
=== 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 0x7f6656441730>

{'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= 10.0000

=== 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: 1.9985 top1= 19.3750
[E 1B20 |  13440/50000 ( 27%) ] Loss: 1.7857 top1= 19.5312
[E 1B30 |  19840/50000 ( 40%) ] Loss: 1.6367 top1= 23.7500
[E 1B40 |  26240/50000 ( 52%) ] Loss: 1.5866 top1= 28.1250
[E 1B50 |  32640/50000 ( 65%) ] Loss: 1.5885 top1= 24.0625
[E 1B60 |  39040/50000 ( 78%) ] Loss: 1.5977 top1= 26.4062
[E 1B70 |  45440/50000 ( 91%) ] Loss: 1.6015 top1= 24.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3071 top1=  9.6655


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=3.8712 top1= 10.0060


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.1407 top1= 14.7035

Train epoch 2
[E 2B0  |    640/50000 (  1%) ] Loss: 1.5756 top1= 24.8438
[E 2B10 |   7040/50000 ( 14%) ] Loss: 1.5612 top1= 29.2188
[E 2B20 |  13440/50000 ( 27%) ] Loss: 1.4780 top1= 32.8125
[E 2B30 |  19840/50000 ( 40%) ] Loss: 1.4328 top1= 38.1250
[E 2B40 |  26240/50000 ( 52%) ] Loss: 1.4902 top1= 31.7188
[E 2B50 |  32640/50000 ( 65%) ] Loss: 1.4936 top1= 33.2812
[E 2B60 |  39040/50000 ( 78%) ] Loss: 1.5868 top1= 33.1250
[E 2B70 |  45440/50000 ( 91%) ] Loss: 1.5107 top1= 31.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3088 top1=  9.9960


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.0890 top1= 13.7821


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.3562 top1= 21.2941

Train epoch 3
[E 3B0  |    640/50000 (  1%) ] Loss: 1.4702 top1= 34.3750
[E 3B10 |   7040/50000 ( 14%) ] Loss: 1.4532 top1= 30.9375
[E 3B20 |  13440/50000 ( 27%) ] Loss: 1.4261 top1= 38.4375
[E 3B30 |  19840/50000 ( 40%) ] Loss: 1.4578 top1= 34.3750
[E 3B40 |  26240/50000 ( 52%) ] Loss: 1.2979 top1= 41.2500
[E 3B50 |  32640/50000 ( 65%) ] Loss: 1.3257 top1= 40.1562
[E 3B60 |  39040/50000 ( 78%) ] Loss: 1.3275 top1= 42.8125
[E 3B70 |  45440/50000 ( 91%) ] Loss: 1.2047 top1= 48.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3367 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5438 top1= 18.6198


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9737 top1= 29.8377

Train epoch 4
[E 4B0  |    640/50000 (  1%) ] Loss: 1.1763 top1= 51.4062
[E 4B10 |   7040/50000 ( 14%) ] Loss: 1.2566 top1= 47.1875
[E 4B20 |  13440/50000 ( 27%) ] Loss: 1.2238 top1= 49.3750
[E 4B30 |  19840/50000 ( 40%) ] Loss: 1.1545 top1= 53.7500
[E 4B40 |  26240/50000 ( 52%) ] Loss: 1.1226 top1= 54.8438
[E 4B50 |  32640/50000 ( 65%) ] Loss: 1.2392 top1= 49.8438
[E 4B60 |  39040/50000 ( 78%) ] Loss: 1.1557 top1= 53.4375
[E 4B70 |  45440/50000 ( 91%) ] Loss: 1.1265 top1= 53.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3528 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5313 top1= 24.9900


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9839 top1= 30.4788

Train epoch 5
[E 5B0  |    640/50000 (  1%) ] Loss: 1.1735 top1= 52.0312
[E 5B10 |   7040/50000 ( 14%) ] Loss: 1.1262 top1= 53.7500
[E 5B20 |  13440/50000 ( 27%) ] Loss: 1.0790 top1= 53.9062
[E 5B30 |  19840/50000 ( 40%) ] Loss: 1.0446 top1= 58.1250
[E 5B40 |  26240/50000 ( 52%) ] Loss: 0.9976 top1= 60.1562
[E 5B50 |  32640/50000 ( 65%) ] Loss: 1.0686 top1= 58.1250
[E 5B60 |  39040/50000 ( 78%) ] Loss: 1.0152 top1= 61.2500
[E 5B70 |  45440/50000 ( 91%) ] Loss: 1.0474 top1= 58.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3672 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.7154 top1= 26.5525


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6112 top1= 31.6106

Train epoch 6
[E 6B0  |    640/50000 (  1%) ] Loss: 1.1110 top1= 53.5938
[E 6B10 |   7040/50000 ( 14%) ] Loss: 1.0131 top1= 60.3125
[E 6B20 |  13440/50000 ( 27%) ] Loss: 0.9650 top1= 61.7188
[E 6B30 |  19840/50000 ( 40%) ] Loss: 0.9933 top1= 60.0000
[E 6B40 |  26240/50000 ( 52%) ] Loss: 1.0064 top1= 59.2188
[E 6B50 |  32640/50000 ( 65%) ] Loss: 0.9978 top1= 60.6250
[E 6B60 |  39040/50000 ( 78%) ] Loss: 0.9446 top1= 64.3750
[E 6B70 |  45440/50000 ( 91%) ] Loss: 0.9705 top1= 59.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3949 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.7442 top1= 27.6943


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6337 top1= 35.5469

Train epoch 7
[E 7B0  |    640/50000 (  1%) ] Loss: 1.0036 top1= 60.0000
[E 7B10 |   7040/50000 ( 14%) ] Loss: 0.9618 top1= 64.8438
[E 7B20 |  13440/50000 ( 27%) ] Loss: 0.9697 top1= 62.5000
[E 7B30 |  19840/50000 ( 40%) ] Loss: 0.9234 top1= 63.1250
[E 7B40 |  26240/50000 ( 52%) ] Loss: 0.9174 top1= 64.2188
[E 7B50 |  32640/50000 ( 65%) ] Loss: 0.9615 top1= 61.4062
[E 7B60 |  39040/50000 ( 78%) ] Loss: 0.9522 top1= 64.8438
[E 7B70 |  45440/50000 ( 91%) ] Loss: 0.9058 top1= 60.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4266 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.5440 top1= 29.9179


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.2814 top1= 34.2748

Train epoch 8
[E 8B0  |    640/50000 (  1%) ] Loss: 0.9472 top1= 61.4062
[E 8B10 |   7040/50000 ( 14%) ] Loss: 0.9067 top1= 66.2500
[E 8B20 |  13440/50000 ( 27%) ] Loss: 0.8465 top1= 68.9062
[E 8B30 |  19840/50000 ( 40%) ] Loss: 0.9006 top1= 65.9375
[E 8B40 |  26240/50000 ( 52%) ] Loss: 0.8127 top1= 69.0625
[E 8B50 |  32640/50000 ( 65%) ] Loss: 0.8585 top1= 67.9688
[E 8B60 |  39040/50000 ( 78%) ] Loss: 0.8469 top1= 66.7188
[E 8B70 |  45440/50000 ( 91%) ] Loss: 0.8622 top1= 64.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4707 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.8264 top1= 30.7792


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6428 top1= 37.9207

Train epoch 9
[E 9B0  |    640/50000 (  1%) ] Loss: 0.8776 top1= 65.7812
[E 9B10 |   7040/50000 ( 14%) ] Loss: 0.9152 top1= 67.0312
[E 9B20 |  13440/50000 ( 27%) ] Loss: 0.7991 top1= 66.4062
[E 9B30 |  19840/50000 ( 40%) ] Loss: 0.7906 top1= 71.4062
[E 9B40 |  26240/50000 ( 52%) ] Loss: 0.7938 top1= 69.6875
[E 9B50 |  32640/50000 ( 65%) ] Loss: 0.8258 top1= 68.7500
[E 9B60 |  39040/50000 ( 78%) ] Loss: 0.8216 top1= 68.4375
[E 9B70 |  45440/50000 ( 91%) ] Loss: 0.7586 top1= 69.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5026 top1= 10.0461


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.7526 top1= 31.9712


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7808 top1= 38.4816

Train epoch 10
[E10B0  |    640/50000 (  1%) ] Loss: 0.8029 top1= 69.0625
[E10B10 |   7040/50000 ( 14%) ] Loss: 0.8331 top1= 67.5000
[E10B20 |  13440/50000 ( 27%) ] Loss: 0.6913 top1= 73.2812
[E10B30 |  19840/50000 ( 40%) ] Loss: 0.7398 top1= 70.7812
[E10B40 |  26240/50000 ( 52%) ] Loss: 0.7726 top1= 68.5938
[E10B50 |  32640/50000 ( 65%) ] Loss: 0.8133 top1= 67.9688
[E10B60 |  39040/50000 ( 78%) ] Loss: 0.7740 top1= 70.1562
[E10B70 |  45440/50000 ( 91%) ] Loss: 0.7665 top1= 70.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5326 top1= 10.0160


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.8535 top1= 32.9728


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6712 top1= 38.2812

Train epoch 11
[E11B0  |    640/50000 (  1%) ] Loss: 0.7581 top1= 70.0000
[E11B10 |   7040/50000 ( 14%) ] Loss: 0.7799 top1= 71.8750
[E11B20 |  13440/50000 ( 27%) ] Loss: 0.6978 top1= 74.5312
[E11B30 |  19840/50000 ( 40%) ] Loss: 0.6901 top1= 72.9688
[E11B40 |  26240/50000 ( 52%) ] Loss: 0.7018 top1= 73.1250
[E11B50 |  32640/50000 ( 65%) ] Loss: 0.7279 top1= 71.5625
[E11B60 |  39040/50000 ( 78%) ] Loss: 0.6421 top1= 75.9375
[E11B70 |  45440/50000 ( 91%) ] Loss: 0.6719 top1= 75.1562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5365 top1= 10.1062


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.8408 top1= 33.0429


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5622 top1= 40.0341

Train epoch 12
[E12B0  |    640/50000 (  1%) ] Loss: 0.7161 top1= 72.6562
[E12B10 |   7040/50000 ( 14%) ] Loss: 0.7358 top1= 73.5938
[E12B20 |  13440/50000 ( 27%) ] Loss: 0.6553 top1= 74.3750
[E12B30 |  19840/50000 ( 40%) ] Loss: 0.6820 top1= 72.6562
[E12B40 |  26240/50000 ( 52%) ] Loss: 0.6345 top1= 74.2188
[E12B50 |  32640/50000 ( 65%) ] Loss: 0.6965 top1= 74.8438
[E12B60 |  39040/50000 ( 78%) ] Loss: 0.6795 top1= 73.7500
[E12B70 |  45440/50000 ( 91%) ] Loss: 0.6132 top1= 74.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5527 top1= 10.0962


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2082 top1= 34.1446


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5972 top1= 41.1158

Train epoch 13
[E13B0  |    640/50000 (  1%) ] Loss: 0.6929 top1= 72.6562
[E13B10 |   7040/50000 ( 14%) ] Loss: 0.6621 top1= 74.3750
[E13B20 |  13440/50000 ( 27%) ] Loss: 0.5977 top1= 76.8750
[E13B30 |  19840/50000 ( 40%) ] Loss: 0.6029 top1= 77.1875
[E13B40 |  26240/50000 ( 52%) ] Loss: 0.6168 top1= 75.9375
[E13B50 |  32640/50000 ( 65%) ] Loss: 0.5717 top1= 78.1250
[E13B60 |  39040/50000 ( 78%) ] Loss: 0.6793 top1= 75.7812
[E13B70 |  45440/50000 ( 91%) ] Loss: 0.6628 top1= 76.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5496 top1= 10.2264


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1541 top1= 34.8357


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6438 top1= 40.9856

Train epoch 14
[E14B0  |    640/50000 (  1%) ] Loss: 0.6372 top1= 76.2500
[E14B10 |   7040/50000 ( 14%) ] Loss: 0.6534 top1= 75.1562
[E14B20 |  13440/50000 ( 27%) ] Loss: 0.5499 top1= 80.1562
[E14B30 |  19840/50000 ( 40%) ] Loss: 0.5807 top1= 77.9688
[E14B40 |  26240/50000 ( 52%) ] Loss: 0.5712 top1= 77.6562
[E14B50 |  32640/50000 ( 65%) ] Loss: 0.6154 top1= 77.0312
[E14B60 |  39040/50000 ( 78%) ] Loss: 0.6047 top1= 79.2188
[E14B70 |  45440/50000 ( 91%) ] Loss: 0.5195 top1= 81.5625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5144 top1= 10.1663


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2138 top1= 35.9175


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7059 top1= 41.5765

Train epoch 15
[E15B0  |    640/50000 (  1%) ] Loss: 0.6127 top1= 77.1875
[E15B10 |   7040/50000 ( 14%) ] Loss: 0.5814 top1= 79.3750
[E15B20 |  13440/50000 ( 27%) ] Loss: 0.5030 top1= 81.8750
[E15B30 |  19840/50000 ( 40%) ] Loss: 0.6001 top1= 77.1875
[E15B40 |  26240/50000 ( 52%) ] Loss: 0.5300 top1= 79.2188
[E15B50 |  32640/50000 ( 65%) ] Loss: 0.5305 top1= 80.4688
[E15B60 |  39040/50000 ( 78%) ] Loss: 0.5099 top1= 80.7812
[E15B70 |  45440/50000 ( 91%) ] Loss: 0.4824 top1= 81.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.5237 top1= 10.4067


