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5502 top1= 41.4964


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.9480 top1= 45.1923

Train epoch 40
[E40B0  |    640/50000 (  1%) ] Loss: 0.2787 top1= 89.0625
[E40B10 |   7040/50000 ( 14%) ] Loss: 0.2610 top1= 90.0000
[E40B20 |  13440/50000 ( 27%) ] Loss: 0.1983 top1= 92.1875
[E40B30 |  19840/50000 ( 40%) ] Loss: 0.2339 top1= 91.0938
[E40B40 |  26240/50000 ( 52%) ] Loss: 0.1630 top1= 94.3750
[E40B50 |  32640/50000 ( 65%) ] Loss: 0.2132 top1= 92.9688
[E40B60 |  39040/50000 ( 78%) ] Loss: 0.1937 top1= 92.5000
[E40B70 |  45440/50000 ( 91%) ] Loss: 0.1945 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.1610 top1= 60.3866


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3046 top1= 41.8570


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.3471 top1= 45.2123

Train epoch 41
[E41B0  |    640/50000 (  1%) ] Loss: 0.2684 top1= 88.4375
[E41B10 |   7040/50000 ( 14%) ] Loss: 0.2474 top1= 90.7812
[E41B20 |  13440/50000 ( 27%) ] Loss: 0.2039 top1= 92.5000
[E41B30 |  19840/50000 ( 40%) ] Loss: 0.1890 top1= 93.4375
[E41B40 |  26240/50000 ( 52%) ] Loss: 0.1665 top1= 94.3750
[E41B50 |  32640/50000 ( 65%) ] Loss: 0.2111 top1= 92.8125
[E41B60 |  39040/50000 ( 78%) ] Loss: 0.2118 top1= 92.8125
[E41B70 |  45440/50000 ( 91%) ] Loss: 0.2347 top1= 91.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2586 top1= 67.6983


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4672 top1= 42.8285


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.8240 top1= 45.4026

Train epoch 42
[E42B0  |    640/50000 (  1%) ] Loss: 0.1847 top1= 94.0625
[E42B10 |   7040/50000 ( 14%) ] Loss: 0.2211 top1= 93.2812
[E42B20 |  13440/50000 ( 27%) ] Loss: 0.1776 top1= 93.5938
[E42B30 |  19840/50000 ( 40%) ] Loss: 0.2381 top1= 91.2500
[E42B40 |  26240/50000 ( 52%) ] Loss: 0.1760 top1= 93.5938
[E42B50 |  32640/50000 ( 65%) ] Loss: 0.2281 top1= 90.9375
[E42B60 |  39040/50000 ( 78%) ] Loss: 0.1789 top1= 93.5938
[E42B70 |  45440/50000 ( 91%) ] Loss: 0.2503 top1= 91.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4469 top1= 66.4263


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3317 top1= 42.0072


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0066 top1= 45.2724

Train epoch 43
[E43B0  |    640/50000 (  1%) ] Loss: 0.2269 top1= 92.5000
[E43B10 |   7040/50000 ( 14%) ] Loss: 0.1976 top1= 92.9688
[E43B20 |  13440/50000 ( 27%) ] Loss: 0.1970 top1= 92.5000
[E43B30 |  19840/50000 ( 40%) ] Loss: 0.2126 top1= 92.1875
[E43B40 |  26240/50000 ( 52%) ] Loss: 0.1995 top1= 92.6562
[E43B50 |  32640/50000 ( 65%) ] Loss: 0.2147 top1= 91.2500
[E43B60 |  39040/50000 ( 78%) ] Loss: 0.2097 top1= 93.2812
[E43B70 |  45440/50000 ( 91%) ] Loss: 0.1728 top1= 93.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9189 top1= 61.8790


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3230 top1= 41.7969


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.7961 top1= 44.6815

Train epoch 44
[E44B0  |    640/50000 (  1%) ] Loss: 0.2781 top1= 90.3125
[E44B10 |   7040/50000 ( 14%) ] Loss: 0.2748 top1= 91.0938
[E44B20 |  13440/50000 ( 27%) ] Loss: 0.2052 top1= 92.3438
[E44B30 |  19840/50000 ( 40%) ] Loss: 0.2124 top1= 92.5000
[E44B40 |  26240/50000 ( 52%) ] Loss: 0.1688 top1= 93.2812
[E44B50 |  32640/50000 ( 65%) ] Loss: 0.1943 top1= 94.2188
[E44B60 |  39040/50000 ( 78%) ] Loss: 0.1571 top1= 94.2188
[E44B70 |  45440/50000 ( 91%) ] Loss: 0.1956 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6546 top1= 64.9639


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4460 top1= 43.0288


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.9408 top1= 45.3926

Train epoch 45
[E45B0  |    640/50000 (  1%) ] Loss: 0.1875 top1= 94.2188
[E45B10 |   7040/50000 ( 14%) ] Loss: 0.2074 top1= 92.8125
[E45B20 |  13440/50000 ( 27%) ] Loss: 0.2039 top1= 93.4375
[E45B30 |  19840/50000 ( 40%) ] Loss: 0.2009 top1= 92.3438
[E45B40 |  26240/50000 ( 52%) ] Loss: 0.1677 top1= 94.6875
[E45B50 |  32640/50000 ( 65%) ] Loss: 0.1954 top1= 91.5625
[E45B60 |  39040/50000 ( 78%) ] Loss: 0.1692 top1= 93.9062
[E45B70 |  45440/50000 ( 91%) ] Loss: 0.2176 top1= 92.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4951 top1= 65.3646


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5560 top1= 42.7484


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.4798 top1= 46.1639

Train epoch 46
[E46B0  |    640/50000 (  1%) ] Loss: 0.2377 top1= 92.0312
[E46B10 |   7040/50000 ( 14%) ] Loss: 0.2213 top1= 92.0312
[E46B20 |  13440/50000 ( 27%) ] Loss: 0.1681 top1= 92.6562
[E46B30 |  19840/50000 ( 40%) ] Loss: 0.1699 top1= 93.9062
[E46B40 |  26240/50000 ( 52%) ] Loss: 0.1907 top1= 93.4375
[E46B50 |  32640/50000 ( 65%) ] Loss: 0.1958 top1= 92.3438
[E46B60 |  39040/50000 ( 78%) ] Loss: 0.2081 top1= 93.2812
[E46B70 |  45440/50000 ( 91%) ] Loss: 0.1870 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4420 top1= 63.1811


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4700 top1= 42.0373


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.7940 top1= 45.6430

Train epoch 47
[E47B0  |    640/50000 (  1%) ] Loss: 0.2008 top1= 92.9688
[E47B10 |   7040/50000 ( 14%) ] Loss: 0.1998 top1= 93.1250
[E47B20 |  13440/50000 ( 27%) ] Loss: 0.2166 top1= 92.5000
[E47B30 |  19840/50000 ( 40%) ] Loss: 0.1791 top1= 93.5938
[E47B40 |  26240/50000 ( 52%) ] Loss: 0.1880 top1= 93.9062
[E47B50 |  32640/50000 ( 65%) ] Loss: 0.2005 top1= 92.8125
[E47B60 |  39040/50000 ( 78%) ] Loss: 0.1879 top1= 94.0625
[E47B70 |  45440/50000 ( 91%) ] Loss: 0.1637 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.9495 top1= 61.0276


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.7348 top1= 42.0373


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.0227 top1= 45.3826

Train epoch 48
[E48B0  |    640/50000 (  1%) ] Loss: 0.2550 top1= 90.9375
[E48B10 |   7040/50000 ( 14%) ] Loss: 0.1946 top1= 92.8125
[E48B20 |  13440/50000 ( 27%) ] Loss: 0.1547 top1= 94.2188
[E48B30 |  19840/50000 ( 40%) ] Loss: 0.2200 top1= 92.3438
[E48B40 |  26240/50000 ( 52%) ] Loss: 0.2047 top1= 93.1250
[E48B50 |  32640/50000 ( 65%) ] Loss: 0.2352 top1= 92.6562
[E48B60 |  39040/50000 ( 78%) ] Loss: 0.2175 top1= 92.6562
[E48B70 |  45440/50000 ( 91%) ] Loss: 0.1746 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4563 top1= 65.8854


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3765 top1= 42.7083


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.9914 top1= 44.6314

Train epoch 49
[E49B0  |    640/50000 (  1%) ] Loss: 0.2077 top1= 93.2812
[E49B10 |   7040/50000 ( 14%) ] Loss: 0.2079 top1= 91.5625
[E49B20 |  13440/50000 ( 27%) ] Loss: 0.1531 top1= 94.8438
[E49B30 |  19840/50000 ( 40%) ] Loss: 0.1982 top1= 92.0312
[E49B40 |  26240/50000 ( 52%) ] Loss: 0.1604 top1= 93.9062
[E49B50 |  32640/50000 ( 65%) ] Loss: 0.1549 top1= 94.6875
[E49B60 |  39040/50000 ( 78%) ] Loss: 0.1946 top1= 92.8125
[E49B70 |  45440/50000 ( 91%) ] Loss: 0.1375 top1= 94.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3003 top1= 68.5296


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6324 top1= 43.6799


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.0830 top1= 45.0621

Train epoch 50
[E50B0  |    640/50000 (  1%) ] Loss: 0.2083 top1= 93.4375
[E50B10 |   7040/50000 ( 14%) ] Loss: 0.2278 top1= 91.2500
[E50B20 |  13440/50000 ( 27%) ] Loss: 0.1827 top1= 92.6562
[E50B30 |  19840/50000 ( 40%) ] Loss: 0.1687 top1= 94.5312
[E50B40 |  26240/50000 ( 52%) ] Loss: 0.1605 top1= 94.5312
[E50B50 |  32640/50000 ( 65%) ] Loss: 0.1797 top1= 94.0625
[E50B60 |  39040/50000 ( 78%) ] Loss: 0.2015 top1= 92.5000
[E50B70 |  45440/50000 ( 91%) ] Loss: 0.1840 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3492 top1= 66.3462


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3246 top1= 41.7268


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.6401 top1= 45.3425

Train epoch 51
[E51B0  |    640/50000 (  1%) ] Loss: 0.2342 top1= 92.9688
[E51B10 |   7040/50000 ( 14%) ] Loss: 0.2168 top1= 92.1875
[E51B20 |  13440/50000 ( 27%) ] Loss: 0.1725 top1= 94.3750
[E51B30 |  19840/50000 ( 40%) ] Loss: 0.1578 top1= 93.4375
[E51B40 |  26240/50000 ( 52%) ] Loss: 0.1615 top1= 93.7500
[E51B50 |  32640/50000 ( 65%) ] Loss: 0.1650 top1= 93.7500
[E51B60 |  39040/50000 ( 78%) ] Loss: 0.1208 top1= 95.3125
[E51B70 |  45440/50000 ( 91%) ] Loss: 0.1686 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6123 top1= 65.1142


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4890 top1= 42.1875


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.0848 top1= 45.5929

Train epoch 52
[E52B0  |    640/50000 (  1%) ] Loss: 0.2076 top1= 90.9375
[E52B10 |   7040/50000 ( 14%) ] Loss: 0.2044 top1= 93.7500
[E52B20 |  13440/50000 ( 27%) ] Loss: 0.1654 top1= 95.1562
[E52B30 |  19840/50000 ( 40%) ] Loss: 0.1505 top1= 95.0000
[E52B40 |  26240/50000 ( 52%) ] Loss: 0.1920 top1= 94.0625
[E52B50 |  32640/50000 ( 65%) ] Loss: 0.1926 top1= 94.2188
[E52B60 |  39040/50000 ( 78%) ] Loss: 0.1927 top1= 92.1875
[E52B70 |  45440/50000 ( 91%) ] Loss: 0.1667 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0001 top1= 61.4083


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5354 top1= 42.7584


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.0491 top1= 44.9920

Train epoch 53
[E53B0  |    640/50000 (  1%) ] Loss: 0.1930 top1= 92.9688
[E53B10 |   7040/50000 ( 14%) ] Loss: 0.1808 top1= 94.5312
[E53B20 |  13440/50000 ( 27%) ] Loss: 0.1873 top1= 92.9688
[E53B30 |  19840/50000 ( 40%) ] Loss: 0.1720 top1= 93.4375
[E53B40 |  26240/50000 ( 52%) ] Loss: 0.1642 top1= 95.4688
[E53B50 |  32640/50000 ( 65%) ] Loss: 0.1917 top1= 92.3438
[E53B60 |  39040/50000 ( 78%) ] Loss: 0.1649 top1= 94.2188
[E53B70 |  45440/50000 ( 91%) ] Loss: 0.1633 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2694 top1= 62.0393