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1688 top1= 35.9475


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6503 top1= 41.9772

Train epoch 16
[E16B0  |    640/50000 (  1%) ] Loss: 0.5846 top1= 77.6562
[E16B10 |   7040/50000 ( 14%) ] Loss: 0.6629 top1= 75.6250
[E16B20 |  13440/50000 ( 27%) ] Loss: 0.5124 top1= 80.0000
[E16B30 |  19840/50000 ( 40%) ] Loss: 0.5310 top1= 81.7188
[E16B40 |  26240/50000 ( 52%) ] Loss: 0.5626 top1= 77.9688
[E16B50 |  32640/50000 ( 65%) ] Loss: 0.6010 top1= 78.1250
[E16B60 |  39040/50000 ( 78%) ] Loss: 0.5770 top1= 78.1250
[E16B70 |  45440/50000 ( 91%) ] Loss: 0.5009 top1= 79.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4405 top1= 11.1979


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=4.9536 top1= 36.5084


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2500 top1= 42.6583

Train epoch 17
[E17B0  |    640/50000 (  1%) ] Loss: 0.5721 top1= 78.2812
[E17B10 |   7040/50000 ( 14%) ] Loss: 0.5527 top1= 80.3125
[E17B20 |  13440/50000 ( 27%) ] Loss: 0.5249 top1= 81.7188
[E17B30 |  19840/50000 ( 40%) ] Loss: 0.5285 top1= 80.7812
[E17B40 |  26240/50000 ( 52%) ] Loss: 0.4863 top1= 81.8750
[E17B50 |  32640/50000 ( 65%) ] Loss: 0.5024 top1= 82.1875
[E17B60 |  39040/50000 ( 78%) ] Loss: 0.5698 top1= 78.5938
[E17B70 |  45440/50000 ( 91%) ] Loss: 0.4932 top1= 83.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4678 top1= 10.6771


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2016 top1= 37.1595


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6552 top1= 41.4463

Train epoch 18
[E18B0  |    640/50000 (  1%) ] Loss: 0.5495 top1= 78.2812
[E18B10 |   7040/50000 ( 14%) ] Loss: 0.5596 top1= 79.8438
[E18B20 |  13440/50000 ( 27%) ] Loss: 0.4981 top1= 82.1875
[E18B30 |  19840/50000 ( 40%) ] Loss: 0.4897 top1= 81.8750
[E18B40 |  26240/50000 ( 52%) ] Loss: 0.5356 top1= 77.9688
[E18B50 |  32640/50000 ( 65%) ] Loss: 0.5087 top1= 79.5312
[E18B60 |  39040/50000 ( 78%) ] Loss: 0.5188 top1= 80.7812
[E18B70 |  45440/50000 ( 91%) ] Loss: 0.4219 top1= 84.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.4056 top1= 14.6334


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2042 top1= 37.6302


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7305 top1= 42.5581

Train epoch 19
[E19B0  |    640/50000 (  1%) ] Loss: 0.5009 top1= 81.2500
[E19B10 |   7040/50000 ( 14%) ] Loss: 0.5899 top1= 79.5312
[E19B20 |  13440/50000 ( 27%) ] Loss: 0.4609 top1= 81.7188
[E19B30 |  19840/50000 ( 40%) ] Loss: 0.4717 top1= 82.5000
[E19B40 |  26240/50000 ( 52%) ] Loss: 0.5060 top1= 80.0000
[E19B50 |  32640/50000 ( 65%) ] Loss: 0.4701 top1= 80.9375
[E19B60 |  39040/50000 ( 78%) ] Loss: 0.5367 top1= 80.9375
[E19B70 |  45440/50000 ( 91%) ] Loss: 0.4442 top1= 82.3438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.3447 top1= 16.1759


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4043 top1= 37.2496


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9132 top1= 43.5697

Train epoch 20
[E20B0  |    640/50000 (  1%) ] Loss: 0.5005 top1= 82.1875
[E20B10 |   7040/50000 ( 14%) ] Loss: 0.5136 top1= 80.7812
[E20B20 |  13440/50000 ( 27%) ] Loss: 0.3872 top1= 85.0000
[E20B30 |  19840/50000 ( 40%) ] Loss: 0.4404 top1= 82.8125
[E20B40 |  26240/50000 ( 52%) ] Loss: 0.5154 top1= 80.1562
[E20B50 |  32640/50000 ( 65%) ] Loss: 0.5021 top1= 80.4688
[E20B60 |  39040/50000 ( 78%) ] Loss: 0.4559 top1= 83.9062
[E20B70 |  45440/50000 ( 91%) ] Loss: 0.4392 top1= 84.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2390 top1= 20.1522


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.6012 top1= 38.4215


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.2977 top1= 43.8001

Train epoch 21
[E21B0  |    640/50000 (  1%) ] Loss: 0.4108 top1= 83.7500
[E21B10 |   7040/50000 ( 14%) ] Loss: 0.4739 top1= 82.5000
[E21B20 |  13440/50000 ( 27%) ] Loss: 0.4449 top1= 83.5938
[E21B30 |  19840/50000 ( 40%) ] Loss: 0.4005 top1= 85.1562
[E21B40 |  26240/50000 ( 52%) ] Loss: 0.4678 top1= 82.5000
[E21B50 |  32640/50000 ( 65%) ] Loss: 0.4438 top1= 82.8125
[E21B60 |  39040/50000 ( 78%) ] Loss: 0.4489 top1= 83.5938
[E21B70 |  45440/50000 ( 91%) ] Loss: 0.3839 top1= 84.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.2738 top1= 20.8834


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3702 top1= 39.5032


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4760 top1= 42.9587

Train epoch 22
[E22B0  |    640/50000 (  1%) ] Loss: 0.4369 top1= 82.0312
[E22B10 |   7040/50000 ( 14%) ] Loss: 0.4622 top1= 83.1250
[E22B20 |  13440/50000 ( 27%) ] Loss: 0.3735 top1= 86.0938
[E22B30 |  19840/50000 ( 40%) ] Loss: 0.4046 top1= 84.8438
[E22B40 |  26240/50000 ( 52%) ] Loss: 0.3749 top1= 85.0000
[E22B50 |  32640/50000 ( 65%) ] Loss: 0.4470 top1= 82.1875
[E22B60 |  39040/50000 ( 78%) ] Loss: 0.4999 top1= 81.5625
[E22B70 |  45440/50000 ( 91%) ] Loss: 0.4064 top1= 85.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0775 top1= 29.6274


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5250 top1= 37.4099


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.5978 top1= 42.4880

Train epoch 23
[E23B0  |    640/50000 (  1%) ] Loss: 0.5413 top1= 79.6875
[E23B10 |   7040/50000 ( 14%) ] Loss: 0.4377 top1= 84.8438
[E23B20 |  13440/50000 ( 27%) ] Loss: 0.3498 top1= 87.1875
[E23B30 |  19840/50000 ( 40%) ] Loss: 0.4023 top1= 85.4688
[E23B40 |  26240/50000 ( 52%) ] Loss: 0.4901 top1= 80.4688
[E23B50 |  32640/50000 ( 65%) ] Loss: 0.4193 top1= 83.4375
[E23B60 |  39040/50000 ( 78%) ] Loss: 0.4318 top1= 83.7500
[E23B70 |  45440/50000 ( 91%) ] Loss: 0.3684 top1= 85.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=2.0503 top1= 29.6775


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8360 top1= 39.7035


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1231 top1= 43.7901

Train epoch 24
[E24B0  |    640/50000 (  1%) ] Loss: 0.4416 top1= 83.1250
[E24B10 |   7040/50000 ( 14%) ] Loss: 0.4542 top1= 82.8125
[E24B20 |  13440/50000 ( 27%) ] Loss: 0.3673 top1= 87.0312
[E24B30 |  19840/50000 ( 40%) ] Loss: 0.4129 top1= 85.4688
[E24B40 |  26240/50000 ( 52%) ] Loss: 0.3903 top1= 86.2500
[E24B50 |  32640/50000 ( 65%) ] Loss: 0.4192 top1= 83.5938
[E24B60 |  39040/50000 ( 78%) ] Loss: 0.4297 top1= 83.5938
[E24B70 |  45440/50000 ( 91%) ] Loss: 0.3685 top1= 86.7188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9450 top1= 34.1046


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4840 top1= 38.7520


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3329 top1= 44.2208

Train epoch 25
[E25B0  |    640/50000 (  1%) ] Loss: 0.3953 top1= 84.2188
[E25B10 |   7040/50000 ( 14%) ] Loss: 0.4145 top1= 85.1562
[E25B20 |  13440/50000 ( 27%) ] Loss: 0.3324 top1= 87.0312
[E25B30 |  19840/50000 ( 40%) ] Loss: 0.3676 top1= 85.0000
[E25B40 |  26240/50000 ( 52%) ] Loss: 0.3774 top1= 85.9375
[E25B50 |  32640/50000 ( 65%) ] Loss: 0.4000 top1= 84.8438
[E25B60 |  39040/50000 ( 78%) ] Loss: 0.3908 top1= 86.5625
[E25B70 |  45440/50000 ( 91%) ] Loss: 0.3473 top1= 85.7812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.9454 top1= 35.0361


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9159 top1= 39.9539


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6036 top1= 44.3409

Train epoch 26
[E26B0  |    640/50000 (  1%) ] Loss: 0.3954 top1= 84.8438
[E26B10 |   7040/50000 ( 14%) ] Loss: 0.3838 top1= 86.0938
[E26B20 |  13440/50000 ( 27%) ] Loss: 0.3312 top1= 87.8125
[E26B30 |  19840/50000 ( 40%) ] Loss: 0.3753 top1= 87.0312
[E26B40 |  26240/50000 ( 52%) ] Loss: 0.3850 top1= 86.4062
[E26B50 |  32640/50000 ( 65%) ] Loss: 0.3383 top1= 88.1250
[E26B60 |  39040/50000 ( 78%) ] Loss: 0.3977 top1= 85.0000
[E26B70 |  45440/50000 ( 91%) ] Loss: 0.3481 top1= 86.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8775 top1= 38.0108


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.7872 top1= 40.6450


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4629 top1= 44.3309

Train epoch 27
[E27B0  |    640/50000 (  1%) ] Loss: 0.3895 top1= 84.8438
[E27B10 |   7040/50000 ( 14%) ] Loss: 0.3880 top1= 86.5625
[E27B20 |  13440/50000 ( 27%) ] Loss: 0.3177 top1= 87.3438
[E27B30 |  19840/50000 ( 40%) ] Loss: 0.3343 top1= 86.5625
[E27B40 |  26240/50000 ( 52%) ] Loss: 0.4025 top1= 85.1562
[E27B50 |  32640/50000 ( 65%) ] Loss: 0.3633 top1= 85.4688
[E27B60 |  39040/50000 ( 78%) ] Loss: 0.3700 top1= 87.0312
[E27B70 |  45440/50000 ( 91%) ] Loss: 0.2990 top1= 89.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.8214 top1= 40.7151


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9067 top1= 39.4431


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.4785 top1= 43.7099

Train epoch 28
[E28B0  |    640/50000 (  1%) ] Loss: 0.3835 top1= 86.2500
[E28B10 |   7040/50000 ( 14%) ] Loss: 0.3894 top1= 85.6250
[E28B20 |  13440/50000 ( 27%) ] Loss: 0.3015 top1= 88.2812
[E28B30 |  19840/50000 ( 40%) ] Loss: 0.3395 top1= 88.1250
[E28B40 |  26240/50000 ( 52%) ] Loss: 0.3191 top1= 88.2812
[E28B50 |  32640/50000 ( 65%) ] Loss: 0.3799 top1= 84.0625
[E28B60 |  39040/50000 ( 78%) ] Loss: 0.3652 top1= 87.8125
[E28B70 |  45440/50000 ( 91%) ] Loss: 0.2898 top1= 89.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7421 top1= 45.9435


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0619 top1= 39.0024


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6307 top1= 44.5312

Train epoch 29
[E29B0  |    640/50000 (  1%) ] Loss: 0.4154 top1= 84.0625
[E29B10 |   7040/50000 ( 14%) ] Loss: 0.3832 top1= 86.2500
[E29B20 |  13440/50000 ( 27%) ] Loss: 0.3283 top1= 87.5000
[E29B30 |  19840/50000 ( 40%) ] Loss: 0.3175 top1= 88.5938
[E29B40 |  26240/50000 ( 52%) ] Loss: 0.3267 top1= 87.5000
[E29B50 |  32640/50000 ( 65%) ] Loss: 0.4300 top1= 85.1562
[E29B60 |  39040/50000 ( 78%) ] Loss: 0.3418 top1= 88.2812
[E29B70 |  45440/50000 ( 91%) ] Loss: 0.3212 top1= 87.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6448 top1= 50.0901


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0230 top1= 40.1542


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.7104 top1= 45.0321

Train epoch 30
[E30B0  |    640/50000 (  1%) ] Loss: 0.3290 top1= 87.0312
[E30B10 |   7040/50000 ( 14%) ] Loss: 0.3646 top1= 85.9375
[E30B20 |  13440/50000 ( 27%) ] Loss: 0.2717 top1= 89.8438
[E30B30 |  19840/50000 ( 40%) ] Loss: 0.2904 top1= 88.5938
[E30B40 |  26240/50000 ( 52%) ] Loss: 0.3850 top1= 85.3125
[E30B50 |  32640/50000 ( 65%) ] Loss: 0.3056 top1= 87.6562
[E30B60 |  39040/50000 ( 78%) ] Loss: 0.3034 top1= 89.2188
[E30B70 |  45440/50000 ( 91%) ] Loss: 0.2298 top1= 91.5625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6118 top1= 51.3221


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7199 top1= 41.0657


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0054 top1= 44.6815

Train epoch 31
[E31B0  |    640/50000 (  1%) ] Loss: 0.3597 top1= 86.2500
[E31B10 |   7040/50000 ( 14%) ] Loss: 0.3671 top1= 88.2812
[E31B20 |  13440/50000 ( 27%) ] Loss: 0.2874 top1= 88.9062
[E31B30 |  19840/50000 ( 40%) ] Loss: 0.2958 top1= 88.7500
[E31B40 |  26240/50000 ( 52%) ] Loss: 0.2792 top1= 89.2188
[E31B50 |  32640/50000 ( 65%) ] Loss: 0.3215 top1= 87.6562
[E31B60 |  39040/50000 ( 78%) ] Loss: 0.3056 top1= 88.9062
[E31B70 |  45440/50000 ( 91%) ] Loss: 0.3324 top1= 85.7812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5571 top1= 53.0849


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8578 top1= 40.8554


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.1363 top1= 45.1522

Train epoch 32
[E32B0  |    640/50000 (  1%) ] Loss: 0.3313 top1= 86.5625
[E32B10 |   7040/50000 ( 14%) ] Loss: 0.3047 top1= 87.9688
[E32B20 |  13440/50000 ( 27%) ] Loss: 0.2847 top1= 88.9062
[E32B30 |  19840/50000 ( 40%) ] Loss: 0.3360 top1= 87.8125
[E32B40 |  26240/50000 ( 52%) ] Loss: 0.3245 top1= 86.7188
[E32B50 |  32640/50000 ( 65%) ] Loss: 0.3393 top1= 86.4062
[E32B60 |  39040/50000 ( 78%) ] Loss: 0.3288 top1= 88.1250
[E32B70 |  45440/50000 ( 91%) ] Loss: 0.2965 top1= 89.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4046 top1= 58.2232


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0442 top1= 41.1659


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0957 top1= 44.8017

Train epoch 33
[E33B0  |    640/50000 (  1%) ] Loss: 0.3179 top1= 87.6562
[E33B10 |   7040/50000 ( 14%) ] Loss: 0.3254 top1= 87.9688
[E33B20 |  13440/50000 ( 27%) ] Loss: 0.2870 top1= 88.7500
[E33B30 |  19840/50000 ( 40%) ] Loss: 0.3434 top1= 87.6562
[E33B40 |  26240/50000 ( 52%) ] Loss: 0.3230 top1= 88.9062
[E33B50 |  32640/50000 ( 65%) ] Loss: 0.2934 top1= 89.0625
[E33B60 |  39040/50000 ( 78%) ] Loss: 0.3001 top1= 88.1250
[E33B70 |  45440/50000 ( 91%) ] Loss: 0.2960 top1= 91.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4431 top1= 56.9010


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.0423 top1= 40.9555


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8720 top1= 44.5913

Train epoch 34
[E34B0  |    640/50000 (  1%) ] Loss: 0.3135 top1= 88.5938
[E34B10 |   7040/50000 ( 14%) ] Loss: 0.3134 top1= 88.9062
[E34B20 |  13440/50000 ( 27%) ] Loss: 0.2829 top1= 89.5312
[E34B30 |  19840/50000 ( 40%) ] Loss: 0.3074 top1= 88.1250
[E34B40 |  26240/50000 ( 52%) ] Loss: 0.2769 top1= 89.6875
[E34B50 |  32640/50000 ( 65%) ] Loss: 0.3558 top1= 86.2500
[E34B60 |  39040/50000 ( 78%) ] Loss: 0.2679 top1= 89.6875
[E34B70 |  45440/50000 ( 91%) ] Loss: 0.2430 top1= 91.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4439 top1= 57.2616


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.3297 top1= 41.2760


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.0088 top1= 44.6915

Train epoch 35
[E35B0  |    640/50000 (  1%) ] Loss: 0.3016 top1= 87.3438
[E35B10 |   7040/50000 ( 14%) ] Loss: 0.2937 top1= 89.2188
[E35B20 |  13440/50000 ( 27%) ] Loss: 0.2439 top1= 91.2500
[E35B30 |  19840/50000 ( 40%) ] Loss: 0.3008 top1= 88.4375
[E35B40 |  26240/50000 ( 52%) ] Loss: 0.2676 top1= 89.5312
[E35B50 |  32640/50000 ( 65%) ] Loss: 0.3429 top1= 87.6562
[E35B60 |  39040/50000 ( 78%) ] Loss: 0.2701 top1= 89.2188
[E35B70 |  45440/50000 ( 91%) ] Loss: 0.2437 top1= 90.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2996 top1= 62.2296


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5833 top1= 40.2644


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4850 top1= 45.0220

Train epoch 36
[E36B0  |    640/50000 (  1%) ] Loss: 0.3542 top1= 87.1875
[E36B10 |   7040/50000 ( 14%) ] Loss: 0.2856 top1= 89.8438
[E36B20 |  13440/50000 ( 27%) ] Loss: 0.2561 top1= 91.2500
[E36B30 |  19840/50000 ( 40%) ] Loss: 0.2961 top1= 89.2188
[E36B40 |  26240/50000 ( 52%) ] Loss: 0.2953 top1= 89.5312
[E36B50 |  32640/50000 ( 65%) ] Loss: 0.3245 top1= 87.8125
[E36B60 |  39040/50000 ( 78%) ] Loss: 0.3131 top1= 89.6875
[E36B70 |  45440/50000 ( 91%) ] Loss: 0.3157 top1= 88.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2890 top1= 62.5801


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1309 top1= 41.1859


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2003 top1= 45.2224

Train epoch 37
[E37B0  |    640/50000 (  1%) ] Loss: 0.2902 top1= 89.6875
[E37B10 |   7040/50000 ( 14%) ] Loss: 0.2826 top1= 89.0625
[E37B20 |  13440/50000 ( 27%) ] Loss: 0.2562 top1= 90.3125
[E37B30 |  19840/50000 ( 40%) ] Loss: 0.2722 top1= 90.7812
[E37B40 |  26240/50000 ( 52%) ] Loss: 0.2790 top1= 89.2188
[E37B50 |  32640/50000 ( 65%) ] Loss: 0.2463 top1= 90.6250
[E37B60 |  39040/50000 ( 78%) ] Loss: 0.2677 top1= 91.4062
[E37B70 |  45440/50000 ( 91%) ] Loss: 0.2540 top1= 91.0938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2849 top1= 62.7304


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9444 top1= 41.7368


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.1846 top1= 45.1522

Train epoch 38
[E38B0  |    640/50000 (  1%) ] Loss: 0.2521 top1= 89.6875
[E38B10 |   7040/50000 ( 14%) ] Loss: 0.2534 top1= 90.9375
[E38B20 |  13440/50000 ( 27%) ] Loss: 0.2381 top1= 91.4062
[E38B30 |  19840/50000 ( 40%) ] Loss: 0.2881 top1= 89.2188
[E38B40 |  26240/50000 ( 52%) ] Loss: 0.3207 top1= 90.3125
[E38B50 |  32640/50000 ( 65%) ] Loss: 0.2700 top1= 89.8438
[E38B60 |  39040/50000 ( 78%) ] Loss: 0.2326 top1= 91.4062
[E38B70 |  45440/50000 ( 91%) ] Loss: 0.2310 top1= 90.4688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3176 top1= 60.8974


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.9510 top1= 40.8353


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1828 top1= 45.4627

Train epoch 39
[E39B0  |    640/50000 (  1%) ] Loss: 0.3098 top1= 88.5938
[E39B10 |   7040/50000 ( 14%) ] Loss: 0.2884 top1= 88.9062
[E39B20 |  13440/50000 ( 27%) ] Loss: 0.2201 top1= 92.3438
[E39B30 |  19840/50000 ( 40%) ] Loss: 0.2611 top1= 91.2500
[E39B40 |  26240/50000 ( 52%) ] Loss: 0.2970 top1= 90.6250
[E39B50 |  32640/50000 ( 65%) ] Loss: 0.2698 top1= 89.8438
[E39B60 |  39040/50000 ( 78%) ] Loss: 0.2623 top1= 90.9375
[E39B70 |  45440/50000 ( 91%) ] Loss: 0.2012 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2934 top1= 61.6887


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7854 top1= 41.3862


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8459 top1= 45.1222

Train epoch 40
[E40B0  |    640/50000 (  1%) ] Loss: 0.2650 top1= 90.0000
[E40B10 |   7040/50000 ( 14%) ] Loss: 0.2612 top1= 89.2188
[E40B20 |  13440/50000 ( 27%) ] Loss: 0.2170 top1= 92.5000
[E40B30 |  19840/50000 ( 40%) ] Loss: 0.2803 top1= 90.7812
[E40B40 |  26240/50000 ( 52%) ] Loss: 0.2226 top1= 91.5625
[E40B50 |  32640/50000 ( 65%) ] Loss: 0.2760 top1= 90.1562
[E40B60 |  39040/50000 ( 78%) ] Loss: 0.2237 top1= 92.6562
[E40B70 |  45440/50000 ( 91%) ] Loss: 0.2551 top1= 90.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2135 top1= 64.1927


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7225 top1= 41.5665


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8379 top1= 45.3626

Train epoch 41
[E41B0  |    640/50000 (  1%) ] Loss: 0.2775 top1= 89.5312
[E41B10 |   7040/50000 ( 14%) ] Loss: 0.2624 top1= 90.3125
[E41B20 |  13440/50000 ( 27%) ] Loss: 0.2027 top1= 92.6562
[E41B30 |  19840/50000 ( 40%) ] Loss: 0.2176 top1= 92.9688
[E41B40 |  26240/50000 ( 52%) ] Loss: 0.2191 top1= 91.7188
[E41B50 |  32640/50000 ( 65%) ] Loss: 0.2254 top1= 91.0938
[E41B60 |  39040/50000 ( 78%) ] Loss: 0.2287 top1= 90.7812
[E41B70 |  45440/50000 ( 91%) ] Loss: 0.2122 top1= 92.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2404 top1= 63.8221


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8294 top1= 41.9772


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.9317 top1= 45.8634

Train epoch 42
[E42B0  |    640/50000 (  1%) ] Loss: 0.2362 top1= 91.2500
[E42B10 |   7040/50000 ( 14%) ] Loss: 0.2400 top1= 90.4688
[E42B20 |  13440/50000 ( 27%) ] Loss: 0.2243 top1= 91.0938
[E42B30 |  19840/50000 ( 40%) ] Loss: 0.2882 top1= 90.1562
[E42B40 |  26240/50000 ( 52%) ] Loss: 0.2571 top1= 90.4688
[E42B50 |  32640/50000 ( 65%) ] Loss: 0.2533 top1= 90.6250
[E42B60 |  39040/50000 ( 78%) ] Loss: 0.2767 top1= 90.3125
[E42B70 |  45440/50000 ( 91%) ] Loss: 0.2530 top1= 91.4062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2451 top1= 63.4415


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7448 top1= 41.2560


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8698 top1= 44.7616

Train epoch 43
[E43B0  |    640/50000 (  1%) ] Loss: 0.3029 top1= 89.8438
[E43B10 |   7040/50000 ( 14%) ] Loss: 0.3042 top1= 89.2188
[E43B20 |  13440/50000 ( 27%) ] Loss: 0.2494 top1= 90.4688
[E43B30 |  19840/50000 ( 40%) ] Loss: 0.2630 top1= 90.6250
[E43B40 |  26240/50000 ( 52%) ] Loss: 0.2195 top1= 92.8125
[E43B50 |  32640/50000 ( 65%) ] Loss: 0.2444 top1= 90.1562
[E43B60 |  39040/50000 ( 78%) ] Loss: 0.2225 top1= 92.6562
[E43B70 |  45440/50000 ( 91%) ] Loss: 0.2506 top1= 91.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2342 top1= 64.4231