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6715 top1= 41.9772


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.8038 top1= 45.1623

Train epoch 54
[E54B0  |    640/50000 (  1%) ] Loss: 0.2334 top1= 92.6562
[E54B10 |   7040/50000 ( 14%) ] Loss: 0.2339 top1= 92.0312
[E54B20 |  13440/50000 ( 27%) ] Loss: 0.1459 top1= 94.6875
[E54B30 |  19840/50000 ( 40%) ] Loss: 0.1921 top1= 92.6562
[E54B40 |  26240/50000 ( 52%) ] Loss: 0.1852 top1= 92.5000
[E54B50 |  32640/50000 ( 65%) ] Loss: 0.1850 top1= 92.9688
[E54B60 |  39040/50000 ( 78%) ] Loss: 0.1580 top1= 94.6875
[E54B70 |  45440/50000 ( 91%) ] Loss: 0.1635 top1= 94.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2251 top1= 69.9419


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5389 top1= 42.4279


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0919 top1= 46.2340

Train epoch 55
[E55B0  |    640/50000 (  1%) ] Loss: 0.1595 top1= 93.9062
[E55B10 |   7040/50000 ( 14%) ] Loss: 0.1865 top1= 93.2812
[E55B20 |  13440/50000 ( 27%) ] Loss: 0.1460 top1= 95.0000
[E55B30 |  19840/50000 ( 40%) ] Loss: 0.1991 top1= 93.2812
[E55B40 |  26240/50000 ( 52%) ] Loss: 0.1607 top1= 94.8438
[E55B50 |  32640/50000 ( 65%) ] Loss: 0.2094 top1= 91.4062
[E55B60 |  39040/50000 ( 78%) ] Loss: 0.1573 top1= 95.0000
[E55B70 |  45440/50000 ( 91%) ] Loss: 0.1496 top1= 94.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5856 top1= 67.7284


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0980 top1= 42.1474


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7525 top1= 45.6030

Train epoch 56
[E56B0  |    640/50000 (  1%) ] Loss: 0.1557 top1= 94.6875
[E56B10 |   7040/50000 ( 14%) ] Loss: 0.1755 top1= 92.5000
[E56B20 |  13440/50000 ( 27%) ] Loss: 0.1656 top1= 93.9062
[E56B30 |  19840/50000 ( 40%) ] Loss: 0.1309 top1= 95.6250
[E56B40 |  26240/50000 ( 52%) ] Loss: 0.1922 top1= 92.8125
[E56B50 |  32640/50000 ( 65%) ] Loss: 0.1564 top1= 94.3750
[E56B60 |  39040/50000 ( 78%) ] Loss: 0.1774 top1= 93.4375
[E56B70 |  45440/50000 ( 91%) ] Loss: 0.2138 top1= 92.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8782 top1= 64.1226


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7204 top1= 41.3962


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.6010 top1= 44.9720

Train epoch 57
[E57B0  |    640/50000 (  1%) ] Loss: 0.2630 top1= 91.2500
[E57B10 |   7040/50000 ( 14%) ] Loss: 0.1568 top1= 94.5312
[E57B20 |  13440/50000 ( 27%) ] Loss: 0.1315 top1= 95.0000
[E57B30 |  19840/50000 ( 40%) ] Loss: 0.1666 top1= 94.5312
[E57B40 |  26240/50000 ( 52%) ] Loss: 0.1680 top1= 93.9062
[E57B50 |  32640/50000 ( 65%) ] Loss: 0.1741 top1= 93.1250
[E57B60 |  39040/50000 ( 78%) ] Loss: 0.1736 top1= 93.5938
[E57B70 |  45440/50000 ( 91%) ] Loss: 0.1358 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4167 top1= 66.2059


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0251 top1= 42.7284


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.7406 top1= 45.3826

Train epoch 58
[E58B0  |    640/50000 (  1%) ] Loss: 0.1626 top1= 95.3125
[E58B10 |   7040/50000 ( 14%) ] Loss: 0.1888 top1= 94.0625
[E58B20 |  13440/50000 ( 27%) ] Loss: 0.1776 top1= 93.1250
[E58B30 |  19840/50000 ( 40%) ] Loss: 0.1564 top1= 94.0625
[E58B40 |  26240/50000 ( 52%) ] Loss: 0.1283 top1= 95.0000
[E58B50 |  32640/50000 ( 65%) ] Loss: 0.1715 top1= 94.6875
[E58B60 |  39040/50000 ( 78%) ] Loss: 0.1235 top1= 95.9375
[E58B70 |  45440/50000 ( 91%) ] Loss: 0.1458 top1= 94.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3070 top1= 68.6498


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.0882 top1= 43.3694


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4858 top1= 44.8017

Train epoch 59
[E59B0  |    640/50000 (  1%) ] Loss: 0.1640 top1= 93.1250
[E59B10 |   7040/50000 ( 14%) ] Loss: 0.1905 top1= 93.7500
[E59B20 |  13440/50000 ( 27%) ] Loss: 0.1501 top1= 93.5938
[E59B30 |  19840/50000 ( 40%) ] Loss: 0.1442 top1= 94.8438
[E59B40 |  26240/50000 ( 52%) ] Loss: 0.1901 top1= 93.4375
[E59B50 |  32640/50000 ( 65%) ] Loss: 0.1683 top1= 94.0625
[E59B60 |  39040/50000 ( 78%) ] Loss: 0.1767 top1= 94.2188
[E59B70 |  45440/50000 ( 91%) ] Loss: 0.1062 top1= 95.7812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4568 top1= 68.0489


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2164 top1= 42.7384


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.5057 top1= 45.4627

Train epoch 60
[E60B0  |    640/50000 (  1%) ] Loss: 0.1486 top1= 94.3750
[E60B10 |   7040/50000 ( 14%) ] Loss: 0.1627 top1= 94.8438
[E60B20 |  13440/50000 ( 27%) ] Loss: 0.1466 top1= 95.1562
[E60B30 |  19840/50000 ( 40%) ] Loss: 0.1617 top1= 94.0625
[E60B40 |  26240/50000 ( 52%) ] Loss: 0.1097 top1= 96.0938
[E60B50 |  32640/50000 ( 65%) ] Loss: 0.1803 top1= 93.5938
[E60B60 |  39040/50000 ( 78%) ] Loss: 0.1985 top1= 93.1250
[E60B70 |  45440/50000 ( 91%) ] Loss: 0.1739 top1= 94.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2844 top1= 67.4279


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7496 top1= 42.1374


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9407 top1= 44.7015

Train epoch 61
[E61B0  |    640/50000 (  1%) ] Loss: 0.2123 top1= 92.0312
[E61B10 |   7040/50000 ( 14%) ] Loss: 0.1614 top1= 94.6875
[E61B20 |  13440/50000 ( 27%) ] Loss: 0.0944 top1= 96.8750
[E61B30 |  19840/50000 ( 40%) ] Loss: 0.1398 top1= 94.6875
[E61B40 |  26240/50000 ( 52%) ] Loss: 0.1731 top1= 95.3125
[E61B50 |  32640/50000 ( 65%) ] Loss: 0.1720 top1= 95.1562
[E61B60 |  39040/50000 ( 78%) ] Loss: 0.1209 top1= 95.6250
[E61B70 |  45440/50000 ( 91%) ] Loss: 0.1454 top1= 95.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6222 top1= 63.3313


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6385 top1= 43.4595


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.2292 top1= 45.4026

Train epoch 62
[E62B0  |    640/50000 (  1%) ] Loss: 0.1274 top1= 96.0938
[E62B10 |   7040/50000 ( 14%) ] Loss: 0.1886 top1= 93.5938
[E62B20 |  13440/50000 ( 27%) ] Loss: 0.1475 top1= 95.1562
[E62B30 |  19840/50000 ( 40%) ] Loss: 0.1788 top1= 95.3125
[E62B40 |  26240/50000 ( 52%) ] Loss: 0.1432 top1= 94.8438
[E62B50 |  32640/50000 ( 65%) ] Loss: 0.1624 top1= 93.5938
[E62B60 |  39040/50000 ( 78%) ] Loss: 0.1387 top1= 95.1562
[E62B70 |  45440/50000 ( 91%) ] Loss: 0.1274 top1= 96.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6261 top1= 66.2260


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.9416 top1= 42.5781


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.0281 top1= 44.9119

Train epoch 63
[E63B0  |    640/50000 (  1%) ] Loss: 0.2056 top1= 92.5000
[E63B10 |   7040/50000 ( 14%) ] Loss: 0.1761 top1= 93.7500
[E63B20 |  13440/50000 ( 27%) ] Loss: 0.1405 top1= 95.4688
[E63B30 |  19840/50000 ( 40%) ] Loss: 0.1379 top1= 94.6875
[E63B40 |  26240/50000 ( 52%) ] Loss: 0.1947 top1= 92.3438
[E63B50 |  32640/50000 ( 65%) ] Loss: 0.1213 top1= 95.7812
[E63B60 |  39040/50000 ( 78%) ] Loss: 0.1196 top1= 96.7188
[E63B70 |  45440/50000 ( 91%) ] Loss: 0.1368 top1= 95.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4057 top1= 68.9804


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2460 top1= 42.5881


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.2760 top1= 45.6931

Train epoch 64
[E64B0  |    640/50000 (  1%) ] Loss: 0.1706 top1= 93.7500
[E64B10 |   7040/50000 ( 14%) ] Loss: 0.1333 top1= 95.0000
[E64B20 |  13440/50000 ( 27%) ] Loss: 0.1165 top1= 95.0000
[E64B30 |  19840/50000 ( 40%) ] Loss: 0.1352 top1= 95.3125
[E64B40 |  26240/50000 ( 52%) ] Loss: 0.1449 top1= 95.1562
[E64B50 |  32640/50000 ( 65%) ] Loss: 0.1653 top1= 95.1562
[E64B60 |  39040/50000 ( 78%) ] Loss: 0.1443 top1= 94.6875
[E64B70 |  45440/50000 ( 91%) ] Loss: 0.1379 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6446 top1= 66.6166


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4526 top1= 42.7083


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.9008 top1= 45.5529

Train epoch 65
[E65B0  |    640/50000 (  1%) ] Loss: 0.2118 top1= 93.4375
[E65B10 |   7040/50000 ( 14%) ] Loss: 0.1323 top1= 95.1562
[E65B20 |  13440/50000 ( 27%) ] Loss: 0.2124 top1= 93.7500
[E65B30 |  19840/50000 ( 40%) ] Loss: 0.1374 top1= 95.3125
[E65B40 |  26240/50000 ( 52%) ] Loss: 0.1701 top1= 94.2188
[E65B50 |  32640/50000 ( 65%) ] Loss: 0.1729 top1= 94.6875
[E65B60 |  39040/50000 ( 78%) ] Loss: 0.1437 top1= 95.3125
[E65B70 |  45440/50000 ( 91%) ] Loss: 0.1350 top1= 96.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3994 top1= 67.8586


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.9766 top1= 42.9587


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3919 top1= 45.4627

Train epoch 66
[E66B0  |    640/50000 (  1%) ] Loss: 0.1760 top1= 94.6875
[E66B10 |   7040/50000 ( 14%) ] Loss: 0.1460 top1= 95.0000
[E66B20 |  13440/50000 ( 27%) ] Loss: 0.1624 top1= 95.4688
[E66B30 |  19840/50000 ( 40%) ] Loss: 0.1791 top1= 94.2188
[E66B40 |  26240/50000 ( 52%) ] Loss: 0.1282 top1= 95.1562
[E66B50 |  32640/50000 ( 65%) ] Loss: 0.1487 top1= 95.6250
[E66B60 |  39040/50000 ( 78%) ] Loss: 0.1155 top1= 96.2500
[E66B70 |  45440/50000 ( 91%) ] Loss: 0.1852 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4267 top1= 67.8285


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0578 top1= 42.9187


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.7042 top1= 45.4527

Train epoch 67
[E67B0  |    640/50000 (  1%) ] Loss: 0.1768 top1= 93.7500
[E67B10 |   7040/50000 ( 14%) ] Loss: 0.2282 top1= 93.5938
[E67B20 |  13440/50000 ( 27%) ] Loss: 0.1256 top1= 95.4688
[E67B30 |  19840/50000 ( 40%) ] Loss: 0.1233 top1= 95.6250
[E67B40 |  26240/50000 ( 52%) ] Loss: 0.1183 top1= 95.7812
[E67B50 |  32640/50000 ( 65%) ] Loss: 0.1343 top1= 95.6250
[E67B60 |  39040/50000 ( 78%) ] Loss: 0.1141 top1= 96.7188
[E67B70 |  45440/50000 ( 91%) ] Loss: 0.0938 top1= 97.1875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4840 top1= 67.7985


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1413 top1= 43.2492


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.6708 top1= 45.5529

Train epoch 68
[E68B0  |    640/50000 (  1%) ] Loss: 0.1508 top1= 95.1562
[E68B10 |   7040/50000 ( 14%) ] Loss: 0.1568 top1= 95.0000
[E68B20 |  13440/50000 ( 27%) ] Loss: 0.1409 top1= 94.8438
[E68B30 |  19840/50000 ( 40%) ] Loss: 0.2239 top1= 92.6562
[E68B40 |  26240/50000 ( 52%) ] Loss: 0.1419 top1= 95.1562
[E68B50 |  32640/50000 ( 65%) ] Loss: 0.2128 top1= 94.2188
[E68B60 |  39040/50000 ( 78%) ] Loss: 0.1408 top1= 95.4688
[E68B70 |  45440/50000 ( 91%) ] Loss: 0.1105 top1= 96.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7927 top1= 67.2676


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.8300 top1= 41.6066


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.4080 top1= 45.5329

Train epoch 69
[E69B0  |    640/50000 (  1%) ] Loss: 0.1861 top1= 93.9062
[E69B10 |   7040/50000 ( 14%) ] Loss: 0.1760 top1= 94.2188
[E69B20 |  13440/50000 ( 27%) ] Loss: 0.1742 top1= 94.2188
[E69B30 |  19840/50000 ( 40%) ] Loss: 0.2023 top1= 92.3438
[E69B40 |  26240/50000 ( 52%) ] Loss: 0.1614 top1= 94.6875
[E69B50 |  32640/50000 ( 65%) ] Loss: 0.1313 top1= 94.6875
[E69B60 |  39040/50000 ( 78%) ] Loss: 0.1560 top1= 95.1562
[E69B70 |  45440/50000 ( 91%) ] Loss: 0.1363 top1= 94.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6517 top1= 67.9988


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.2828 top1= 42.6282


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.2204 top1= 46.8950

Train epoch 70
[E70B0  |    640/50000 (  1%) ] Loss: 0.1650 top1= 94.8438
[E70B10 |   7040/50000 ( 14%) ] Loss: 0.1543 top1= 94.8438
[E70B20 |  13440/50000 ( 27%) ] Loss: 0.1709 top1= 93.5938
[E70B30 |  19840/50000 ( 40%) ] Loss: 0.1929 top1= 94.2188
[E70B40 |  26240/50000 ( 52%) ] Loss: 0.1874 top1= 93.2812
[E70B50 |  32640/50000 ( 65%) ] Loss: 0.1224 top1= 95.7812
[E70B60 |  39040/50000 ( 78%) ] Loss: 0.0910 top1= 96.7188
[E70B70 |  45440/50000 ( 91%) ] Loss: 0.1525 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3877 top1= 68.5397


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.9776 top1= 43.4495


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.5580 top1= 45.5529

Train epoch 71
[E71B0  |    640/50000 (  1%) ] Loss: 0.1779 top1= 94.3750
[E71B10 |   7040/50000 ( 14%) ] Loss: 0.1529 top1= 95.1562
[E71B20 |  13440/50000 ( 27%) ] Loss: 0.1801 top1= 93.7500
[E71B30 |  19840/50000 ( 40%) ] Loss: 0.1297 top1= 95.1562
[E71B40 |  26240/50000 ( 52%) ] Loss: 0.1056 top1= 96.0938
[E71B50 |  32640/50000 ( 65%) ] Loss: 0.1628 top1= 94.5312
[E71B60 |  39040/50000 ( 78%) ] Loss: 0.1400 top1= 95.7812
[E71B70 |  45440/50000 ( 91%) ] Loss: 0.1225 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.6752 top1= 62.7404


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.8723 top1= 43.1190


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.5453 top1= 45.4026

Train epoch 72
[E72B0  |    640/50000 (  1%) ] Loss: 0.1531 top1= 94.5312
[E72B10 |   7040/50000 ( 14%) ] Loss: 0.1499 top1= 94.5312
[E72B20 |  13440/50000 ( 27%) ] Loss: 0.0916 top1= 96.8750
[E72B30 |  19840/50000 ( 40%) ] Loss: 0.1970 top1= 93.4375
[E72B40 |  26240/50000 ( 52%) ] Loss: 0.1100 top1= 96.0938
[E72B50 |  32640/50000 ( 65%) ] Loss: 0.2009 top1= 93.4375
[E72B60 |  39040/50000 ( 78%) ] Loss: 0.1547 top1= 94.8438
[E72B70 |  45440/50000 ( 91%) ] Loss: 0.1578 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7604 top1= 66.8870


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4869 top1= 43.3193


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.2098 top1= 45.4327

Train epoch 73
[E73B0  |    640/50000 (  1%) ] Loss: 0.2532 top1= 92.3438
[E73B10 |   7040/50000 ( 14%) ] Loss: 0.1891 top1= 94.5312
[E73B20 |  13440/50000 ( 27%) ] Loss: 0.1763 top1= 94.8438
[E73B30 |  19840/50000 ( 40%) ] Loss: 0.1702 top1= 94.3750
[E73B40 |  26240/50000 ( 52%) ] Loss: 0.1520 top1= 95.1562
[E73B50 |  32640/50000 ( 65%) ] Loss: 0.1811 top1= 94.5312
[E73B60 |  39040/50000 ( 78%) ] Loss: 0.1855 top1= 94.2188
[E73B70 |  45440/50000 ( 91%) ] Loss: 0.1408 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4710 top1= 68.0389


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4089 top1= 42.5381


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4829 top1= 45.7131

Train epoch 74
[E74B0  |    640/50000 (  1%) ] Loss: 0.1753 top1= 94.2188
[E74B10 |   7040/50000 ( 14%) ] Loss: 0.1813 top1= 93.4375
[E74B20 |  13440/50000 ( 27%) ] Loss: 0.1720 top1= 94.8438
[E74B30 |  19840/50000 ( 40%) ] Loss: 0.1989 top1= 93.9062
[E74B40 |  26240/50000 ( 52%) ] Loss: 0.1755 top1= 93.5938
[E74B50 |  32640/50000 ( 65%) ] Loss: 0.1994 top1= 94.5312
[E74B60 |  39040/50000 ( 78%) ] Loss: 0.1462 top1= 95.7812
[E74B70 |  45440/50000 ( 91%) ] Loss: 0.1763 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4141 top1= 68.9303


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4656 top1= 42.7384


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=13.0635 top1= 45.6931

Train epoch 75
[E75B0  |    640/50000 (  1%) ] Loss: 0.1438 top1= 94.0625
[E75B10 |   7040/50000 ( 14%) ] Loss: 0.1324 top1= 97.0312
[E75B20 |  13440/50000 ( 27%) ] Loss: 0.1246 top1= 95.4688
[E75B30 |  19840/50000 ( 40%) ] Loss: 0.1656 top1= 95.1562
[E75B40 |  26240/50000 ( 52%) ] Loss: 0.1275 top1= 94.6875
[E75B50 |  32640/50000 ( 65%) ] Loss: 0.1529 top1= 95.4688
[E75B60 |  39040/50000 ( 78%) ] Loss: 0.1677 top1= 94.6875
[E75B70 |  45440/50000 ( 91%) ] Loss: 0.2942 top1= 95.7812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2292 top1= 67.7083


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.1733 top1= 43.2192


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.9594 top1= 46.1338

Train epoch 76
[E76B0  |    640/50000 (  1%) ] Loss: 0.2088 top1= 93.2812
[E76B10 |   7040/50000 ( 14%) ] Loss: 0.1447 top1= 96.0938
[E76B20 |  13440/50000 ( 27%) ] Loss: 0.1295 top1= 96.2500
[E76B30 |  19840/50000 ( 40%) ] Loss: 0.1613 top1= 93.1250
[E76B40 |  26240/50000 ( 52%) ] Loss: 0.1808 top1= 95.1562
[E76B50 |  32640/50000 ( 65%) ] Loss: 0.1534 top1= 94.2188
[E76B60 |  39040/50000 ( 78%) ] Loss: 0.1470 top1= 95.4688
[E76B70 |  45440/50000 ( 91%) ] Loss: 0.1693 top1= 94.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3672 top1= 68.2592


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2598 top1= 43.1290


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.4199 top1= 45.8834

Train epoch 77
[E77B0  |    640/50000 (  1%) ] Loss: 0.1662 top1= 93.7500
[E77B10 |   7040/50000 ( 14%) ] Loss: 0.1322 top1= 96.4062
[E77B20 |  13440/50000 ( 27%) ] Loss: 0.0945 top1= 96.7188
[E77B30 |  19840/50000 ( 40%) ] Loss: 0.1563 top1= 93.9062
[E77B40 |  26240/50000 ( 52%) ] Loss: 0.1755 top1= 95.3125
[E77B50 |  32640/50000 ( 65%) ] Loss: 0.1336 top1= 95.0000
[E77B60 |  39040/50000 ( 78%) ] Loss: 0.1203 top1= 95.7812
[E77B70 |  45440/50000 ( 91%) ] Loss: 0.1201 top1= 95.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5190 top1= 67.8986


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0302 top1= 43.3894


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.0611 top1= 45.7632

Train epoch 78
[E78B0  |    640/50000 (  1%) ] Loss: 0.1395 top1= 94.5312
[E78B10 |   7040/50000 ( 14%) ] Loss: 0.1328 top1= 95.0000
[E78B20 |  13440/50000 ( 27%) ] Loss: 0.1371 top1= 95.6250
[E78B30 |  19840/50000 ( 40%) ] Loss: 0.1212 top1= 96.4062
[E78B40 |  26240/50000 ( 52%) ] Loss: 0.1547 top1= 94.3750
[E78B50 |  32640/50000 ( 65%) ] Loss: 0.1392 top1= 94.8438
[E78B60 |  39040/50000 ( 78%) ] Loss: 0.1366 top1= 95.6250
[E78B70 |  45440/50000 ( 91%) ] Loss: 0.0913 top1= 97.1875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4934 top1= 68.0489


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4054 top1= 43.6899


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4959 top1= 47.4058

Train epoch 79
[E79B0  |    640/50000 (  1%) ] Loss: 0.1568 top1= 95.6250
[E79B10 |   7040/50000 ( 14%) ] Loss: 0.1878 top1= 94.0625
[E79B20 |  13440/50000 ( 27%) ] Loss: 0.1677 top1= 94.6875
[E79B30 |  19840/50000 ( 40%) ] Loss: 0.1625 top1= 94.5312
[E79B40 |  26240/50000 ( 52%) ] Loss: 0.1401 top1= 95.1562
[E79B50 |  32640/50000 ( 65%) ] Loss: 0.1372 top1= 95.3125
[E79B60 |  39040/50000 ( 78%) ] Loss: 0.1949 top1= 94.2188
[E79B70 |  45440/50000 ( 91%) ] Loss: 0.1617 top1= 95.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4888 top1= 66.9371


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.9314 top1= 43.0188


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7619 top1= 46.9752

Train epoch 80
[E80B0  |    640/50000 (  1%) ] Loss: 0.1602 top1= 93.9062
[E80B10 |   7040/50000 ( 14%) ] Loss: 0.1304 top1= 95.9375
[E80B20 |  13440/50000 ( 27%) ] Loss: 0.1511 top1= 95.6250
[E80B30 |  19840/50000 ( 40%) ] Loss: 0.1370 top1= 95.3125
[E80B40 |  26240/50000 ( 52%) ] Loss: 0.1654 top1= 95.1562
[E80B50 |  32640/50000 ( 65%) ] Loss: 0.1407 top1= 94.2188
[E80B60 |  39040/50000 ( 78%) ] Loss: 0.1954 top1= 94.3750
[E80B70 |  45440/50000 ( 91%) ] Loss: 0.1195 top1= 96.8750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4462 top1= 67.7985