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2973 top1= 41.5264


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0501 top1= 45.5929

Train epoch 44
[E44B0  |    640/50000 (  1%) ] Loss: 0.2294 top1= 91.2500
[E44B10 |   7040/50000 ( 14%) ] Loss: 0.2430 top1= 91.5625
[E44B20 |  13440/50000 ( 27%) ] Loss: 0.2712 top1= 90.6250
[E44B30 |  19840/50000 ( 40%) ] Loss: 0.2597 top1= 90.6250
[E44B40 |  26240/50000 ( 52%) ] Loss: 0.2246 top1= 93.2812
[E44B50 |  32640/50000 ( 65%) ] Loss: 0.2446 top1= 90.7812
[E44B60 |  39040/50000 ( 78%) ] Loss: 0.2326 top1= 90.7812
[E44B70 |  45440/50000 ( 91%) ] Loss: 0.2299 top1= 91.2500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2576 top1= 65.0140


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.0492 top1= 42.3177


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7977 top1= 45.7332

Train epoch 45
[E45B0  |    640/50000 (  1%) ] Loss: 0.1950 top1= 92.6562
[E45B10 |   7040/50000 ( 14%) ] Loss: 0.2560 top1= 90.7812
[E45B20 |  13440/50000 ( 27%) ] Loss: 0.1911 top1= 92.5000
[E45B30 |  19840/50000 ( 40%) ] Loss: 0.2336 top1= 92.1875
[E45B40 |  26240/50000 ( 52%) ] Loss: 0.2103 top1= 92.3438
[E45B50 |  32640/50000 ( 65%) ] Loss: 0.2550 top1= 89.5312
[E45B60 |  39040/50000 ( 78%) ] Loss: 0.2194 top1= 91.7188
[E45B70 |  45440/50000 ( 91%) ] Loss: 0.2254 top1= 92.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2822 top1= 62.9708


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.5648 top1= 41.3862


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.9129 top1= 44.8518

Train epoch 46
[E46B0  |    640/50000 (  1%) ] Loss: 0.2781 top1= 89.5312
[E46B10 |   7040/50000 ( 14%) ] Loss: 0.2563 top1= 90.7812
[E46B20 |  13440/50000 ( 27%) ] Loss: 0.2494 top1= 92.1875
[E46B30 |  19840/50000 ( 40%) ] Loss: 0.2204 top1= 92.0312
[E46B40 |  26240/50000 ( 52%) ] Loss: 0.2382 top1= 91.0938
[E46B50 |  32640/50000 ( 65%) ] Loss: 0.2345 top1= 90.4688
[E46B60 |  39040/50000 ( 78%) ] Loss: 0.1795 top1= 93.9062
[E46B70 |  45440/50000 ( 91%) ] Loss: 0.2325 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2324 top1= 65.0741


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.2730 top1= 41.9571


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.1987 top1= 45.6731

Train epoch 47
[E47B0  |    640/50000 (  1%) ] Loss: 0.2234 top1= 92.0312
[E47B10 |   7040/50000 ( 14%) ] Loss: 0.1963 top1= 92.3438
[E47B20 |  13440/50000 ( 27%) ] Loss: 0.2046 top1= 92.3438
[E47B30 |  19840/50000 ( 40%) ] Loss: 0.2543 top1= 91.0938
[E47B40 |  26240/50000 ( 52%) ] Loss: 0.2306 top1= 92.0312
[E47B50 |  32640/50000 ( 65%) ] Loss: 0.2244 top1= 92.1875
[E47B60 |  39040/50000 ( 78%) ] Loss: 0.2419 top1= 91.2500
[E47B70 |  45440/50000 ( 91%) ] Loss: 0.2055 top1= 92.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2358 top1= 65.4046


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8031 top1= 41.7268


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.9877 top1= 45.5228

Train epoch 48
[E48B0  |    640/50000 (  1%) ] Loss: 0.2547 top1= 89.0625
[E48B10 |   7040/50000 ( 14%) ] Loss: 0.2555 top1= 91.7188
[E48B20 |  13440/50000 ( 27%) ] Loss: 0.2417 top1= 91.5625
[E48B30 |  19840/50000 ( 40%) ] Loss: 0.2137 top1= 91.7188
[E48B40 |  26240/50000 ( 52%) ] Loss: 0.2007 top1= 92.6562
[E48B50 |  32640/50000 ( 65%) ] Loss: 0.2161 top1= 92.5000
[E48B60 |  39040/50000 ( 78%) ] Loss: 0.2128 top1= 92.6562
[E48B70 |  45440/50000 ( 91%) ] Loss: 0.2098 top1= 92.1875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2292 top1= 66.1258


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.6610 top1= 42.3277


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.2096 top1= 45.1823

Train epoch 49
[E49B0  |    640/50000 (  1%) ] Loss: 0.2338 top1= 91.0938
[E49B10 |   7040/50000 ( 14%) ] Loss: 0.2011 top1= 92.3438
[E49B20 |  13440/50000 ( 27%) ] Loss: 0.2416 top1= 90.9375
[E49B30 |  19840/50000 ( 40%) ] Loss: 0.2383 top1= 90.9375
[E49B40 |  26240/50000 ( 52%) ] Loss: 0.2432 top1= 92.0312
[E49B50 |  32640/50000 ( 65%) ] Loss: 0.2041 top1= 92.9688
[E49B60 |  39040/50000 ( 78%) ] Loss: 0.2356 top1= 92.3438
[E49B70 |  45440/50000 ( 91%) ] Loss: 0.2569 top1= 89.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2235 top1= 65.8454


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8903 top1= 40.5549


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.8235 top1= 44.8518

Train epoch 50
[E50B0  |    640/50000 (  1%) ] Loss: 0.2964 top1= 89.6875
[E50B10 |   7040/50000 ( 14%) ] Loss: 0.2302 top1= 91.2500
[E50B20 |  13440/50000 ( 27%) ] Loss: 0.2329 top1= 92.1875
[E50B30 |  19840/50000 ( 40%) ] Loss: 0.2230 top1= 91.8750
[E50B40 |  26240/50000 ( 52%) ] Loss: 0.2186 top1= 92.0312
[E50B50 |  32640/50000 ( 65%) ] Loss: 0.2085 top1= 92.1875
[E50B60 |  39040/50000 ( 78%) ] Loss: 0.2125 top1= 93.1250
[E50B70 |  45440/50000 ( 91%) ] Loss: 0.2175 top1= 91.5625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2456 top1= 65.3446


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5047 top1= 42.4980


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4929 top1= 44.9519

Train epoch 51
[E51B0  |    640/50000 (  1%) ] Loss: 0.2517 top1= 91.7188
[E51B10 |   7040/50000 ( 14%) ] Loss: 0.2421 top1= 91.5625
[E51B20 |  13440/50000 ( 27%) ] Loss: 0.1997 top1= 92.5000
[E51B30 |  19840/50000 ( 40%) ] Loss: 0.2285 top1= 93.2812
[E51B40 |  26240/50000 ( 52%) ] Loss: 0.2172 top1= 92.5000
[E51B50 |  32640/50000 ( 65%) ] Loss: 0.2153 top1= 91.8750
[E51B60 |  39040/50000 ( 78%) ] Loss: 0.2042 top1= 92.8125
[E51B70 |  45440/50000 ( 91%) ] Loss: 0.2161 top1= 90.7812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1875 top1= 66.9772


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8763 top1= 42.3778


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3172 top1= 45.6330

Train epoch 52
[E52B0  |    640/50000 (  1%) ] Loss: 0.2179 top1= 91.4062
[E52B10 |   7040/50000 ( 14%) ] Loss: 0.2153 top1= 92.0312
[E52B20 |  13440/50000 ( 27%) ] Loss: 0.2032 top1= 93.1250
[E52B30 |  19840/50000 ( 40%) ] Loss: 0.2380 top1= 91.4062
[E52B40 |  26240/50000 ( 52%) ] Loss: 0.1588 top1= 95.4688
[E52B50 |  32640/50000 ( 65%) ] Loss: 0.2189 top1= 92.1875
[E52B60 |  39040/50000 ( 78%) ] Loss: 0.2433 top1= 91.8750
[E52B70 |  45440/50000 ( 91%) ] Loss: 0.2249 top1= 92.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2113 top1= 67.5180


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.3934 top1= 42.6482


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7118 top1= 45.1623

Train epoch 53
[E53B0  |    640/50000 (  1%) ] Loss: 0.2107 top1= 92.9688
[E53B10 |   7040/50000 ( 14%) ] Loss: 0.1999 top1= 93.9062
[E53B20 |  13440/50000 ( 27%) ] Loss: 0.2046 top1= 93.2812
[E53B30 |  19840/50000 ( 40%) ] Loss: 0.2659 top1= 90.7812
[E53B40 |  26240/50000 ( 52%) ] Loss: 0.1464 top1= 95.4688
[E53B50 |  32640/50000 ( 65%) ] Loss: 0.1904 top1= 92.9688
[E53B60 |  39040/50000 ( 78%) ] Loss: 0.1913 top1= 93.1250
[E53B70 |  45440/50000 ( 91%) ] Loss: 0.1955 top1= 92.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2217 top1= 66.8470


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.6884 top1= 41.4163


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4568 top1= 45.3626

Train epoch 54
[E54B0  |    640/50000 (  1%) ] Loss: 0.2042 top1= 92.1875
[E54B10 |   7040/50000 ( 14%) ] Loss: 0.2352 top1= 90.4688
[E54B20 |  13440/50000 ( 27%) ] Loss: 0.1976 top1= 93.7500
[E54B30 |  19840/50000 ( 40%) ] Loss: 0.2104 top1= 90.6250
[E54B40 |  26240/50000 ( 52%) ] Loss: 0.1930 top1= 92.5000
[E54B50 |  32640/50000 ( 65%) ] Loss: 0.2315 top1= 92.3438
[E54B60 |  39040/50000 ( 78%) ] Loss: 0.2000 top1= 92.6562
[E54B70 |  45440/50000 ( 91%) ] Loss: 0.2386 top1= 92.0312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2405 top1= 66.3161


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.7836 top1= 42.1975


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.3801 top1= 45.8033

Train epoch 55
[E55B0  |    640/50000 (  1%) ] Loss: 0.2345 top1= 91.5625
[E55B10 |   7040/50000 ( 14%) ] Loss: 0.2361 top1= 91.8750
[E55B20 |  13440/50000 ( 27%) ] Loss: 0.2405 top1= 91.0938
[E55B30 |  19840/50000 ( 40%) ] Loss: 0.2811 top1= 89.8438
[E55B40 |  26240/50000 ( 52%) ] Loss: 0.1953 top1= 93.2812
[E55B50 |  32640/50000 ( 65%) ] Loss: 0.2774 top1= 89.8438
[E55B60 |  39040/50000 ( 78%) ] Loss: 0.2321 top1= 91.5625
[E55B70 |  45440/50000 ( 91%) ] Loss: 0.1772 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1869 top1= 67.5080


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.1954 top1= 41.9371


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.5325 top1= 45.2123

Train epoch 56
[E56B0  |    640/50000 (  1%) ] Loss: 0.2128 top1= 91.8750
[E56B10 |   7040/50000 ( 14%) ] Loss: 0.2708 top1= 90.0000
[E56B20 |  13440/50000 ( 27%) ] Loss: 0.1801 top1= 93.4375
[E56B30 |  19840/50000 ( 40%) ] Loss: 0.2518 top1= 90.6250
[E56B40 |  26240/50000 ( 52%) ] Loss: 0.2566 top1= 90.9375
[E56B50 |  32640/50000 ( 65%) ] Loss: 0.2204 top1= 91.7188
[E56B60 |  39040/50000 ( 78%) ] Loss: 0.2340 top1= 92.8125
[E56B70 |  45440/50000 ( 91%) ] Loss: 0.1640 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2771 top1= 66.3962


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.5889 top1= 42.6583


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6879 top1= 44.9920

Train epoch 57
[E57B0  |    640/50000 (  1%) ] Loss: 0.2332 top1= 90.7812
[E57B10 |   7040/50000 ( 14%) ] Loss: 0.2168 top1= 92.1875
[E57B20 |  13440/50000 ( 27%) ] Loss: 0.1799 top1= 92.8125
[E57B30 |  19840/50000 ( 40%) ] Loss: 0.1952 top1= 92.1875
[E57B40 |  26240/50000 ( 52%) ] Loss: 0.1879 top1= 93.1250
[E57B50 |  32640/50000 ( 65%) ] Loss: 0.2426 top1= 90.7812
[E57B60 |  39040/50000 ( 78%) ] Loss: 0.1852 top1= 93.1250
[E57B70 |  45440/50000 ( 91%) ] Loss: 0.1976 top1= 93.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2025 top1= 68.0188


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.6206 top1= 42.7584


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.4422 top1= 45.7532

Train epoch 58
[E58B0  |    640/50000 (  1%) ] Loss: 0.1911 top1= 93.5938
[E58B10 |   7040/50000 ( 14%) ] Loss: 0.1960 top1= 93.2812
[E58B20 |  13440/50000 ( 27%) ] Loss: 0.1733 top1= 94.5312
[E58B30 |  19840/50000 ( 40%) ] Loss: 0.2140 top1= 92.5000
[E58B40 |  26240/50000 ( 52%) ] Loss: 0.2027 top1= 92.0312
[E58B50 |  32640/50000 ( 65%) ] Loss: 0.2302 top1= 91.4062
[E58B60 |  39040/50000 ( 78%) ] Loss: 0.2138 top1= 93.1250
[E58B70 |  45440/50000 ( 91%) ] Loss: 0.1852 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2327 top1= 67.6583


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5968 top1= 42.4579


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.0113 top1= 45.4327

Train epoch 59
[E59B0  |    640/50000 (  1%) ] Loss: 0.2178 top1= 92.3438
[E59B10 |   7040/50000 ( 14%) ] Loss: 0.1631 top1= 93.5938
[E59B20 |  13440/50000 ( 27%) ] Loss: 0.2013 top1= 92.5000
[E59B30 |  19840/50000 ( 40%) ] Loss: 0.2361 top1= 91.4062
[E59B40 |  26240/50000 ( 52%) ] Loss: 0.1906 top1= 93.5938
[E59B50 |  32640/50000 ( 65%) ] Loss: 0.2010 top1= 92.0312
[E59B60 |  39040/50000 ( 78%) ] Loss: 0.2087 top1= 92.5000
[E59B70 |  45440/50000 ( 91%) ] Loss: 0.1728 top1= 93.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1863 top1= 68.0589


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7828 top1= 42.4479


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7495 top1= 45.7232

Train epoch 60
[E60B0  |    640/50000 (  1%) ] Loss: 0.1826 top1= 93.1250
[E60B10 |   7040/50000 ( 14%) ] Loss: 0.1840 top1= 92.5000
[E60B20 |  13440/50000 ( 27%) ] Loss: 0.1781 top1= 93.4375
[E60B30 |  19840/50000 ( 40%) ] Loss: 0.2109 top1= 93.2812
[E60B40 |  26240/50000 ( 52%) ] Loss: 0.2069 top1= 93.2812
[E60B50 |  32640/50000 ( 65%) ] Loss: 0.2236 top1= 91.5625
[E60B60 |  39040/50000 ( 78%) ] Loss: 0.2063 top1= 92.8125
[E60B70 |  45440/50000 ( 91%) ] Loss: 0.1803 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2206 top1= 67.4679


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.1457 top1= 42.1875


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3352 top1= 44.9820

Train epoch 61
[E61B0  |    640/50000 (  1%) ] Loss: 0.2618 top1= 91.5625
[E61B10 |   7040/50000 ( 14%) ] Loss: 0.2194 top1= 91.8750
[E61B20 |  13440/50000 ( 27%) ] Loss: 0.1608 top1= 94.2188
[E61B30 |  19840/50000 ( 40%) ] Loss: 0.1716 top1= 92.3438
[E61B40 |  26240/50000 ( 52%) ] Loss: 0.1844 top1= 93.9062
[E61B50 |  32640/50000 ( 65%) ] Loss: 0.1770 top1= 93.7500
[E61B60 |  39040/50000 ( 78%) ] Loss: 0.1967 top1= 92.8125
[E61B70 |  45440/50000 ( 91%) ] Loss: 0.1857 top1= 92.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1779 top1= 68.5897


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6801 top1= 41.4964


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4226 top1= 45.0120

Train epoch 62
[E62B0  |    640/50000 (  1%) ] Loss: 0.2776 top1= 91.2500
[E62B10 |   7040/50000 ( 14%) ] Loss: 0.2142 top1= 92.6562
[E62B20 |  13440/50000 ( 27%) ] Loss: 0.1525 top1= 94.5312
[E62B30 |  19840/50000 ( 40%) ] Loss: 0.1808 top1= 93.4375
[E62B40 |  26240/50000 ( 52%) ] Loss: 0.2094 top1= 92.3438
[E62B50 |  32640/50000 ( 65%) ] Loss: 0.1984 top1= 93.5938
[E62B60 |  39040/50000 ( 78%) ] Loss: 0.1926 top1= 92.0312
[E62B70 |  45440/50000 ( 91%) ] Loss: 0.1501 top1= 94.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2193 top1= 68.4495


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2194 top1= 42.1975


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.4428 top1= 45.5228

Train epoch 63
[E63B0  |    640/50000 (  1%) ] Loss: 0.1645 top1= 93.9062
[E63B10 |   7040/50000 ( 14%) ] Loss: 0.2028 top1= 93.2812
[E63B20 |  13440/50000 ( 27%) ] Loss: 0.2147 top1= 93.2812
[E63B30 |  19840/50000 ( 40%) ] Loss: 0.2382 top1= 90.6250
[E63B40 |  26240/50000 ( 52%) ] Loss: 0.2024 top1= 92.3438
[E63B50 |  32640/50000 ( 65%) ] Loss: 0.1979 top1= 91.8750
[E63B60 |  39040/50000 ( 78%) ] Loss: 0.2058 top1= 93.4375
[E63B70 |  45440/50000 ( 91%) ] Loss: 0.1623 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1861 top1= 68.9804


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.9439 top1= 42.2977


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6276 top1= 45.4227

Train epoch 64
[E64B0  |    640/50000 (  1%) ] Loss: 0.1939 top1= 92.8125
[E64B10 |   7040/50000 ( 14%) ] Loss: 0.1899 top1= 93.9062
[E64B20 |  13440/50000 ( 27%) ] Loss: 0.1710 top1= 94.5312
[E64B30 |  19840/50000 ( 40%) ] Loss: 0.2028 top1= 92.6562
[E64B40 |  26240/50000 ( 52%) ] Loss: 0.2190 top1= 90.9375
[E64B50 |  32640/50000 ( 65%) ] Loss: 0.1863 top1= 92.1875
[E64B60 |  39040/50000 ( 78%) ] Loss: 0.1760 top1= 92.3438
[E64B70 |  45440/50000 ( 91%) ] Loss: 0.1966 top1= 92.3438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2502 top1= 68.9002


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.2073 top1= 41.5064


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3985 top1= 45.6631

Train epoch 65
[E65B0  |    640/50000 (  1%) ] Loss: 0.2192 top1= 91.2500
[E65B10 |   7040/50000 ( 14%) ] Loss: 0.1620 top1= 94.0625
[E65B20 |  13440/50000 ( 27%) ] Loss: 0.1871 top1= 92.6562
[E65B30 |  19840/50000 ( 40%) ] Loss: 0.1722 top1= 94.5312
[E65B40 |  26240/50000 ( 52%) ] Loss: 0.2135 top1= 92.3438
[E65B50 |  32640/50000 ( 65%) ] Loss: 0.2063 top1= 90.9375
[E65B60 |  39040/50000 ( 78%) ] Loss: 0.1858 top1= 94.5312
[E65B70 |  45440/50000 ( 91%) ] Loss: 0.1615 top1= 94.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2675 top1= 68.5597


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.6490 top1= 42.3177


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.5543 top1= 45.4728

Train epoch 66
[E66B0  |    640/50000 (  1%) ] Loss: 0.1881 top1= 93.9062
[E66B10 |   7040/50000 ( 14%) ] Loss: 0.1904 top1= 93.1250
[E66B20 |  13440/50000 ( 27%) ] Loss: 0.2110 top1= 90.7812
[E66B30 |  19840/50000 ( 40%) ] Loss: 0.1734 top1= 93.1250
[E66B40 |  26240/50000 ( 52%) ] Loss: 0.1466 top1= 95.0000
[E66B50 |  32640/50000 ( 65%) ] Loss: 0.1911 top1= 92.5000
[E66B60 |  39040/50000 ( 78%) ] Loss: 0.1483 top1= 94.3750
[E66B70 |  45440/50000 ( 91%) ] Loss: 0.1691 top1= 93.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2151 top1= 68.3093


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.5737 top1= 42.3377


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.7033 top1= 45.5128

Train epoch 67
[E67B0  |    640/50000 (  1%) ] Loss: 0.2233 top1= 91.7188
[E67B10 |   7040/50000 ( 14%) ] Loss: 0.1639 top1= 94.6875
[E67B20 |  13440/50000 ( 27%) ] Loss: 0.1897 top1= 92.6562
[E67B30 |  19840/50000 ( 40%) ] Loss: 0.2058 top1= 92.8125
[E67B40 |  26240/50000 ( 52%) ] Loss: 0.2001 top1= 94.5312
[E67B50 |  32640/50000 ( 65%) ] Loss: 0.1618 top1= 93.2812
[E67B60 |  39040/50000 ( 78%) ] Loss: 0.1870 top1= 93.5938
[E67B70 |  45440/50000 ( 91%) ] Loss: 0.1842 top1= 93.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1795 top1= 69.6014


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.5029 top1= 42.6282


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1424 top1= 45.9435

Train epoch 68
[E68B0  |    640/50000 (  1%) ] Loss: 0.1650 top1= 94.6875
[E68B10 |   7040/50000 ( 14%) ] Loss: 0.1902 top1= 94.2188
[E68B20 |  13440/50000 ( 27%) ] Loss: 0.1744 top1= 92.9688
[E68B30 |  19840/50000 ( 40%) ] Loss: 0.1629 top1= 94.5312
[E68B40 |  26240/50000 ( 52%) ] Loss: 0.1835 top1= 93.7500
[E68B50 |  32640/50000 ( 65%) ] Loss: 0.1632 top1= 94.8438
[E68B60 |  39040/50000 ( 78%) ] Loss: 0.1777 top1= 94.3750
[E68B70 |  45440/50000 ( 91%) ] Loss: 0.1687 top1= 93.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2371 top1= 68.6398


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.7393 top1= 41.6166


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0810 top1= 45.8433

Train epoch 69
[E69B0  |    640/50000 (  1%) ] Loss: 0.1761 top1= 93.1250
[E69B10 |   7040/50000 ( 14%) ] Loss: 0.2649 top1= 90.7812
[E69B20 |  13440/50000 ( 27%) ] Loss: 0.1524 top1= 94.6875
[E69B30 |  19840/50000 ( 40%) ] Loss: 0.1724 top1= 93.7500
[E69B40 |  26240/50000 ( 52%) ] Loss: 0.1423 top1= 94.8438
[E69B50 |  32640/50000 ( 65%) ] Loss: 0.1746 top1= 94.0625
[E69B60 |  39040/50000 ( 78%) ] Loss: 0.1928 top1= 93.7500
[E69B70 |  45440/50000 ( 91%) ] Loss: 0.1826 top1= 93.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2761 top1= 68.4495


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.6495 top1= 42.1675


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6359 top1= 45.6731

Train epoch 70
[E70B0  |    640/50000 (  1%) ] Loss: 0.2314 top1= 91.8750
[E70B10 |   7040/50000 ( 14%) ] Loss: 0.2426 top1= 92.3438
[E70B20 |  13440/50000 ( 27%) ] Loss: 0.1479 top1= 95.7812
[E70B30 |  19840/50000 ( 40%) ] Loss: 0.1931 top1= 94.5312
[E70B40 |  26240/50000 ( 52%) ] Loss: 0.1762 top1= 93.4375
[E70B50 |  32640/50000 ( 65%) ] Loss: 0.1987 top1= 93.2812
[E70B60 |  39040/50000 ( 78%) ] Loss: 0.1595 top1= 93.5938
[E70B70 |  45440/50000 ( 91%) ] Loss: 0.1775 top1= 92.8125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1799 top1= 69.5813


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4468 top1= 42.4880


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.3785 top1= 45.3325

Train epoch 71
[E71B0  |    640/50000 (  1%) ] Loss: 0.1983 top1= 94.0625
[E71B10 |   7040/50000 ( 14%) ] Loss: 0.2063 top1= 91.8750
[E71B20 |  13440/50000 ( 27%) ] Loss: 0.2005 top1= 93.7500
[E71B30 |  19840/50000 ( 40%) ] Loss: 0.1991 top1= 92.3438
[E71B40 |  26240/50000 ( 52%) ] Loss: 0.1483 top1= 95.1562
[E71B50 |  32640/50000 ( 65%) ] Loss: 0.1498 top1= 93.4375
[E71B60 |  39040/50000 ( 78%) ] Loss: 0.2011 top1= 92.6562
[E71B70 |  45440/50000 ( 91%) ] Loss: 0.1452 top1= 95.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1371 top1= 70.2123


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.9105 top1= 42.2576


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.7402 top1= 45.7031

Train epoch 72
[E72B0  |    640/50000 (  1%) ] Loss: 0.1965 top1= 93.9062
[E72B10 |   7040/50000 ( 14%) ] Loss: 0.1365 top1= 94.5312
[E72B20 |  13440/50000 ( 27%) ] Loss: 0.1560 top1= 95.0000
[E72B30 |  19840/50000 ( 40%) ] Loss: 0.1726 top1= 92.8125
[E72B40 |  26240/50000 ( 52%) ] Loss: 0.1369 top1= 95.1562
[E72B50 |  32640/50000 ( 65%) ] Loss: 0.1887 top1= 92.3438
[E72B60 |  39040/50000 ( 78%) ] Loss: 0.1686 top1= 94.0625
[E72B70 |  45440/50000 ( 91%) ] Loss: 0.1399 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1719 top1= 69.6214