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8980 top1= 43.9303


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1959 top1= 45.5729

Train epoch 81
[E81B0  |    640/50000 (  1%) ] Loss: 0.1582 top1= 94.8438
[E81B10 |   7040/50000 ( 14%) ] Loss: 0.1431 top1= 95.9375
[E81B20 |  13440/50000 ( 27%) ] Loss: 0.1057 top1= 96.7188
[E81B30 |  19840/50000 ( 40%) ] Loss: 0.1240 top1= 96.4062
[E81B40 |  26240/50000 ( 52%) ] Loss: 0.1089 top1= 97.1875
[E81B50 |  32640/50000 ( 65%) ] Loss: 0.0765 top1= 97.8125
[E81B60 |  39040/50000 ( 78%) ] Loss: 0.1001 top1= 96.8750
[E81B70 |  45440/50000 ( 91%) ] Loss: 0.0623 top1= 97.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1435 top1= 73.5777


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1828 top1= 48.8982


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8158 top1= 50.5609

Train epoch 82
[E82B0  |    640/50000 (  1%) ] Loss: 0.0685 top1= 97.6562
[E82B10 |   7040/50000 ( 14%) ] Loss: 0.0963 top1= 96.7188
[E82B20 |  13440/50000 ( 27%) ] Loss: 0.0814 top1= 97.8125
[E82B30 |  19840/50000 ( 40%) ] Loss: 0.0851 top1= 97.5000
[E82B40 |  26240/50000 ( 52%) ] Loss: 0.0548 top1= 98.5938
[E82B50 |  32640/50000 ( 65%) ] Loss: 0.0814 top1= 97.0312
[E82B60 |  39040/50000 ( 78%) ] Loss: 0.0513 top1= 98.7500
[E82B70 |  45440/50000 ( 91%) ] Loss: 0.0318 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1280 top1= 73.9784


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.0214 top1= 48.8982


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8924 top1= 51.5825

Train epoch 83
[E83B0  |    640/50000 (  1%) ] Loss: 0.0807 top1= 97.3438
[E83B10 |   7040/50000 ( 14%) ] Loss: 0.0629 top1= 98.1250
[E83B20 |  13440/50000 ( 27%) ] Loss: 0.0563 top1= 97.9688
[E83B30 |  19840/50000 ( 40%) ] Loss: 0.0707 top1= 98.4375
[E83B40 |  26240/50000 ( 52%) ] Loss: 0.0323 top1= 99.0625
[E83B50 |  32640/50000 ( 65%) ] Loss: 0.0662 top1= 97.6562
[E83B60 |  39040/50000 ( 78%) ] Loss: 0.0690 top1= 98.1250
[E83B70 |  45440/50000 ( 91%) ] Loss: 0.0420 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1273 top1= 74.0986


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3133 top1= 48.3774


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0939 top1= 50.8814

Train epoch 84
[E84B0  |    640/50000 (  1%) ] Loss: 0.0598 top1= 97.9688
[E84B10 |   7040/50000 ( 14%) ] Loss: 0.0503 top1= 98.4375
[E84B20 |  13440/50000 ( 27%) ] Loss: 0.0501 top1= 98.4375
[E84B30 |  19840/50000 ( 40%) ] Loss: 0.0499 top1= 98.2812
[E84B40 |  26240/50000 ( 52%) ] Loss: 0.0654 top1= 98.4375
[E84B50 |  32640/50000 ( 65%) ] Loss: 0.0549 top1= 97.8125
[E84B60 |  39040/50000 ( 78%) ] Loss: 0.0481 top1= 98.4375
[E84B70 |  45440/50000 ( 91%) ] Loss: 0.0374 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2346 top1= 73.8882


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5404 top1= 48.5477


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6055 top1= 52.8245

Train epoch 85
[E85B0  |    640/50000 (  1%) ] Loss: 0.0665 top1= 98.4375
[E85B10 |   7040/50000 ( 14%) ] Loss: 0.0518 top1= 97.8125
[E85B20 |  13440/50000 ( 27%) ] Loss: 0.0382 top1= 98.5938
[E85B30 |  19840/50000 ( 40%) ] Loss: 0.0483 top1= 97.8125
[E85B40 |  26240/50000 ( 52%) ] Loss: 0.0517 top1= 98.4375
[E85B50 |  32640/50000 ( 65%) ] Loss: 0.0607 top1= 97.6562
[E85B60 |  39040/50000 ( 78%) ] Loss: 0.0383 top1= 98.4375
[E85B70 |  45440/50000 ( 91%) ] Loss: 0.0290 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1787 top1= 73.9483


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4562 top1= 48.1571


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.9314 top1= 49.8998

Train epoch 86
[E86B0  |    640/50000 (  1%) ] Loss: 0.0403 top1= 98.7500
[E86B10 |   7040/50000 ( 14%) ] Loss: 0.0350 top1= 98.9062
[E86B20 |  13440/50000 ( 27%) ] Loss: 0.0372 top1= 98.7500
[E86B30 |  19840/50000 ( 40%) ] Loss: 0.0382 top1= 99.0625
[E86B40 |  26240/50000 ( 52%) ] Loss: 0.0306 top1= 99.0625
[E86B50 |  32640/50000 ( 65%) ] Loss: 0.0394 top1= 98.7500
[E86B60 |  39040/50000 ( 78%) ] Loss: 0.0515 top1= 97.9688
[E86B70 |  45440/50000 ( 91%) ] Loss: 0.0404 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2983 top1= 73.2372


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8028 top1= 47.8666


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.2124 top1= 55.9395

Train epoch 87
[E87B0  |    640/50000 (  1%) ] Loss: 0.0540 top1= 98.7500
[E87B10 |   7040/50000 ( 14%) ] Loss: 0.0307 top1= 99.2188
[E87B20 |  13440/50000 ( 27%) ] Loss: 0.0366 top1= 98.4375
[E87B30 |  19840/50000 ( 40%) ] Loss: 0.0331 top1= 98.5938
[E87B40 |  26240/50000 ( 52%) ] Loss: 0.0388 top1= 98.7500
[E87B50 |  32640/50000 ( 65%) ] Loss: 0.0527 top1= 98.4375
[E87B60 |  39040/50000 ( 78%) ] Loss: 0.0294 top1= 98.9062
[E87B70 |  45440/50000 ( 91%) ] Loss: 0.0483 top1= 97.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2861 top1= 74.2188


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4530 top1= 47.1755


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6555 top1= 51.7929

Train epoch 88
[E88B0  |    640/50000 (  1%) ] Loss: 0.0452 top1= 98.2812
[E88B10 |   7040/50000 ( 14%) ] Loss: 0.0744 top1= 97.6562
[E88B20 |  13440/50000 ( 27%) ] Loss: 0.0428 top1= 98.1250
[E88B30 |  19840/50000 ( 40%) ] Loss: 0.0480 top1= 98.5938
[E88B40 |  26240/50000 ( 52%) ] Loss: 0.0323 top1= 99.0625
[E88B50 |  32640/50000 ( 65%) ] Loss: 0.0338 top1= 98.4375
[E88B60 |  39040/50000 ( 78%) ] Loss: 0.0382 top1= 98.7500
[E88B70 |  45440/50000 ( 91%) ] Loss: 0.0391 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2989 top1= 74.1887


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3322 top1= 49.0184


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2290 top1= 50.9415

Train epoch 89
[E89B0  |    640/50000 (  1%) ] Loss: 0.0580 top1= 97.8125
[E89B10 |   7040/50000 ( 14%) ] Loss: 0.0292 top1= 99.2188
[E89B20 |  13440/50000 ( 27%) ] Loss: 0.0306 top1= 98.9062
[E89B30 |  19840/50000 ( 40%) ] Loss: 0.0398 top1= 98.5938
[E89B40 |  26240/50000 ( 52%) ] Loss: 0.0389 top1= 99.0625
[E89B50 |  32640/50000 ( 65%) ] Loss: 0.0661 top1= 98.5938
[E89B60 |  39040/50000 ( 78%) ] Loss: 0.0445 top1= 98.5938
[E89B70 |  45440/50000 ( 91%) ] Loss: 0.0283 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3457 top1= 72.9667


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2279 top1= 47.0954


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2442 top1= 53.2752

Train epoch 90
[E90B0  |    640/50000 (  1%) ] Loss: 0.0389 top1= 98.7500
[E90B10 |   7040/50000 ( 14%) ] Loss: 0.0488 top1= 98.5938
[E90B20 |  13440/50000 ( 27%) ] Loss: 0.0284 top1= 99.5312
[E90B30 |  19840/50000 ( 40%) ] Loss: 0.0591 top1= 98.2812
[E90B40 |  26240/50000 ( 52%) ] Loss: 0.0212 top1= 99.2188
[E90B50 |  32640/50000 ( 65%) ] Loss: 0.0318 top1= 98.9062
[E90B60 |  39040/50000 ( 78%) ] Loss: 0.0451 top1= 98.5938
[E90B70 |  45440/50000 ( 91%) ] Loss: 0.0520 top1= 98.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3497 top1= 74.0885


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8040 top1= 49.2288


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1190 top1= 50.1903

Train epoch 91
[E91B0  |    640/50000 (  1%) ] Loss: 0.0408 top1= 99.3750
[E91B10 |   7040/50000 ( 14%) ] Loss: 0.0465 top1= 98.5938
[E91B20 |  13440/50000 ( 27%) ] Loss: 0.0240 top1= 99.2188
[E91B30 |  19840/50000 ( 40%) ] Loss: 0.0196 top1= 99.5312
[E91B40 |  26240/50000 ( 52%) ] Loss: 0.0153 top1= 99.5312
[E91B50 |  32640/50000 ( 65%) ] Loss: 0.0608 top1= 98.5938
[E91B60 |  39040/50000 ( 78%) ] Loss: 0.0332 top1= 98.9062
[E91B70 |  45440/50000 ( 91%) ] Loss: 0.0154 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2789 top1= 73.9283


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0414 top1= 48.4675


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3966 top1= 51.3822

Train epoch 92
[E92B0  |    640/50000 (  1%) ] Loss: 0.0466 top1= 98.7500
[E92B10 |   7040/50000 ( 14%) ] Loss: 0.0293 top1= 98.4375
[E92B20 |  13440/50000 ( 27%) ] Loss: 0.0248 top1= 98.7500
[E92B30 |  19840/50000 ( 40%) ] Loss: 0.0274 top1= 99.3750
[E92B40 |  26240/50000 ( 52%) ] Loss: 0.0195 top1= 99.3750
[E92B50 |  32640/50000 ( 65%) ] Loss: 0.0181 top1= 99.5312
[E92B60 |  39040/50000 ( 78%) ] Loss: 0.0363 top1= 99.2188
[E92B70 |  45440/50000 ( 91%) ] Loss: 0.0393 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3866 top1= 73.7881


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6807 top1= 47.6462


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4795 top1= 53.3153

Train epoch 93
[E93B0  |    640/50000 (  1%) ] Loss: 0.0456 top1= 98.2812
[E93B10 |   7040/50000 ( 14%) ] Loss: 0.0403 top1= 98.2812
[E93B20 |  13440/50000 ( 27%) ] Loss: 0.0217 top1= 99.2188
[E93B30 |  19840/50000 ( 40%) ] Loss: 0.0413 top1= 98.5938
[E93B40 |  26240/50000 ( 52%) ] Loss: 0.0106 top1= 99.6875
[E93B50 |  32640/50000 ( 65%) ] Loss: 0.0393 top1= 98.9062
[E93B60 |  39040/50000 ( 78%) ] Loss: 0.0328 top1= 99.0625
[E93B70 |  45440/50000 ( 91%) ] Loss: 0.0278 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3421 top1= 74.3990


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2875 top1= 48.5577


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4719 top1= 53.9563

Train epoch 94
[E94B0  |    640/50000 (  1%) ] Loss: 0.0322 top1= 98.5938
[E94B10 |   7040/50000 ( 14%) ] Loss: 0.0366 top1= 98.7500
[E94B20 |  13440/50000 ( 27%) ] Loss: 0.0213 top1= 99.3750
[E94B30 |  19840/50000 ( 40%) ] Loss: 0.0299 top1= 98.9062
[E94B40 |  26240/50000 ( 52%) ] Loss: 0.0226 top1= 98.9062
[E94B50 |  32640/50000 ( 65%) ] Loss: 0.0274 top1= 99.2188
[E94B60 |  39040/50000 ( 78%) ] Loss: 0.0277 top1= 99.2188
[E94B70 |  45440/50000 ( 91%) ] Loss: 0.0268 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4253 top1= 73.7981