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.8696 top1= 42.2977


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.0566 top1= 45.6430

Train epoch 73
[E73B0  |    640/50000 (  1%) ] Loss: 0.1851 top1= 92.9688
[E73B10 |   7040/50000 ( 14%) ] Loss: 0.1483 top1= 94.5312
[E73B20 |  13440/50000 ( 27%) ] Loss: 0.1757 top1= 93.5938
[E73B30 |  19840/50000 ( 40%) ] Loss: 0.1791 top1= 94.2188
[E73B40 |  26240/50000 ( 52%) ] Loss: 0.1446 top1= 94.5312
[E73B50 |  32640/50000 ( 65%) ] Loss: 0.1915 top1= 92.6562
[E73B60 |  39040/50000 ( 78%) ] Loss: 0.1940 top1= 92.9688
[E73B70 |  45440/50000 ( 91%) ] Loss: 0.1780 top1= 92.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2114 top1= 69.3109


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.1040 top1= 42.0974


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.5664 top1= 45.7432

Train epoch 74
[E74B0  |    640/50000 (  1%) ] Loss: 0.1993 top1= 93.1250
[E74B10 |   7040/50000 ( 14%) ] Loss: 0.1888 top1= 93.1250
[E74B20 |  13440/50000 ( 27%) ] Loss: 0.1850 top1= 93.9062
[E74B30 |  19840/50000 ( 40%) ] Loss: 0.1889 top1= 93.9062
[E74B40 |  26240/50000 ( 52%) ] Loss: 0.1313 top1= 95.3125
[E74B50 |  32640/50000 ( 65%) ] Loss: 0.2248 top1= 92.1875
[E74B60 |  39040/50000 ( 78%) ] Loss: 0.1925 top1= 93.7500
[E74B70 |  45440/50000 ( 91%) ] Loss: 0.1748 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1913 top1= 69.2007


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.3899 top1= 42.1174


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.1582 top1= 45.9635

Train epoch 75
[E75B0  |    640/50000 (  1%) ] Loss: 0.2067 top1= 93.7500
[E75B10 |   7040/50000 ( 14%) ] Loss: 0.1310 top1= 95.3125
[E75B20 |  13440/50000 ( 27%) ] Loss: 0.1223 top1= 95.0000
[E75B30 |  19840/50000 ( 40%) ] Loss: 0.1943 top1= 92.9688
[E75B40 |  26240/50000 ( 52%) ] Loss: 0.1489 top1= 94.8438
[E75B50 |  32640/50000 ( 65%) ] Loss: 0.2027 top1= 93.4375
[E75B60 |  39040/50000 ( 78%) ] Loss: 0.1951 top1= 93.4375
[E75B70 |  45440/50000 ( 91%) ] Loss: 0.1803 top1= 94.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1674 top1= 70.2624


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.7398 top1= 43.0489


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.5861 top1= 45.2825

Train epoch 76
[E76B0  |    640/50000 (  1%) ] Loss: 0.1837 top1= 93.7500
[E76B10 |   7040/50000 ( 14%) ] Loss: 0.1549 top1= 93.4375
[E76B20 |  13440/50000 ( 27%) ] Loss: 0.1573 top1= 94.0625
[E76B30 |  19840/50000 ( 40%) ] Loss: 0.1822 top1= 93.7500
[E76B40 |  26240/50000 ( 52%) ] Loss: 0.2158 top1= 93.2812
[E76B50 |  32640/50000 ( 65%) ] Loss: 0.1606 top1= 94.8438
[E76B60 |  39040/50000 ( 78%) ] Loss: 0.1200 top1= 95.7812
[E76B70 |  45440/50000 ( 91%) ] Loss: 0.1307 top1= 94.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2088 top1= 70.3025


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.4588 top1= 42.2977


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.1459 top1= 45.5128

Train epoch 77
[E77B0  |    640/50000 (  1%) ] Loss: 0.1968 top1= 92.6562
[E77B10 |   7040/50000 ( 14%) ] Loss: 0.1529 top1= 95.3125
[E77B20 |  13440/50000 ( 27%) ] Loss: 0.1590 top1= 93.5938
[E77B30 |  19840/50000 ( 40%) ] Loss: 0.1502 top1= 94.8438
[E77B40 |  26240/50000 ( 52%) ] Loss: 0.1497 top1= 95.1562
[E77B50 |  32640/50000 ( 65%) ] Loss: 0.1860 top1= 92.9688
[E77B60 |  39040/50000 ( 78%) ] Loss: 0.1594 top1= 93.9062
[E77B70 |  45440/50000 ( 91%) ] Loss: 0.1586 top1= 94.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2985 top1= 69.2308


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.3771 top1= 41.9271


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=8.3872 top1= 45.7031

Train epoch 78
[E78B0  |    640/50000 (  1%) ] Loss: 0.1628 top1= 94.3750
[E78B10 |   7040/50000 ( 14%) ] Loss: 0.1736 top1= 94.2188
[E78B20 |  13440/50000 ( 27%) ] Loss: 0.1503 top1= 94.2188
[E78B30 |  19840/50000 ( 40%) ] Loss: 0.1427 top1= 94.2188
[E78B40 |  26240/50000 ( 52%) ] Loss: 0.1811 top1= 93.7500
[E78B50 |  32640/50000 ( 65%) ] Loss: 0.2011 top1= 92.3438
[E78B60 |  39040/50000 ( 78%) ] Loss: 0.1694 top1= 95.6250
[E78B70 |  45440/50000 ( 91%) ] Loss: 0.1405 top1= 95.6250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1456 top1= 70.5729


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.1239 top1= 42.5080


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.5564 top1= 46.0136

Train epoch 79
[E79B0  |    640/50000 (  1%) ] Loss: 0.1812 top1= 93.2812
[E79B10 |   7040/50000 ( 14%) ] Loss: 0.1534 top1= 94.6875
[E79B20 |  13440/50000 ( 27%) ] Loss: 0.1221 top1= 94.8438
[E79B30 |  19840/50000 ( 40%) ] Loss: 0.1616 top1= 92.6562
[E79B40 |  26240/50000 ( 52%) ] Loss: 0.1492 top1= 95.3125
[E79B50 |  32640/50000 ( 65%) ] Loss: 0.2103 top1= 93.4375
[E79B60 |  39040/50000 ( 78%) ] Loss: 0.1993 top1= 93.4375
[E79B70 |  45440/50000 ( 91%) ] Loss: 0.1458 top1= 94.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2176 top1= 70.4227


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.0895 top1= 42.6983


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.8042 top1= 45.9435

Train epoch 80
[E80B0  |    640/50000 (  1%) ] Loss: 0.1898 top1= 95.0000
[E80B10 |   7040/50000 ( 14%) ] Loss: 0.1448 top1= 94.3750
[E80B20 |  13440/50000 ( 27%) ] Loss: 0.1160 top1= 96.0938
[E80B30 |  19840/50000 ( 40%) ] Loss: 0.1542 top1= 94.8438
[E80B40 |  26240/50000 ( 52%) ] Loss: 0.1049 top1= 96.7188
[E80B50 |  32640/50000 ( 65%) ] Loss: 0.2169 top1= 93.1250
[E80B60 |  39040/50000 ( 78%) ] Loss: 0.1358 top1= 95.7812
[E80B70 |  45440/50000 ( 91%) ] Loss: 0.1629 top1= 95.3125

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2515 top1= 68.9103


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.4170 top1= 42.2376


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.7259 top1= 45.9836

Train epoch 81
[E81B0  |    640/50000 (  1%) ] Loss: 0.1885 top1= 92.5000
[E81B10 |   7040/50000 ( 14%) ] Loss: 0.1595 top1= 93.4375
[E81B20 |  13440/50000 ( 27%) ] Loss: 0.1281 top1= 95.4688
[E81B30 |  19840/50000 ( 40%) ] Loss: 0.1109 top1= 95.7812
[E81B40 |  26240/50000 ( 52%) ] Loss: 0.0737 top1= 97.1875
[E81B50 |  32640/50000 ( 65%) ] Loss: 0.0676 top1= 97.8125
[E81B60 |  39040/50000 ( 78%) ] Loss: 0.0836 top1= 97.1875
[E81B70 |  45440/50000 ( 91%) ] Loss: 0.0593 top1= 97.5000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1806 top1= 71.1839


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=8.7054 top1= 44.2608


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.7009 top1= 46.5946

Train epoch 82
[E82B0  |    640/50000 (  1%) ] Loss: 0.0954 top1= 96.8750
[E82B10 |   7040/50000 ( 14%) ] Loss: 0.0718 top1= 97.0312
[E82B20 |  13440/50000 ( 27%) ] Loss: 0.0818 top1= 97.6562
[E82B30 |  19840/50000 ( 40%) ] Loss: 0.1060 top1= 95.9375
[E82B40 |  26240/50000 ( 52%) ] Loss: 0.0752 top1= 97.3438
[E82B50 |  32640/50000 ( 65%) ] Loss: 0.0699 top1= 97.5000
[E82B60 |  39040/50000 ( 78%) ] Loss: 0.0531 top1= 98.1250
[E82B70 |  45440/50000 ( 91%) ] Loss: 0.0521 top1= 97.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1532 top1= 71.8049


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.1182 top1= 44.3710


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.3173 top1= 46.6647

Train epoch 83
[E83B0  |    640/50000 (  1%) ] Loss: 0.0858 top1= 96.7188
[E83B10 |   7040/50000 ( 14%) ] Loss: 0.0751 top1= 97.3438
[E83B20 |  13440/50000 ( 27%) ] Loss: 0.0725 top1= 97.8125
[E83B30 |  19840/50000 ( 40%) ] Loss: 0.0920 top1= 96.5625
[E83B40 |  26240/50000 ( 52%) ] Loss: 0.0409 top1= 98.9062
[E83B50 |  32640/50000 ( 65%) ] Loss: 0.0563 top1= 97.8125
[E83B60 |  39040/50000 ( 78%) ] Loss: 0.0450 top1= 98.2812
[E83B70 |  45440/50000 ( 91%) ] Loss: 0.0674 top1= 97.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1824 top1= 71.6146


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.8378 top1= 44.1306


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.8269 top1= 46.6046

Train epoch 84
[E84B0  |    640/50000 (  1%) ] Loss: 0.0669 top1= 97.6562
[E84B10 |   7040/50000 ( 14%) ] Loss: 0.0612 top1= 98.2812
[E84B20 |  13440/50000 ( 27%) ] Loss: 0.0456 top1= 98.4375
[E84B30 |  19840/50000 ( 40%) ] Loss: 0.0601 top1= 97.9688
[E84B40 |  26240/50000 ( 52%) ] Loss: 0.0394 top1= 98.7500
[E84B50 |  32640/50000 ( 65%) ] Loss: 0.0524 top1= 98.4375
[E84B60 |  39040/50000 ( 78%) ] Loss: 0.0457 top1= 98.1250
[E84B70 |  45440/50000 ( 91%) ] Loss: 0.0204 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.1788 top1= 71.8650


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.9570 top1= 44.3109


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.0925 top1= 46.5946

Train epoch 85
[E85B0  |    640/50000 (  1%) ] Loss: 0.0388 top1= 99.2188
[E85B10 |   7040/50000 ( 14%) ] Loss: 0.0641 top1= 98.1250
[E85B20 |  13440/50000 ( 27%) ] Loss: 0.0381 top1= 98.7500
[E85B30 |  19840/50000 ( 40%) ] Loss: 0.0569 top1= 98.2812
[E85B40 |  26240/50000 ( 52%) ] Loss: 0.0342 top1= 99.0625
[E85B50 |  32640/50000 ( 65%) ] Loss: 0.0351 top1= 98.5938
[E85B60 |  39040/50000 ( 78%) ] Loss: 0.0334 top1= 98.7500
[E85B70 |  45440/50000 ( 91%) ] Loss: 0.0390 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2093 top1= 71.7047


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.3684 top1= 44.3409


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.5426 top1= 46.6947

Train epoch 86
[E86B0  |    640/50000 (  1%) ] Loss: 0.0538 top1= 97.5000
[E86B10 |   7040/50000 ( 14%) ] Loss: 0.0451 top1= 98.2812
[E86B20 |  13440/50000 ( 27%) ] Loss: 0.0390 top1= 98.5938
[E86B30 |  19840/50000 ( 40%) ] Loss: 0.0352 top1= 98.7500
[E86B40 |  26240/50000 ( 52%) ] Loss: 0.0434 top1= 98.7500
[E86B50 |  32640/50000 ( 65%) ] Loss: 0.0419 top1= 98.4375
[E86B60 |  39040/50000 ( 78%) ] Loss: 0.0277 top1= 99.0625
[E86B70 |  45440/50000 ( 91%) ] Loss: 0.0234 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2088 top1= 71.8750


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.6381 top1= 44.6114


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.9455 top1= 46.6546

Train epoch 87
[E87B0  |    640/50000 (  1%) ] Loss: 0.0373 top1= 98.9062
[E87B10 |   7040/50000 ( 14%) ] Loss: 0.0388 top1= 98.5938
[E87B20 |  13440/50000 ( 27%) ] Loss: 0.0315 top1= 98.9062
[E87B30 |  19840/50000 ( 40%) ] Loss: 0.0444 top1= 98.1250
[E87B40 |  26240/50000 ( 52%) ] Loss: 0.0344 top1= 99.0625
[E87B50 |  32640/50000 ( 65%) ] Loss: 0.0251 top1= 99.5312
[E87B60 |  39040/50000 ( 78%) ] Loss: 0.0199 top1= 98.9062
[E87B70 |  45440/50000 ( 91%) ] Loss: 0.0271 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2209 top1= 71.7949