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9049 top1= 47.2356


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7434 top1= 51.2620

Train epoch 95
[E95B0  |    640/50000 (  1%) ] Loss: 0.0317 top1= 98.9062
[E95B10 |   7040/50000 ( 14%) ] Loss: 0.0394 top1= 98.1250
[E95B20 |  13440/50000 ( 27%) ] Loss: 0.0300 top1= 99.0625
[E95B30 |  19840/50000 ( 40%) ] Loss: 0.0204 top1= 99.3750
[E95B40 |  26240/50000 ( 52%) ] Loss: 0.0212 top1= 99.3750
[E95B50 |  32640/50000 ( 65%) ] Loss: 0.0297 top1= 99.0625
[E95B60 |  39040/50000 ( 78%) ] Loss: 0.0299 top1= 99.2188
[E95B70 |  45440/50000 ( 91%) ] Loss: 0.0348 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3781 top1= 73.6979


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2931 top1= 49.6895


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4978 top1= 51.9431

Train epoch 96
[E96B0  |    640/50000 (  1%) ] Loss: 0.0285 top1= 98.7500
[E96B10 |   7040/50000 ( 14%) ] Loss: 0.0317 top1= 98.7500
[E96B20 |  13440/50000 ( 27%) ] Loss: 0.0254 top1= 98.9062
[E96B30 |  19840/50000 ( 40%) ] Loss: 0.0186 top1= 99.0625
[E96B40 |  26240/50000 ( 52%) ] Loss: 0.0340 top1= 99.3750
[E96B50 |  32640/50000 ( 65%) ] Loss: 0.0261 top1= 98.9062
[E96B60 |  39040/50000 ( 78%) ] Loss: 0.0267 top1= 98.9062
[E96B70 |  45440/50000 ( 91%) ] Loss: 0.0130 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4601 top1= 74.0284


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0411 top1= 51.1919


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1362 top1= 54.9079

Train epoch 97
[E97B0  |    640/50000 (  1%) ] Loss: 0.0267 top1= 98.9062
[E97B10 |   7040/50000 ( 14%) ] Loss: 0.0174 top1= 99.3750
[E97B20 |  13440/50000 ( 27%) ] Loss: 0.0108 top1= 99.6875
[E97B30 |  19840/50000 ( 40%) ] Loss: 0.0109 top1= 99.6875
[E97B40 |  26240/50000 ( 52%) ] Loss: 0.0157 top1= 99.5312
[E97B50 |  32640/50000 ( 65%) ] Loss: 0.0228 top1= 99.2188
[E97B60 |  39040/50000 ( 78%) ] Loss: 0.0247 top1= 99.0625
[E97B70 |  45440/50000 ( 91%) ] Loss: 0.0150 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4237 top1= 74.7196


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4999 top1= 48.8982


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3559 top1= 52.5240

Train epoch 98
[E98B0  |    640/50000 (  1%) ] Loss: 0.0290 top1= 98.9062
[E98B10 |   7040/50000 ( 14%) ] Loss: 0.0258 top1= 99.2188
[E98B20 |  13440/50000 ( 27%) ] Loss: 0.0169 top1= 99.3750
[E98B30 |  19840/50000 ( 40%) ] Loss: 0.0165 top1= 99.2188
[E98B40 |  26240/50000 ( 52%) ] Loss: 0.0362 top1= 98.9062
[E98B50 |  32640/50000 ( 65%) ] Loss: 0.0228 top1= 99.0625
[E98B60 |  39040/50000 ( 78%) ] Loss: 0.0099 top1= 99.8438
[E98B70 |  45440/50000 ( 91%) ] Loss: 0.0252 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6199 top1= 72.5962


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7072 top1= 46.2240


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0772 top1= 53.0349

Train epoch 99
[E99B0  |    640/50000 (  1%) ] Loss: 0.0325 top1= 98.7500
[E99B10 |   7040/50000 ( 14%) ] Loss: 0.0171 top1= 99.3750
[E99B20 |  13440/50000 ( 27%) ] Loss: 0.0101 top1= 99.8438
[E99B30 |  19840/50000 ( 40%) ] Loss: 0.0322 top1= 98.9062
[E99B40 |  26240/50000 ( 52%) ] Loss: 0.0336 top1= 98.9062
[E99B50 |  32640/50000 ( 65%) ] Loss: 0.0403 top1= 98.5938
[E99B60 |  39040/50000 ( 78%) ] Loss: 0.0207 top1= 99.5312
[E99B70 |  45440/50000 ( 91%) ] Loss: 0.0110 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5844 top1= 73.3073


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7273 top1= 47.1054


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8364 top1= 52.3037

Train epoch 100
[E100B0  |    640/50000 (  1%) ] Loss: 0.0201 top1= 99.3750
[E100B10 |   7040/50000 ( 14%) ] Loss: 0.0152 top1= 99.6875
[E100B20 |  13440/50000 ( 27%) ] Loss: 0.0213 top1= 99.2188
[E100B30 |  19840/50000 ( 40%) ] Loss: 0.0261 top1= 99.2188
[E100B40 |  26240/50000 ( 52%) ] Loss: 0.0063 top1= 99.6875
[E100B50 |  32640/50000 ( 65%) ] Loss: 0.0318 top1= 99.0625
[E100B60 |  39040/50000 ( 78%) ] Loss: 0.0208 top1= 99.2188
[E100B70 |  45440/50000 ( 91%) ] Loss: 0.0177 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4813 top1= 74.1587


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0266 top1= 47.2155


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1825 top1= 51.1418

Train epoch 101
[E101B0  |    640/50000 (  1%) ] Loss: 0.0250 top1= 99.0625
[E101B10 |   7040/50000 ( 14%) ] Loss: 0.0327 top1= 98.9062
[E101B20 |  13440/50000 ( 27%) ] Loss: 0.0151 top1= 99.5312
[E101B30 |  19840/50000 ( 40%) ] Loss: 0.0152 top1= 99.0625
[E101B40 |  26240/50000 ( 52%) ] Loss: 0.0036 top1=100.0000
[E101B50 |  32640/50000 ( 65%) ] Loss: 0.0414 top1= 98.7500
[E101B60 |  39040/50000 ( 78%) ] Loss: 0.0116 top1= 99.3750
[E101B70 |  45440/50000 ( 91%) ] Loss: 0.0181 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2174 top1= 69.7015


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3544 top1= 48.1871


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4521 top1= 53.5457

Train epoch 102
[E102B0  |    640/50000 (  1%) ] Loss: 0.0298 top1= 99.2188
[E102B10 |   7040/50000 ( 14%) ] Loss: 0.0220 top1= 98.9062
[E102B20 |  13440/50000 ( 27%) ] Loss: 0.0137 top1= 99.6875
[E102B30 |  19840/50000 ( 40%) ] Loss: 0.0257 top1= 99.0625
[E102B40 |  26240/50000 ( 52%) ] Loss: 0.0149 top1= 99.5312
[E102B50 |  32640/50000 ( 65%) ] Loss: 0.0174 top1= 99.6875
[E102B60 |  39040/50000 ( 78%) ] Loss: 0.0192 top1= 99.0625
[E102B70 |  45440/50000 ( 91%) ] Loss: 0.0168 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4870 top1= 74.2488


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.0478 top1= 47.9968


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2594 top1= 51.4022

Train epoch 103
[E103B0  |    640/50000 (  1%) ] Loss: 0.0326 top1= 99.0625
[E103B10 |   7040/50000 ( 14%) ] Loss: 0.0123 top1= 99.6875
[E103B20 |  13440/50000 ( 27%) ] Loss: 0.0165 top1= 99.5312
[E103B30 |  19840/50000 ( 40%) ] Loss: 0.0186 top1= 99.2188
[E103B40 |  26240/50000 ( 52%) ] Loss: 0.0132 top1= 99.3750
[E103B50 |  32640/50000 ( 65%) ] Loss: 0.0071 top1= 99.8438
[E103B60 |  39040/50000 ( 78%) ] Loss: 0.0143 top1= 99.3750
[E103B70 |  45440/50000 ( 91%) ] Loss: 0.0083 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5741 top1= 73.6178


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6679 top1= 49.5292


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0211 top1= 52.2336

Train epoch 104
[E104B0  |    640/50000 (  1%) ] Loss: 0.0194 top1= 99.2188
[E104B10 |   7040/50000 ( 14%) ] Loss: 0.0244 top1= 99.3750
[E104B20 |  13440/50000 ( 27%) ] Loss: 0.0240 top1= 99.2188
[E104B30 |  19840/50000 ( 40%) ] Loss: 0.0138 top1= 99.2188
[E104B40 |  26240/50000 ( 52%) ] Loss: 0.0122 top1= 99.6875
[E104B50 |  32640/50000 ( 65%) ] Loss: 0.0184 top1= 99.3750
[E104B60 |  39040/50000 ( 78%) ] Loss: 0.0250 top1= 99.2188
[E104B70 |  45440/50000 ( 91%) ] Loss: 0.0334 top1= 98.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0294 top1= 70.8834


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3171 top1= 47.0753


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.3471 top1= 47.7564

Train epoch 105
[E105B0  |    640/50000 (  1%) ] Loss: 0.0154 top1= 99.3750
[E105B10 |   7040/50000 ( 14%) ] Loss: 0.0308 top1= 99.2188
[E105B20 |  13440/50000 ( 27%) ] Loss: 0.0166 top1= 99.6875
[E105B30 |  19840/50000 ( 40%) ] Loss: 0.0257 top1= 99.2188
[E105B40 |  26240/50000 ( 52%) ] Loss: 0.0098 top1= 99.6875
[E105B50 |  32640/50000 ( 65%) ] Loss: 0.0398 top1= 98.7500
[E105B60 |  39040/50000 ( 78%) ] Loss: 0.0220 top1= 99.0625
[E105B70 |  45440/50000 ( 91%) ] Loss: 0.0170 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6963 top1= 73.2572


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3539 top1= 47.3357


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0492 top1= 52.3738

Train epoch 106
[E106B0  |    640/50000 (  1%) ] Loss: 0.0181 top1= 99.3750
[E106B10 |   7040/50000 ( 14%) ] Loss: 0.0135 top1= 99.6875
[E106B20 |  13440/50000 ( 27%) ] Loss: 0.0310 top1= 98.4375
[E106B30 |  19840/50000 ( 40%) ] Loss: 0.0138 top1= 99.5312
[E106B40 |  26240/50000 ( 52%) ] Loss: 0.0152 top1= 99.2188
[E106B50 |  32640/50000 ( 65%) ] Loss: 0.0114 top1= 99.6875
[E106B60 |  39040/50000 ( 78%) ] Loss: 0.0074 top1= 99.6875
[E106B70 |  45440/50000 ( 91%) ] Loss: 0.0226 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7803 top1= 72.9267


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.8926 top1= 46.6446


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.7586 top1= 50.6110

Train epoch 107
[E107B0  |    640/50000 (  1%) ] Loss: 0.0222 top1= 99.3750
[E107B10 |   7040/50000 ( 14%) ] Loss: 0.0159 top1= 99.0625
[E107B20 |  13440/50000 ( 27%) ] Loss: 0.0063 top1= 99.8438
[E107B30 |  19840/50000 ( 40%) ] Loss: 0.0110 top1= 99.6875
[E107B40 |  26240/50000 ( 52%) ] Loss: 0.0182 top1= 99.3750
[E107B50 |  32640/50000 ( 65%) ] Loss: 0.0133 top1= 99.3750
[E107B60 |  39040/50000 ( 78%) ] Loss: 0.0179 top1= 99.6875
[E107B70 |  45440/50000 ( 91%) ] Loss: 0.0197 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6251 top1= 74.1587


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4712 top1= 49.8397


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.4286 top1= 50.2604

Train epoch 108
[E108B0  |    640/50000 (  1%) ] Loss: 0.0138 top1= 99.6875
[E108B10 |   7040/50000 ( 14%) ] Loss: 0.0217 top1= 99.0625
[E108B20 |  13440/50000 ( 27%) ] Loss: 0.0078 top1= 99.6875
[E108B30 |  19840/50000 ( 40%) ] Loss: 0.0225 top1= 99.5312
[E108B40 |  26240/50000 ( 52%) ] Loss: 0.0107 top1= 99.5312
[E108B50 |  32640/50000 ( 65%) ] Loss: 0.0219 top1= 99.0625
[E108B60 |  39040/50000 ( 78%) ] Loss: 0.0048 top1= 99.8438
[E108B70 |  45440/50000 ( 91%) ] Loss: 0.0102 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6495 top1= 74.7696


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7577 top1= 49.4892


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.2857 top1= 52.1835

Train epoch 109
[E109B0  |    640/50000 (  1%) ] Loss: 0.0232 top1= 99.2188
[E109B10 |   7040/50000 ( 14%) ] Loss: 0.0157 top1= 99.3750
[E109B20 |  13440/50000 ( 27%) ] Loss: 0.0305 top1= 99.3750
[E109B30 |  19840/50000 ( 40%) ] Loss: 0.0261 top1= 99.0625
[E109B40 |  26240/50000 ( 52%) ] Loss: 0.0326 top1= 99.0625
[E109B50 |  32640/50000 ( 65%) ] Loss: 0.0089 top1= 99.6875
[E109B60 |  39040/50000 ( 78%) ] Loss: 0.0023 top1=100.0000
[E109B70 |  45440/50000 ( 91%) ] Loss: 0.0128 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5881 top1= 74.3389


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3994 top1= 50.6110


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7703 top1= 50.9014

Train epoch 110
[E110B0  |    640/50000 (  1%) ] Loss: 0.0147 top1= 99.5312
[E110B10 |   7040/50000 ( 14%) ] Loss: 0.0158 top1= 99.5312
[E110B20 |  13440/50000 ( 27%) ] Loss: 0.0122 top1= 99.6875
[E110B30 |  19840/50000 ( 40%) ] Loss: 0.0227 top1= 98.9062
[E110B40 |  26240/50000 ( 52%) ] Loss: 0.0127 top1= 99.8438
[E110B50 |  32640/50000 ( 65%) ] Loss: 0.0396 top1= 99.0625
[E110B60 |  39040/50000 ( 78%) ] Loss: 0.0169 top1= 99.2188
[E110B70 |  45440/50000 ( 91%) ] Loss: 0.0182 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7160 top1= 73.6478


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3707 top1= 48.3474


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7549 top1= 51.3421

Train epoch 111
[E111B0  |    640/50000 (  1%) ] Loss: 0.0216 top1= 99.0625
[E111B10 |   7040/50000 ( 14%) ] Loss: 0.0160 top1= 99.3750
[E111B20 |  13440/50000 ( 27%) ] Loss: 0.0088 top1= 99.5312
[E111B30 |  19840/50000 ( 40%) ] Loss: 0.0212 top1= 99.5312
[E111B40 |  26240/50000 ( 52%) ] Loss: 0.0142 top1= 99.5312
[E111B50 |  32640/50000 ( 65%) ] Loss: 0.0371 top1= 98.4375
[E111B60 |  39040/50000 ( 78%) ] Loss: 0.0059 top1= 99.8438
[E111B70 |  45440/50000 ( 91%) ] Loss: 0.0102 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6795 top1= 73.9583


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7306 top1= 49.6194


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7487 top1= 51.2620

Train epoch 112
[E112B0  |    640/50000 (  1%) ] Loss: 0.0314 top1= 98.5938
[E112B10 |   7040/50000 ( 14%) ] Loss: 0.0259 top1= 99.0625
[E112B20 |  13440/50000 ( 27%) ] Loss: 0.0129 top1= 99.5312
[E112B30 |  19840/50000 ( 40%) ] Loss: 0.0237 top1= 99.3750
[E112B40 |  26240/50000 ( 52%) ] Loss: 0.0199 top1= 99.2188
[E112B50 |  32640/50000 ( 65%) ] Loss: 0.0174 top1= 99.3750
[E112B60 |  39040/50000 ( 78%) ] Loss: 0.0260 top1= 99.5312
[E112B70 |  45440/50000 ( 91%) ] Loss: 0.0060 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7959 top1= 73.4776


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3167 top1= 49.5292


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3209 top1= 53.9363

Train epoch 113
[E113B0  |    640/50000 (  1%) ] Loss: 0.0195 top1= 99.5312
[E113B10 |   7040/50000 ( 14%) ] Loss: 0.0152 top1= 99.5312
[E113B20 |  13440/50000 ( 27%) ] Loss: 0.0118 top1= 99.3750
[E113B30 |  19840/50000 ( 40%) ] Loss: 0.0131 top1= 99.6875
[E113B40 |  26240/50000 ( 52%) ] Loss: 0.0233 top1= 98.9062
[E113B50 |  32640/50000 ( 65%) ] Loss: 0.0121 top1= 99.6875
[E113B60 |  39040/50000 ( 78%) ] Loss: 0.0208 top1= 99.6875
[E113B70 |  45440/50000 ( 91%) ] Loss: 0.0188 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8394 top1= 73.5477


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7728 top1= 48.6278


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6501 top1= 54.3870

Train epoch 114
[E114B0  |    640/50000 (  1%) ] Loss: 0.0363 top1= 99.0625
[E114B10 |   7040/50000 ( 14%) ] Loss: 0.0170 top1= 99.5312
[E114B20 |  13440/50000 ( 27%) ] Loss: 0.0055 top1= 99.8438
[E114B30 |  19840/50000 ( 40%) ] Loss: 0.0072 top1= 99.8438
[E114B40 |  26240/50000 ( 52%) ] Loss: 0.0185 top1= 99.2188
[E114B50 |  32640/50000 ( 65%) ] Loss: 0.0226 top1= 99.2188
[E114B60 |  39040/50000 ( 78%) ] Loss: 0.0136 top1= 99.3750
[E114B70 |  45440/50000 ( 91%) ] Loss: 0.0096 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7759 top1= 73.3874


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8921 top1= 48.0869


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.3023 top1= 50.5509

Train epoch 115
[E115B0  |    640/50000 (  1%) ] Loss: 0.0180 top1= 99.0625
[E115B10 |   7040/50000 ( 14%) ] Loss: 0.0088 top1= 99.5312
[E115B20 |  13440/50000 ( 27%) ] Loss: 0.0134 top1= 99.3750
[E115B30 |  19840/50000 ( 40%) ] Loss: 0.0088 top1= 99.5312
[E115B40 |  26240/50000 ( 52%) ] Loss: 0.0108 top1= 99.5312
[E115B50 |  32640/50000 ( 65%) ] Loss: 0.0135 top1= 99.5312
[E115B60 |  39040/50000 ( 78%) ] Loss: 0.0100 top1= 99.6875
[E115B70 |  45440/50000 ( 91%) ] Loss: 0.0381 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7546 top1= 74.0785


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.8821 top1= 47.3057


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.6097 top1= 51.1919

Train epoch 116
[E116B0  |    640/50000 (  1%) ] Loss: 0.0138 top1= 99.3750
[E116B10 |   7040/50000 ( 14%) ] Loss: 0.0208 top1= 99.5312
[E116B20 |  13440/50000 ( 27%) ] Loss: 0.0126 top1= 99.6875
[E116B30 |  19840/50000 ( 40%) ] Loss: 0.0061 top1= 99.8438
[E116B40 |  26240/50000 ( 52%) ] Loss: 0.0134 top1= 99.5312
[E116B50 |  32640/50000 ( 65%) ] Loss: 0.0143 top1= 99.6875
[E116B60 |  39040/50000 ( 78%) ] Loss: 0.0323 top1= 99.5312
[E116B70 |  45440/50000 ( 91%) ] Loss: 0.0144 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6901 top1= 74.2188


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8232 top1= 49.9199


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.6071 top1= 50.8514

Train epoch 117
[E117B0  |    640/50000 (  1%) ] Loss: 0.0129 top1= 99.6875
[E117B10 |   7040/50000 ( 14%) ] Loss: 0.0092 top1= 99.8438
[E117B20 |  13440/50000 ( 27%) ] Loss: 0.0067 top1=100.0000
[E117B30 |  19840/50000 ( 40%) ] Loss: 0.0092 top1= 99.5312
[E117B40 |  26240/50000 ( 52%) ] Loss: 0.0197 top1= 99.3750
[E117B50 |  32640/50000 ( 65%) ] Loss: 0.0139 top1= 99.6875
[E117B60 |  39040/50000 ( 78%) ] Loss: 0.0104 top1= 99.5312
[E117B70 |  45440/50000 ( 91%) ] Loss: 0.0187 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8650 top1= 73.0970


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.7286 top1= 48.0268


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3635 top1= 51.4022

Train epoch 118
[E118B0  |    640/50000 (  1%) ] Loss: 0.0139 top1= 99.5312
[E118B10 |   7040/50000 ( 14%) ] Loss: 0.0113 top1= 99.5312
[E118B20 |  13440/50000 ( 27%) ] Loss: 0.0118 top1= 99.5312
[E118B30 |  19840/50000 ( 40%) ] Loss: 0.0176 top1= 99.5312
[E118B40 |  26240/50000 ( 52%) ] Loss: 0.0088 top1= 99.6875
[E118B50 |  32640/50000 ( 65%) ] Loss: 0.0069 top1=100.0000
[E118B60 |  39040/50000 ( 78%) ] Loss: 0.0049 top1=100.0000
[E118B70 |  45440/50000 ( 91%) ] Loss: 0.0225 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9259 top1= 73.1370


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.5730 top1= 48.3173


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0449 top1= 51.7127

Train epoch 119
[E119B0  |    640/50000 (  1%) ] Loss: 0.0270 top1= 99.2188
[E119B10 |   7040/50000 ( 14%) ] Loss: 0.0165 top1= 99.0625
[E119B20 |  13440/50000 ( 27%) ] Loss: 0.0019 top1=100.0000
[E119B30 |  19840/50000 ( 40%) ] Loss: 0.0163 top1= 99.3750
[E119B40 |  26240/50000 ( 52%) ] Loss: 0.0095 top1= 99.8438
[E119B50 |  32640/50000 ( 65%) ] Loss: 0.0150 top1= 99.2188
[E119B60 |  39040/50000 ( 78%) ] Loss: 0.0139 top1= 99.5312
[E119B70 |  45440/50000 ( 91%) ] Loss: 0.0175 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7438 top1= 74.3189


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3838 top1= 49.6094


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1605 top1= 52.3538

Train epoch 120
[E120B0  |    640/50000 (  1%) ] Loss: 0.0259 top1= 99.3750
[E120B10 |   7040/50000 ( 14%) ] Loss: 0.0169 top1= 99.5312
[E120B20 |  13440/50000 ( 27%) ] Loss: 0.0057 top1= 99.6875
[E120B30 |  19840/50000 ( 40%) ] Loss: 0.0250 top1= 99.0625
[E120B40 |  26240/50000 ( 52%) ] Loss: 0.0213 top1= 98.9062
[E120B50 |  32640/50000 ( 65%) ] Loss: 0.0054 top1= 99.8438
[E120B60 |  39040/50000 ( 78%) ] Loss: 0.0028 top1=100.0000
[E120B70 |  45440/50000 ( 91%) ] Loss: 0.0260 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7078 top1= 74.9800


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7080 top1= 48.9283


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.6101 top1= 51.6526

Train epoch 121
[E121B0  |    640/50000 (  1%) ] Loss: 0.0122 top1= 99.3750
[E121B10 |   7040/50000 ( 14%) ] Loss: 0.0224 top1= 99.0625
[E121B20 |  13440/50000 ( 27%) ] Loss: 0.0447 top1= 98.2812
[E121B30 |  19840/50000 ( 40%) ] Loss: 0.0214 top1= 99.5312
[E121B40 |  26240/50000 ( 52%) ] Loss: 0.0126 top1= 99.5312
[E121B50 |  32640/50000 ( 65%) ] Loss: 0.0093 top1= 99.8438
[E121B60 |  39040/50000 ( 78%) ] Loss: 0.0151 top1= 99.5312
[E121B70 |  45440/50000 ( 91%) ] Loss: 0.0250 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7213 top1= 74.7596


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9941 top1= 55.1583


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3091 top1= 58.0729

Train epoch 122
[E122B0  |    640/50000 (  1%) ] Loss: 0.0228 top1= 99.2188
[E122B10 |   7040/50000 ( 14%) ] Loss: 0.0308 top1= 99.0625
[E122B20 |  13440/50000 ( 27%) ] Loss: 0.0450 top1= 98.5938
[E122B30 |  19840/50000 ( 40%) ] Loss: 0.0412 top1= 98.7500
[E122B40 |  26240/50000 ( 52%) ] Loss: 0.0117 top1= 99.5312
[E122B50 |  32640/50000 ( 65%) ] Loss: 0.0406 top1= 99.0625
[E122B60 |  39040/50000 ( 78%) ] Loss: 0.0046 top1=100.0000
[E122B70 |  45440/50000 ( 91%) ] Loss: 0.0152 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7848 top1= 74.7196