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.8748 top1= 44.4511


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.7756 top1= 46.6246

Train epoch 88
[E88B0  |    640/50000 (  1%) ] Loss: 0.0569 top1= 98.1250
[E88B10 |   7040/50000 ( 14%) ] Loss: 0.0701 top1= 97.3438
[E88B20 |  13440/50000 ( 27%) ] Loss: 0.0304 top1= 98.7500
[E88B30 |  19840/50000 ( 40%) ] Loss: 0.0541 top1= 98.2812
[E88B40 |  26240/50000 ( 52%) ] Loss: 0.0407 top1= 99.0625
[E88B50 |  32640/50000 ( 65%) ] Loss: 0.0296 top1= 99.2188
[E88B60 |  39040/50000 ( 78%) ] Loss: 0.0433 top1= 98.7500
[E88B70 |  45440/50000 ( 91%) ] Loss: 0.0222 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2455 top1= 71.9451


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.0734 top1= 44.5413


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.9762 top1= 46.7248

Train epoch 89
[E89B0  |    640/50000 (  1%) ] Loss: 0.0295 top1= 99.3750
[E89B10 |   7040/50000 ( 14%) ] Loss: 0.0345 top1= 99.0625
[E89B20 |  13440/50000 ( 27%) ] Loss: 0.0350 top1= 99.2188
[E89B30 |  19840/50000 ( 40%) ] Loss: 0.0328 top1= 98.7500
[E89B40 |  26240/50000 ( 52%) ] Loss: 0.0341 top1= 98.9062
[E89B50 |  32640/50000 ( 65%) ] Loss: 0.0214 top1= 99.0625
[E89B60 |  39040/50000 ( 78%) ] Loss: 0.0304 top1= 99.0625
[E89B70 |  45440/50000 ( 91%) ] Loss: 0.0202 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2581 top1= 71.7348


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.0151 top1= 44.4611


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.1451 top1= 46.6146

Train epoch 90
[E90B0  |    640/50000 (  1%) ] Loss: 0.0377 top1= 98.7500
[E90B10 |   7040/50000 ( 14%) ] Loss: 0.0224 top1= 99.6875
[E90B20 |  13440/50000 ( 27%) ] Loss: 0.0202 top1= 99.0625
[E90B30 |  19840/50000 ( 40%) ] Loss: 0.0315 top1= 98.9062
[E90B40 |  26240/50000 ( 52%) ] Loss: 0.0247 top1= 99.3750
[E90B50 |  32640/50000 ( 65%) ] Loss: 0.0279 top1= 98.9062
[E90B60 |  39040/50000 ( 78%) ] Loss: 0.0331 top1= 98.7500
[E90B70 |  45440/50000 ( 91%) ] Loss: 0.0226 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2670 top1= 72.0353


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.2182 top1= 44.5613


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.0388 top1= 46.7047

Train epoch 91
[E91B0  |    640/50000 (  1%) ] Loss: 0.0492 top1= 98.7500
[E91B10 |   7040/50000 ( 14%) ] Loss: 0.0398 top1= 98.4375
[E91B20 |  13440/50000 ( 27%) ] Loss: 0.0306 top1= 98.7500
[E91B30 |  19840/50000 ( 40%) ] Loss: 0.0268 top1= 98.9062
[E91B40 |  26240/50000 ( 52%) ] Loss: 0.0172 top1= 99.3750
[E91B50 |  32640/50000 ( 65%) ] Loss: 0.0466 top1= 98.1250
[E91B60 |  39040/50000 ( 78%) ] Loss: 0.0194 top1= 99.2188
[E91B70 |  45440/50000 ( 91%) ] Loss: 0.0194 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.2857 top1= 72.2356


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.4886 top1= 44.5112


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.0271 top1= 46.7147

Train epoch 92
[E92B0  |    640/50000 (  1%) ] Loss: 0.0525 top1= 98.2812
[E92B10 |   7040/50000 ( 14%) ] Loss: 0.0259 top1= 98.9062
[E92B20 |  13440/50000 ( 27%) ] Loss: 0.0241 top1= 99.2188
[E92B30 |  19840/50000 ( 40%) ] Loss: 0.0349 top1= 99.2188
[E92B40 |  26240/50000 ( 52%) ] Loss: 0.0277 top1= 99.5312
[E92B50 |  32640/50000 ( 65%) ] Loss: 0.0276 top1= 99.0625
[E92B60 |  39040/50000 ( 78%) ] Loss: 0.0173 top1= 99.6875
[E92B70 |  45440/50000 ( 91%) ] Loss: 0.0204 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3168 top1= 72.3758


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.0506 top1= 44.5513


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.9307 top1= 46.5745

Train epoch 93
[E93B0  |    640/50000 (  1%) ] Loss: 0.0315 top1= 98.7500
[E93B10 |   7040/50000 ( 14%) ] Loss: 0.0430 top1= 98.5938
[E93B20 |  13440/50000 ( 27%) ] Loss: 0.0256 top1= 99.3750
[E93B30 |  19840/50000 ( 40%) ] Loss: 0.0213 top1= 99.3750
[E93B40 |  26240/50000 ( 52%) ] Loss: 0.0115 top1= 99.6875
[E93B50 |  32640/50000 ( 65%) ] Loss: 0.0102 top1= 99.6875
[E93B60 |  39040/50000 ( 78%) ] Loss: 0.0322 top1= 98.9062
[E93B70 |  45440/50000 ( 91%) ] Loss: 0.0152 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3253 top1= 71.9151


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6519 top1= 44.5513


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.2900 top1= 46.8149

Train epoch 94
[E94B0  |    640/50000 (  1%) ] Loss: 0.0482 top1= 98.1250
[E94B10 |   7040/50000 ( 14%) ] Loss: 0.0159 top1= 99.8438
[E94B20 |  13440/50000 ( 27%) ] Loss: 0.0264 top1= 99.0625
[E94B30 |  19840/50000 ( 40%) ] Loss: 0.0182 top1= 99.3750
[E94B40 |  26240/50000 ( 52%) ] Loss: 0.0222 top1= 99.3750
[E94B50 |  32640/50000 ( 65%) ] Loss: 0.0190 top1= 99.5312
[E94B60 |  39040/50000 ( 78%) ] Loss: 0.0139 top1= 99.8438
[E94B70 |  45440/50000 ( 91%) ] Loss: 0.0094 top1=100.0000

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3279 top1= 72.4058


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.0488 top1= 44.6114


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.8468 top1= 46.7348

Train epoch 95
[E95B0  |    640/50000 (  1%) ] Loss: 0.0190 top1= 99.6875
[E95B10 |   7040/50000 ( 14%) ] Loss: 0.0357 top1= 98.7500
[E95B20 |  13440/50000 ( 27%) ] Loss: 0.0283 top1= 98.9062
[E95B30 |  19840/50000 ( 40%) ] Loss: 0.0194 top1= 99.6875
[E95B40 |  26240/50000 ( 52%) ] Loss: 0.0227 top1= 99.2188
[E95B50 |  32640/50000 ( 65%) ] Loss: 0.0287 top1= 99.2188
[E95B60 |  39040/50000 ( 78%) ] Loss: 0.0244 top1= 99.0625
[E95B70 |  45440/50000 ( 91%) ] Loss: 0.0304 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3322 top1= 72.1955


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8117 top1= 44.5413


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.2330 top1= 46.7047

Train epoch 96
[E96B0  |    640/50000 (  1%) ] Loss: 0.0299 top1= 99.0625
[E96B10 |   7040/50000 ( 14%) ] Loss: 0.0293 top1= 99.2188
[E96B20 |  13440/50000 ( 27%) ] Loss: 0.0189 top1= 99.3750
[E96B30 |  19840/50000 ( 40%) ] Loss: 0.0255 top1= 99.5312
[E96B40 |  26240/50000 ( 52%) ] Loss: 0.0303 top1= 99.5312
[E96B50 |  32640/50000 ( 65%) ] Loss: 0.0146 top1= 99.6875
[E96B60 |  39040/50000 ( 78%) ] Loss: 0.0216 top1= 99.5312
[E96B70 |  45440/50000 ( 91%) ] Loss: 0.0175 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3355 top1= 72.2756


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6433 top1= 44.5713


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.5612 top1= 46.6947

Train epoch 97
[E97B0  |    640/50000 (  1%) ] Loss: 0.0253 top1= 99.3750
[E97B10 |   7040/50000 ( 14%) ] Loss: 0.0268 top1= 99.0625
[E97B20 |  13440/50000 ( 27%) ] Loss: 0.0181 top1= 99.2188
[E97B30 |  19840/50000 ( 40%) ] Loss: 0.0274 top1= 99.0625
[E97B40 |  26240/50000 ( 52%) ] Loss: 0.0284 top1= 99.5312
[E97B50 |  32640/50000 ( 65%) ] Loss: 0.0325 top1= 99.0625
[E97B60 |  39040/50000 ( 78%) ] Loss: 0.0245 top1= 99.0625
[E97B70 |  45440/50000 ( 91%) ] Loss: 0.0164 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3351 top1= 72.6262


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6351 top1= 44.6815


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.1505 top1= 46.8349

Train epoch 98
[E98B0  |    640/50000 (  1%) ] Loss: 0.0249 top1= 99.0625
[E98B10 |   7040/50000 ( 14%) ] Loss: 0.0238 top1= 99.5312
[E98B20 |  13440/50000 ( 27%) ] Loss: 0.0204 top1= 99.3750
[E98B30 |  19840/50000 ( 40%) ] Loss: 0.0218 top1= 99.3750
[E98B40 |  26240/50000 ( 52%) ] Loss: 0.0177 top1= 99.2188
[E98B50 |  32640/50000 ( 65%) ] Loss: 0.0243 top1= 99.5312
[E98B60 |  39040/50000 ( 78%) ] Loss: 0.0138 top1= 99.5312
[E98B70 |  45440/50000 ( 91%) ] Loss: 0.0164 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3413 top1= 72.5761


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.5290 top1= 44.5813


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.7606 top1= 46.9151

Train epoch 99
[E99B0  |    640/50000 (  1%) ] Loss: 0.0278 top1= 98.9062
[E99B10 |   7040/50000 ( 14%) ] Loss: 0.0197 top1= 99.5312
[E99B20 |  13440/50000 ( 27%) ] Loss: 0.0144 top1= 99.5312
[E99B30 |  19840/50000 ( 40%) ] Loss: 0.0281 top1= 99.2188
[E99B40 |  26240/50000 ( 52%) ] Loss: 0.0170 top1= 99.3750
[E99B50 |  32640/50000 ( 65%) ] Loss: 0.0392 top1= 98.7500
[E99B60 |  39040/50000 ( 78%) ] Loss: 0.0292 top1= 99.5312
[E99B70 |  45440/50000 ( 91%) ] Loss: 0.0218 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3822 top1= 72.5361


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.9218 top1= 44.5913


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.9450 top1= 46.9151

Train epoch 100
[E100B0  |    640/50000 (  1%) ] Loss: 0.0145 top1= 99.5312
[E100B10 |   7040/50000 ( 14%) ] Loss: 0.0179 top1= 99.5312
[E100B20 |  13440/50000 ( 27%) ] Loss: 0.0392 top1= 98.9062
[E100B30 |  19840/50000 ( 40%) ] Loss: 0.0201 top1= 99.3750
[E100B40 |  26240/50000 ( 52%) ] Loss: 0.0175 top1= 99.5312
[E100B50 |  32640/50000 ( 65%) ] Loss: 0.0326 top1= 98.9062
[E100B60 |  39040/50000 ( 78%) ] Loss: 0.0144 top1= 99.6875
[E100B70 |  45440/50000 ( 91%) ] Loss: 0.0155 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.3958 top1= 72.4159


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.9488 top1= 44.5813


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.7848 top1= 46.9451

Train epoch 101
[E101B0  |    640/50000 (  1%) ] Loss: 0.0250 top1= 99.2188
[E101B10 |   7040/50000 ( 14%) ] Loss: 0.0178 top1= 99.3750
[E101B20 |  13440/50000 ( 27%) ] Loss: 0.0201 top1= 99.3750
[E101B30 |  19840/50000 ( 40%) ] Loss: 0.0223 top1= 99.0625
[E101B40 |  26240/50000 ( 52%) ] Loss: 0.0209 top1= 99.5312
[E101B50 |  32640/50000 ( 65%) ] Loss: 0.0201 top1= 99.2188
[E101B60 |  39040/50000 ( 78%) ] Loss: 0.0381 top1= 99.3750
[E101B70 |  45440/50000 ( 91%) ] Loss: 0.0085 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4117 top1= 72.2155