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9545 top1= 52.3137


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9603 top1= 59.4151

Train epoch 123
[E123B0  |    640/50000 (  1%) ] Loss: 0.0190 top1= 99.6875
[E123B10 |   7040/50000 ( 14%) ] Loss: 0.0170 top1= 99.3750
[E123B20 |  13440/50000 ( 27%) ] Loss: 0.0152 top1= 99.2188
[E123B30 |  19840/50000 ( 40%) ] Loss: 0.0167 top1= 99.0625
[E123B40 |  26240/50000 ( 52%) ] Loss: 0.0147 top1= 99.3750
[E123B50 |  32640/50000 ( 65%) ] Loss: 0.0170 top1= 99.3750
[E123B60 |  39040/50000 ( 78%) ] Loss: 0.0227 top1= 99.5312
[E123B70 |  45440/50000 ( 91%) ] Loss: 0.0245 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7777 top1= 74.7897


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4928 top1= 53.6558


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6153 top1= 60.6671

Train epoch 124
[E124B0  |    640/50000 (  1%) ] Loss: 0.0216 top1= 99.0625
[E124B10 |   7040/50000 ( 14%) ] Loss: 0.0287 top1= 98.9062
[E124B20 |  13440/50000 ( 27%) ] Loss: 0.0131 top1= 99.6875
[E124B30 |  19840/50000 ( 40%) ] Loss: 0.0169 top1= 99.2188
[E124B40 |  26240/50000 ( 52%) ] Loss: 0.0019 top1=100.0000
[E124B50 |  32640/50000 ( 65%) ] Loss: 0.0172 top1= 99.5312
[E124B60 |  39040/50000 ( 78%) ] Loss: 0.0224 top1= 99.2188
[E124B70 |  45440/50000 ( 91%) ] Loss: 0.0140 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7532 top1= 74.8197


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7593 top1= 53.2552


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5438 top1= 57.7424

Train epoch 125
[E125B0  |    640/50000 (  1%) ] Loss: 0.0098 top1= 99.8438
[E125B10 |   7040/50000 ( 14%) ] Loss: 0.0152 top1= 99.3750
[E125B20 |  13440/50000 ( 27%) ] Loss: 0.0122 top1= 99.8438
[E125B30 |  19840/50000 ( 40%) ] Loss: 0.0206 top1= 99.2188
[E125B40 |  26240/50000 ( 52%) ] Loss: 0.0230 top1= 98.5938
[E125B50 |  32640/50000 ( 65%) ] Loss: 0.0153 top1= 99.6875
[E125B60 |  39040/50000 ( 78%) ] Loss: 0.0084 top1= 99.6875
[E125B70 |  45440/50000 ( 91%) ] Loss: 0.0191 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6851 top1= 75.1502


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2641 top1= 54.1266


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6911 top1= 57.0112

Train epoch 126
[E126B0  |    640/50000 (  1%) ] Loss: 0.0297 top1= 99.2188
[E126B10 |   7040/50000 ( 14%) ] Loss: 0.0222 top1= 99.2188
[E126B20 |  13440/50000 ( 27%) ] Loss: 0.0065 top1= 99.8438
[E126B30 |  19840/50000 ( 40%) ] Loss: 0.0218 top1= 99.3750
[E126B40 |  26240/50000 ( 52%) ] Loss: 0.0078 top1= 99.6875
[E126B50 |  32640/50000 ( 65%) ] Loss: 0.0146 top1= 99.6875
[E126B60 |  39040/50000 ( 78%) ] Loss: 0.0154 top1= 99.6875
[E126B70 |  45440/50000 ( 91%) ] Loss: 0.0152 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6900 top1= 75.0100


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6244 top1= 53.1450


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2751 top1= 55.6190

Train epoch 127
[E127B0  |    640/50000 (  1%) ] Loss: 0.0093 top1= 99.6875
[E127B10 |   7040/50000 ( 14%) ] Loss: 0.0445 top1= 98.5938
[E127B20 |  13440/50000 ( 27%) ] Loss: 0.0198 top1= 99.3750
[E127B30 |  19840/50000 ( 40%) ] Loss: 0.0345 top1= 99.3750
[E127B40 |  26240/50000 ( 52%) ] Loss: 0.0132 top1= 99.5312
[E127B50 |  32640/50000 ( 65%) ] Loss: 0.0155 top1= 99.6875
[E127B60 |  39040/50000 ( 78%) ] Loss: 0.0118 top1= 99.5312
[E127B70 |  45440/50000 ( 91%) ] Loss: 0.0044 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7860 top1= 74.4792


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9648 top1= 52.5040


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0155 top1= 58.9042

Train epoch 128
[E128B0  |    640/50000 (  1%) ] Loss: 0.0246 top1= 98.9062
[E128B10 |   7040/50000 ( 14%) ] Loss: 0.0166 top1= 99.3750
[E128B20 |  13440/50000 ( 27%) ] Loss: 0.0165 top1= 99.6875
[E128B30 |  19840/50000 ( 40%) ] Loss: 0.0215 top1= 99.3750
[E128B40 |  26240/50000 ( 52%) ] Loss: 0.0077 top1= 99.6875
[E128B50 |  32640/50000 ( 65%) ] Loss: 0.0091 top1= 99.6875
[E128B60 |  39040/50000 ( 78%) ] Loss: 0.0136 top1= 99.6875
[E128B70 |  45440/50000 ( 91%) ] Loss: 0.0094 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8017 top1= 74.2788


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1958 top1= 51.7027


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1876 top1= 58.2232

Train epoch 129
[E129B0  |    640/50000 (  1%) ] Loss: 0.0168 top1= 99.5312
[E129B10 |   7040/50000 ( 14%) ] Loss: 0.0318 top1= 98.9062
[E129B20 |  13440/50000 ( 27%) ] Loss: 0.0034 top1=100.0000
[E129B30 |  19840/50000 ( 40%) ] Loss: 0.0131 top1= 99.5312
[E129B40 |  26240/50000 ( 52%) ] Loss: 0.0074 top1= 99.5312
[E129B50 |  32640/50000 ( 65%) ] Loss: 0.0143 top1= 99.5312
[E129B60 |  39040/50000 ( 78%) ] Loss: 0.0153 top1= 99.5312
[E129B70 |  45440/50000 ( 91%) ] Loss: 0.0083 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7773 top1= 74.4591


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3298 top1= 53.4355


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.2283 top1= 61.2981

Train epoch 130
[E130B0  |    640/50000 (  1%) ] Loss: 0.0117 top1= 99.6875
[E130B10 |   7040/50000 ( 14%) ] Loss: 0.0058 top1= 99.8438
[E130B20 |  13440/50000 ( 27%) ] Loss: 0.0184 top1= 99.5312
[E130B30 |  19840/50000 ( 40%) ] Loss: 0.0321 top1= 98.5938
[E130B40 |  26240/50000 ( 52%) ] Loss: 0.0124 top1= 99.5312
[E130B50 |  32640/50000 ( 65%) ] Loss: 0.0171 top1= 99.5312
[E130B60 |  39040/50000 ( 78%) ] Loss: 0.0052 top1= 99.8438
[E130B70 |  45440/50000 ( 91%) ] Loss: 0.0106 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6846 top1= 74.5994


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5364 top1= 53.0048


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6938 top1= 56.3602

Train epoch 131
[E131B0  |    640/50000 (  1%) ] Loss: 0.0111 top1= 99.6875
[E131B10 |   7040/50000 ( 14%) ] Loss: 0.0135 top1= 99.6875
[E131B20 |  13440/50000 ( 27%) ] Loss: 0.0183 top1= 99.3750
[E131B30 |  19840/50000 ( 40%) ] Loss: 0.0086 top1= 99.6875
[E131B40 |  26240/50000 ( 52%) ] Loss: 0.0135 top1= 99.6875
[E131B50 |  32640/50000 ( 65%) ] Loss: 0.0131 top1= 99.3750
[E131B60 |  39040/50000 ( 78%) ] Loss: 0.0124 top1= 99.5312
[E131B70 |  45440/50000 ( 91%) ] Loss: 0.0184 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6763 top1= 74.7997


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4628 top1= 53.0248


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8376 top1= 56.3201

Train epoch 132
[E132B0  |    640/50000 (  1%) ] Loss: 0.0071 top1=100.0000
[E132B10 |   7040/50000 ( 14%) ] Loss: 0.0113 top1= 99.5312
[E132B20 |  13440/50000 ( 27%) ] Loss: 0.0078 top1= 99.8438
[E132B30 |  19840/50000 ( 40%) ] Loss: 0.0116 top1= 99.5312
[E132B40 |  26240/50000 ( 52%) ] Loss: 0.0190 top1= 99.5312
[E132B50 |  32640/50000 ( 65%) ] Loss: 0.0134 top1= 99.5312
[E132B60 |  39040/50000 ( 78%) ] Loss: 0.0083 top1= 99.8438
[E132B70 |  45440/50000 ( 91%) ] Loss: 0.0169 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7065 top1= 74.7296


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8659 top1= 54.7776


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5634 top1= 60.3065

Train epoch 133
[E133B0  |    640/50000 (  1%) ] Loss: 0.0169 top1= 99.2188
[E133B10 |   7040/50000 ( 14%) ] Loss: 0.0084 top1= 99.8438
[E133B20 |  13440/50000 ( 27%) ] Loss: 0.0061 top1= 99.6875
[E133B30 |  19840/50000 ( 40%) ] Loss: 0.0077 top1=100.0000
[E133B40 |  26240/50000 ( 52%) ] Loss: 0.0290 top1= 99.2188
[E133B50 |  32640/50000 ( 65%) ] Loss: 0.0101 top1= 99.6875
[E133B60 |  39040/50000 ( 78%) ] Loss: 0.0359 top1= 99.2188
[E133B70 |  45440/50000 ( 91%) ] Loss: 0.0056 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6923 top1= 74.9099


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1456 top1= 53.8061


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4538 top1= 57.5421

Train epoch 134
[E134B0  |    640/50000 (  1%) ] Loss: 0.0133 top1= 99.6875
[E134B10 |   7040/50000 ( 14%) ] Loss: 0.0211 top1= 99.5312
[E134B20 |  13440/50000 ( 27%) ] Loss: 0.0112 top1= 99.6875
[E134B30 |  19840/50000 ( 40%) ] Loss: 0.0160 top1= 99.2188
[E134B40 |  26240/50000 ( 52%) ] Loss: 0.0087 top1= 99.8438
[E134B50 |  32640/50000 ( 65%) ] Loss: 0.0047 top1=100.0000
[E134B60 |  39040/50000 ( 78%) ] Loss: 0.0088 top1= 99.6875
[E134B70 |  45440/50000 ( 91%) ] Loss: 0.0217 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6943 top1= 74.6294


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5534 top1= 55.4788


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6025 top1= 56.9211

Train epoch 135
[E135B0  |    640/50000 (  1%) ] Loss: 0.0150 top1= 99.3750
[E135B10 |   7040/50000 ( 14%) ] Loss: 0.0109 top1= 99.5312
[E135B20 |  13440/50000 ( 27%) ] Loss: 0.0161 top1= 99.3750
[E135B30 |  19840/50000 ( 40%) ] Loss: 0.0047 top1= 99.8438
[E135B40 |  26240/50000 ( 52%) ] Loss: 0.0405 top1= 99.5312
[E135B50 |  32640/50000 ( 65%) ] Loss: 0.0196 top1= 99.6875
[E135B60 |  39040/50000 ( 78%) ] Loss: 0.0064 top1= 99.8438
[E135B70 |  45440/50000 ( 91%) ] Loss: 0.0333 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7391 top1= 74.3289


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8643 top1= 52.2135


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3690 top1= 57.8125

Train epoch 136
[E136B0  |    640/50000 (  1%) ] Loss: 0.0153 top1= 99.5312
[E136B10 |   7040/50000 ( 14%) ] Loss: 0.0230 top1= 99.3750
[E136B20 |  13440/50000 ( 27%) ] Loss: 0.0166 top1= 99.5312
[E136B30 |  19840/50000 ( 40%) ] Loss: 0.0092 top1= 99.6875
[E136B40 |  26240/50000 ( 52%) ] Loss: 0.0113 top1= 99.3750
[E136B50 |  32640/50000 ( 65%) ] Loss: 0.0103 top1= 99.8438
[E136B60 |  39040/50000 ( 78%) ] Loss: 0.0092 top1= 99.6875
[E136B70 |  45440/50000 ( 91%) ] Loss: 0.0179 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6943 top1= 74.6695


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8367 top1= 54.9179


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2976 top1= 58.3834

Train epoch 137
[E137B0  |    640/50000 (  1%) ] Loss: 0.0095 top1= 99.6875
[E137B10 |   7040/50000 ( 14%) ] Loss: 0.0130 top1= 99.5312
[E137B20 |  13440/50000 ( 27%) ] Loss: 0.0095 top1= 99.6875
[E137B30 |  19840/50000 ( 40%) ] Loss: 0.0086 top1= 99.8438
[E137B40 |  26240/50000 ( 52%) ] Loss: 0.0177 top1= 99.6875
[E137B50 |  32640/50000 ( 65%) ] Loss: 0.0124 top1= 99.5312
[E137B60 |  39040/50000 ( 78%) ] Loss: 0.0206 top1= 99.3750
[E137B70 |  45440/50000 ( 91%) ] Loss: 0.0089 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8298 top1= 74.1486


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.9035 top1= 51.9932


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.3527 top1= 61.0978

Train epoch 138
[E138B0  |    640/50000 (  1%) ] Loss: 0.0239 top1= 99.0625
[E138B10 |   7040/50000 ( 14%) ] Loss: 0.0118 top1= 99.6875
[E138B20 |  13440/50000 ( 27%) ] Loss: 0.0058 top1=100.0000
[E138B30 |  19840/50000 ( 40%) ] Loss: 0.0314 top1= 99.3750
[E138B40 |  26240/50000 ( 52%) ] Loss: 0.0107 top1= 99.5312
[E138B50 |  32640/50000 ( 65%) ] Loss: 0.0188 top1= 99.5312
[E138B60 |  39040/50000 ( 78%) ] Loss: 0.0103 top1= 99.8438
[E138B70 |  45440/50000 ( 91%) ] Loss: 0.0208 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6835 top1= 74.8698


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5153 top1= 53.0849


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7710 top1= 56.5605

Train epoch 139
[E139B0  |    640/50000 (  1%) ] Loss: 0.0199 top1= 99.5312
[E139B10 |   7040/50000 ( 14%) ] Loss: 0.0123 top1= 99.5312
[E139B20 |  13440/50000 ( 27%) ] Loss: 0.0340 top1= 99.3750
[E139B30 |  19840/50000 ( 40%) ] Loss: 0.0132 top1= 99.5312
[E139B40 |  26240/50000 ( 52%) ] Loss: 0.0143 top1= 99.2188
[E139B50 |  32640/50000 ( 65%) ] Loss: 0.0137 top1= 99.5312
[E139B60 |  39040/50000 ( 78%) ] Loss: 0.0133 top1= 99.3750
[E139B70 |  45440/50000 ( 91%) ] Loss: 0.0087 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7629 top1= 74.6194


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6859 top1= 52.6342


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8904 top1= 59.1847

Train epoch 140
[E140B0  |    640/50000 (  1%) ] Loss: 0.0184 top1= 99.6875
[E140B10 |   7040/50000 ( 14%) ] Loss: 0.0151 top1= 99.3750
[E140B20 |  13440/50000 ( 27%) ] Loss: 0.0095 top1= 99.6875
[E140B30 |  19840/50000 ( 40%) ] Loss: 0.0140 top1= 99.3750
[E140B40 |  26240/50000 ( 52%) ] Loss: 0.0053 top1= 99.8438
[E140B50 |  32640/50000 ( 65%) ] Loss: 0.0109 top1= 99.8438
[E140B60 |  39040/50000 ( 78%) ] Loss: 0.0084 top1= 99.5312
[E140B70 |  45440/50000 ( 91%) ] Loss: 0.0117 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6790 top1= 74.9900


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0626 top1= 54.0064


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3667 top1= 57.7624

Train epoch 141
[E141B0  |    640/50000 (  1%) ] Loss: 0.0295 top1= 99.0625
[E141B10 |   7040/50000 ( 14%) ] Loss: 0.0154 top1= 99.3750
[E141B20 |  13440/50000 ( 27%) ] Loss: 0.0110 top1= 99.6875
[E141B30 |  19840/50000 ( 40%) ] Loss: 0.0090 top1= 99.5312
[E141B40 |  26240/50000 ( 52%) ] Loss: 0.0069 top1= 99.6875
[E141B50 |  32640/50000 ( 65%) ] Loss: 0.0069 top1= 99.8438
[E141B60 |  39040/50000 ( 78%) ] Loss: 0.0032 top1=100.0000
[E141B70 |  45440/50000 ( 91%) ] Loss: 0.0033 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7284 top1= 74.7396


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9000 top1= 54.4872


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8195 top1= 59.4451

Train epoch 142
[E142B0  |    640/50000 (  1%) ] Loss: 0.0083 top1= 99.8438
[E142B10 |   7040/50000 ( 14%) ] Loss: 0.0085 top1= 99.8438
[E142B20 |  13440/50000 ( 27%) ] Loss: 0.0083 top1= 99.8438
[E142B30 |  19840/50000 ( 40%) ] Loss: 0.0246 top1= 99.3750
[E142B40 |  26240/50000 ( 52%) ] Loss: 0.0105 top1= 99.5312
[E142B50 |  32640/50000 ( 65%) ] Loss: 0.0122 top1= 99.3750
[E142B60 |  39040/50000 ( 78%) ] Loss: 0.0077 top1= 99.6875
[E142B70 |  45440/50000 ( 91%) ] Loss: 0.0063 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6959 top1= 75.0501


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0146 top1= 54.2067


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2771 top1= 57.9327

Train epoch 143
[E143B0  |    640/50000 (  1%) ] Loss: 0.0155 top1= 99.6875
[E143B10 |   7040/50000 ( 14%) ] Loss: 0.0126 top1= 99.5312
[E143B20 |  13440/50000 ( 27%) ] Loss: 0.0071 top1= 99.8438
[E143B30 |  19840/50000 ( 40%) ] Loss: 0.0098 top1= 99.5312
[E143B40 |  26240/50000 ( 52%) ] Loss: 0.0171 top1= 99.2188
[E143B50 |  32640/50000 ( 65%) ] Loss: 0.0102 top1= 99.6875
[E143B60 |  39040/50000 ( 78%) ] Loss: 0.0079 top1= 99.8438
[E143B70 |  45440/50000 ( 91%) ] Loss: 0.0247 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7105 top1= 75.0100


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6854 top1= 52.4940


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5668 top1= 57.3918

Train epoch 144
[E144B0  |    640/50000 (  1%) ] Loss: 0.0077 top1= 99.6875
[E144B10 |   7040/50000 ( 14%) ] Loss: 0.0274 top1= 99.2188
[E144B20 |  13440/50000 ( 27%) ] Loss: 0.0058 top1= 99.8438
[E144B30 |  19840/50000 ( 40%) ] Loss: 0.0113 top1= 99.6875
[E144B40 |  26240/50000 ( 52%) ] Loss: 0.0249 top1= 99.5312
[E144B50 |  32640/50000 ( 65%) ] Loss: 0.0118 top1= 99.6875
[E144B60 |  39040/50000 ( 78%) ] Loss: 0.0075 top1= 99.6875
[E144B70 |  45440/50000 ( 91%) ] Loss: 0.0070 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7628 top1= 74.6294


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6721 top1= 52.6042


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0744 top1= 58.6939

Train epoch 145
[E145B0  |    640/50000 (  1%) ] Loss: 0.0087 top1= 99.5312
[E145B10 |   7040/50000 ( 14%) ] Loss: 0.0147 top1= 99.5312
[E145B20 |  13440/50000 ( 27%) ] Loss: 0.0176 top1= 99.5312
[E145B30 |  19840/50000 ( 40%) ] Loss: 0.0063 top1= 99.8438
[E145B40 |  26240/50000 ( 52%) ] Loss: 0.0070 top1= 99.6875
[E145B50 |  32640/50000 ( 65%) ] Loss: 0.0156 top1= 99.5312
[E145B60 |  39040/50000 ( 78%) ] Loss: 0.0058 top1= 99.8438
[E145B70 |  45440/50000 ( 91%) ] Loss: 0.0098 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7064 top1= 74.5493


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3332 top1= 53.5457


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2347 top1= 55.6290

Train epoch 146
[E146B0  |    640/50000 (  1%) ] Loss: 0.0059 top1=100.0000
[E146B10 |   7040/50000 ( 14%) ] Loss: 0.0120 top1= 99.3750
[E146B20 |  13440/50000 ( 27%) ] Loss: 0.0263 top1= 99.2188
[E146B30 |  19840/50000 ( 40%) ] Loss: 0.0123 top1= 99.5312
[E146B40 |  26240/50000 ( 52%) ] Loss: 0.0048 top1= 99.8438
[E146B50 |  32640/50000 ( 65%) ] Loss: 0.0170 top1= 99.5312
[E146B60 |  39040/50000 ( 78%) ] Loss: 0.0029 top1=100.0000
[E146B70 |  45440/50000 ( 91%) ] Loss: 0.0091 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8820 top1= 73.4976


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2555 top1= 51.3522


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5392 top1= 60.3065

Train epoch 147
[E147B0  |    640/50000 (  1%) ] Loss: 0.0198 top1= 99.3750
[E147B10 |   7040/50000 ( 14%) ] Loss: 0.0098 top1= 99.8438
[E147B20 |  13440/50000 ( 27%) ] Loss: 0.0213 top1= 99.3750
[E147B30 |  19840/50000 ( 40%) ] Loss: 0.0123 top1= 99.3750
[E147B40 |  26240/50000 ( 52%) ] Loss: 0.0064 top1= 99.8438
[E147B50 |  32640/50000 ( 65%) ] Loss: 0.0139 top1= 99.8438
[E147B60 |  39040/50000 ( 78%) ] Loss: 0.0081 top1= 99.3750
[E147B70 |  45440/50000 ( 91%) ] Loss: 0.0208 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8313 top1= 74.1386


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.4442 top1= 50.9716


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.0222 top1= 58.8542

Train epoch 148
[E148B0  |    640/50000 (  1%) ] Loss: 0.0217 top1= 99.2188
[E148B10 |   7040/50000 ( 14%) ] Loss: 0.0208 top1= 98.9062
[E148B20 |  13440/50000 ( 27%) ] Loss: 0.0114 top1= 99.5312
[E148B30 |  19840/50000 ( 40%) ] Loss: 0.0138 top1= 99.6875
[E148B40 |  26240/50000 ( 52%) ] Loss: 0.0083 top1= 99.8438
[E148B50 |  32640/50000 ( 65%) ] Loss: 0.0142 top1= 99.6875
[E148B60 |  39040/50000 ( 78%) ] Loss: 0.0056 top1= 99.8438
[E148B70 |  45440/50000 ( 91%) ] Loss: 0.0122 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8452 top1= 74.0885


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1975 top1= 51.4623


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7382 top1= 59.8758

Train epoch 149
[E149B0  |    640/50000 (  1%) ] Loss: 0.0208 top1= 98.9062
[E149B10 |   7040/50000 ( 14%) ] Loss: 0.0092 top1= 99.3750
[E149B20 |  13440/50000 ( 27%) ] Loss: 0.0075 top1= 99.8438
[E149B30 |  19840/50000 ( 40%) ] Loss: 0.0053 top1= 99.8438
[E149B40 |  26240/50000 ( 52%) ] Loss: 0.0126 top1= 99.5312
[E149B50 |  32640/50000 ( 65%) ] Loss: 0.0179 top1= 99.6875
[E149B60 |  39040/50000 ( 78%) ] Loss: 0.0060 top1= 99.6875
[E149B70 |  45440/50000 ( 91%) ] Loss: 0.0128 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7172 top1= 74.8297


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7994 top1= 52.2135


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5338 top1= 57.4319

Train epoch 150
[E150B0  |    640/50000 (  1%) ] Loss: 0.0195 top1= 99.2188
[E150B10 |   7040/50000 ( 14%) ] Loss: 0.0087 top1= 99.6875
[E150B20 |  13440/50000 ( 27%) ] Loss: 0.0063 top1=100.0000
[E150B30 |  19840/50000 ( 40%) ] Loss: 0.0148 top1= 99.5312
[E150B40 |  26240/50000 ( 52%) ] Loss: 0.0074 top1= 99.6875
[E150B50 |  32640/50000 ( 65%) ] Loss: 0.0116 top1= 99.5312
[E150B60 |  39040/50000 ( 78%) ] Loss: 0.0048 top1=100.0000
[E150B70 |  45440/50000 ( 91%) ] Loss: 0.0222 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7145 top1= 74.9199


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.4067 top1= 53.3454


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3887 top1= 57.7925