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.9494 top1= 44.6615


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=12.0422 top1= 46.9251

Train epoch 102
[E102B0  |    640/50000 (  1%) ] Loss: 0.0160 top1= 99.2188
[E102B10 |   7040/50000 ( 14%) ] Loss: 0.0348 top1= 98.7500
[E102B20 |  13440/50000 ( 27%) ] Loss: 0.0177 top1= 99.2188
[E102B30 |  19840/50000 ( 40%) ] Loss: 0.0231 top1= 99.2188
[E102B40 |  26240/50000 ( 52%) ] Loss: 0.0127 top1= 99.5312
[E102B50 |  32640/50000 ( 65%) ] Loss: 0.0210 top1= 99.2188
[E102B60 |  39040/50000 ( 78%) ] Loss: 0.0147 top1= 99.6875
[E102B70 |  45440/50000 ( 91%) ] Loss: 0.0105 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4493 top1= 72.0252


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.1907 top1= 44.6715


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.6947 top1= 46.9351

Train epoch 103
[E103B0  |    640/50000 (  1%) ] Loss: 0.0385 top1= 98.5938
[E103B10 |   7040/50000 ( 14%) ] Loss: 0.0245 top1= 99.2188
[E103B20 |  13440/50000 ( 27%) ] Loss: 0.0087 top1= 99.8438
[E103B30 |  19840/50000 ( 40%) ] Loss: 0.0228 top1= 98.9062
[E103B40 |  26240/50000 ( 52%) ] Loss: 0.0135 top1= 99.5312
[E103B50 |  32640/50000 ( 65%) ] Loss: 0.0162 top1= 99.2188
[E103B60 |  39040/50000 ( 78%) ] Loss: 0.0192 top1= 99.2188
[E103B70 |  45440/50000 ( 91%) ] Loss: 0.0183 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4697 top1= 72.1955


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.3858 top1= 44.5613


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.5588 top1= 46.7949

Train epoch 104
[E104B0  |    640/50000 (  1%) ] Loss: 0.0249 top1= 99.0625
[E104B10 |   7040/50000 ( 14%) ] Loss: 0.0314 top1= 98.5938
[E104B20 |  13440/50000 ( 27%) ] Loss: 0.0150 top1= 99.3750
[E104B30 |  19840/50000 ( 40%) ] Loss: 0.0279 top1= 99.0625
[E104B40 |  26240/50000 ( 52%) ] Loss: 0.0086 top1=100.0000
[E104B50 |  32640/50000 ( 65%) ] Loss: 0.0275 top1= 99.3750
[E104B60 |  39040/50000 ( 78%) ] Loss: 0.0179 top1= 99.5312
[E104B70 |  45440/50000 ( 91%) ] Loss: 0.0199 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4934 top1= 72.3357


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=12.1199 top1= 44.7216


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.6633 top1= 46.9151

Train epoch 105
[E105B0  |    640/50000 (  1%) ] Loss: 0.0156 top1= 99.5312
[E105B10 |   7040/50000 ( 14%) ] Loss: 0.0318 top1= 99.2188
[E105B20 |  13440/50000 ( 27%) ] Loss: 0.0065 top1=100.0000
[E105B30 |  19840/50000 ( 40%) ] Loss: 0.0234 top1= 99.3750
[E105B40 |  26240/50000 ( 52%) ] Loss: 0.0080 top1= 99.6875
[E105B50 |  32640/50000 ( 65%) ] Loss: 0.0102 top1= 99.8438
[E105B60 |  39040/50000 ( 78%) ] Loss: 0.0134 top1= 99.6875
[E105B70 |  45440/50000 ( 91%) ] Loss: 0.0095 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.4745 top1= 72.6562


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8396 top1= 44.7716


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.6062 top1= 46.9651

Train epoch 106
[E106B0  |    640/50000 (  1%) ] Loss: 0.0404 top1= 98.5938
[E106B10 |   7040/50000 ( 14%) ] Loss: 0.0157 top1= 99.6875
[E106B20 |  13440/50000 ( 27%) ] Loss: 0.0224 top1= 99.2188
[E106B30 |  19840/50000 ( 40%) ] Loss: 0.0175 top1= 99.3750
[E106B40 |  26240/50000 ( 52%) ] Loss: 0.0156 top1= 99.6875
[E106B50 |  32640/50000 ( 65%) ] Loss: 0.0251 top1= 99.2188
[E106B60 |  39040/50000 ( 78%) ] Loss: 0.0250 top1= 99.3750
[E106B70 |  45440/50000 ( 91%) ] Loss: 0.0103 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5068 top1= 72.2957


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.7443 top1= 44.6915


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.5690 top1= 46.9651

Train epoch 107
[E107B0  |    640/50000 (  1%) ] Loss: 0.0274 top1= 99.0625
[E107B10 |   7040/50000 ( 14%) ] Loss: 0.0216 top1= 99.0625
[E107B20 |  13440/50000 ( 27%) ] Loss: 0.0167 top1= 99.3750
[E107B30 |  19840/50000 ( 40%) ] Loss: 0.0291 top1= 98.9062
[E107B40 |  26240/50000 ( 52%) ] Loss: 0.0164 top1= 99.2188
[E107B50 |  32640/50000 ( 65%) ] Loss: 0.0071 top1= 99.8438
[E107B60 |  39040/50000 ( 78%) ] Loss: 0.0160 top1= 99.6875
[E107B70 |  45440/50000 ( 91%) ] Loss: 0.0092 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5267 top1= 72.2556


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6577 top1= 44.6815


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.1837 top1= 47.0653

Train epoch 108
[E108B0  |    640/50000 (  1%) ] Loss: 0.0298 top1= 98.9062
[E108B10 |   7040/50000 ( 14%) ] Loss: 0.0216 top1= 99.0625
[E108B20 |  13440/50000 ( 27%) ] Loss: 0.0152 top1= 99.3750
[E108B30 |  19840/50000 ( 40%) ] Loss: 0.0111 top1= 99.6875
[E108B40 |  26240/50000 ( 52%) ] Loss: 0.0123 top1= 99.6875
[E108B50 |  32640/50000 ( 65%) ] Loss: 0.0175 top1= 99.3750
[E108B60 |  39040/50000 ( 78%) ] Loss: 0.0129 top1= 99.5312
[E108B70 |  45440/50000 ( 91%) ] Loss: 0.0172 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5496 top1= 72.2556


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.9319 top1= 44.5913


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.1177 top1= 47.0954

Train epoch 109
[E109B0  |    640/50000 (  1%) ] Loss: 0.0247 top1= 99.3750
[E109B10 |   7040/50000 ( 14%) ] Loss: 0.0160 top1= 99.0625
[E109B20 |  13440/50000 ( 27%) ] Loss: 0.0252 top1= 98.5938
[E109B30 |  19840/50000 ( 40%) ] Loss: 0.0256 top1= 98.9062
[E109B40 |  26240/50000 ( 52%) ] Loss: 0.0218 top1= 99.3750
[E109B50 |  32640/50000 ( 65%) ] Loss: 0.0172 top1= 99.6875
[E109B60 |  39040/50000 ( 78%) ] Loss: 0.0121 top1= 99.8438
[E109B70 |  45440/50000 ( 91%) ] Loss: 0.0183 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5508 top1= 72.5561


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8133 top1= 44.7416


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.1180 top1= 47.1955

Train epoch 110
[E110B0  |    640/50000 (  1%) ] Loss: 0.0211 top1= 99.2188
[E110B10 |   7040/50000 ( 14%) ] Loss: 0.0190 top1= 99.3750
[E110B20 |  13440/50000 ( 27%) ] Loss: 0.0176 top1= 99.5312
[E110B30 |  19840/50000 ( 40%) ] Loss: 0.0161 top1= 99.6875
[E110B40 |  26240/50000 ( 52%) ] Loss: 0.0245 top1= 99.2188
[E110B50 |  32640/50000 ( 65%) ] Loss: 0.0100 top1= 99.6875
[E110B60 |  39040/50000 ( 78%) ] Loss: 0.0123 top1= 99.5312
[E110B70 |  45440/50000 ( 91%) ] Loss: 0.0183 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5763 top1= 72.4159


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.8950 top1= 44.6715


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=11.0676 top1= 47.3357

Train epoch 111
[E111B0  |    640/50000 (  1%) ] Loss: 0.0271 top1= 99.2188
[E111B10 |   7040/50000 ( 14%) ] Loss: 0.0184 top1= 99.3750
[E111B20 |  13440/50000 ( 27%) ] Loss: 0.0299 top1= 99.5312
[E111B30 |  19840/50000 ( 40%) ] Loss: 0.0119 top1= 99.6875
[E111B40 |  26240/50000 ( 52%) ] Loss: 0.0108 top1= 99.3750
[E111B50 |  32640/50000 ( 65%) ] Loss: 0.0173 top1= 99.3750
[E111B60 |  39040/50000 ( 78%) ] Loss: 0.0101 top1= 99.8438
[E111B70 |  45440/50000 ( 91%) ] Loss: 0.0118 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5575 top1= 72.7063


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6276 top1= 44.7516


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.8601 top1= 47.2356

Train epoch 112
[E112B0  |    640/50000 (  1%) ] Loss: 0.0192 top1= 99.3750
[E112B10 |   7040/50000 ( 14%) ] Loss: 0.0285 top1= 99.3750
[E112B20 |  13440/50000 ( 27%) ] Loss: 0.0163 top1= 99.3750
[E112B30 |  19840/50000 ( 40%) ] Loss: 0.0197 top1= 99.3750
[E112B40 |  26240/50000 ( 52%) ] Loss: 0.0161 top1= 99.5312
[E112B50 |  32640/50000 ( 65%) ] Loss: 0.0145 top1= 99.3750
[E112B60 |  39040/50000 ( 78%) ] Loss: 0.0138 top1= 99.3750
[E112B70 |  45440/50000 ( 91%) ] Loss: 0.0293 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6011 top1= 72.4159


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.6892 top1= 44.7416


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.7534 top1= 47.3257

Train epoch 113
[E113B0  |    640/50000 (  1%) ] Loss: 0.0239 top1= 99.3750
[E113B10 |   7040/50000 ( 14%) ] Loss: 0.0185 top1= 99.5312
[E113B20 |  13440/50000 ( 27%) ] Loss: 0.0158 top1= 99.0625
[E113B30 |  19840/50000 ( 40%) ] Loss: 0.0189 top1= 99.2188
[E113B40 |  26240/50000 ( 52%) ] Loss: 0.0184 top1= 99.3750
[E113B50 |  32640/50000 ( 65%) ] Loss: 0.0201 top1= 99.3750
[E113B60 |  39040/50000 ( 78%) ] Loss: 0.0140 top1= 99.3750
[E113B70 |  45440/50000 ( 91%) ] Loss: 0.0069 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6115 top1= 72.5962


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.3920 top1= 44.7616


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.3752 top1= 47.4860

Train epoch 114
[E114B0  |    640/50000 (  1%) ] Loss: 0.0178 top1= 99.2188
[E114B10 |   7040/50000 ( 14%) ] Loss: 0.0260 top1= 98.7500
[E114B20 |  13440/50000 ( 27%) ] Loss: 0.0099 top1= 99.5312
[E114B30 |  19840/50000 ( 40%) ] Loss: 0.0065 top1=100.0000
[E114B40 |  26240/50000 ( 52%) ] Loss: 0.0170 top1= 99.5312
[E114B50 |  32640/50000 ( 65%) ] Loss: 0.0133 top1= 99.5312
[E114B60 |  39040/50000 ( 78%) ] Loss: 0.0167 top1= 99.2188
[E114B70 |  45440/50000 ( 91%) ] Loss: 0.0109 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6173 top1= 72.5461


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.2035 top1= 44.7015


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.8453 top1= 47.5861

Train epoch 115
[E115B0  |    640/50000 (  1%) ] Loss: 0.0289 top1= 98.9062
[E115B10 |   7040/50000 ( 14%) ] Loss: 0.0248 top1= 99.3750
[E115B20 |  13440/50000 ( 27%) ] Loss: 0.0176 top1= 99.0625
[E115B30 |  19840/50000 ( 40%) ] Loss: 0.0147 top1= 99.6875
[E115B40 |  26240/50000 ( 52%) ] Loss: 0.0122 top1= 99.6875
[E115B50 |  32640/50000 ( 65%) ] Loss: 0.0114 top1= 99.5312
[E115B60 |  39040/50000 ( 78%) ] Loss: 0.0154 top1= 99.3750
[E115B70 |  45440/50000 ( 91%) ] Loss: 0.0177 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6299 top1= 72.7965


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.5763 top1= 44.9119


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.6182 top1= 47.5561

Train epoch 116
[E116B0  |    640/50000 (  1%) ] Loss: 0.0154 top1= 99.5312
[E116B10 |   7040/50000 ( 14%) ] Loss: 0.0148 top1= 99.5312
[E116B20 |  13440/50000 ( 27%) ] Loss: 0.0103 top1= 99.6875
[E116B30 |  19840/50000 ( 40%) ] Loss: 0.0345 top1= 98.7500
[E116B40 |  26240/50000 ( 52%) ] Loss: 0.0092 top1= 99.5312
[E116B50 |  32640/50000 ( 65%) ] Loss: 0.0165 top1= 99.3750
[E116B60 |  39040/50000 ( 78%) ] Loss: 0.0168 top1= 99.6875
[E116B70 |  45440/50000 ( 91%) ] Loss: 0.0220 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6718 top1= 72.8966


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.5861 top1= 44.7316


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.3248 top1= 47.8365

Train epoch 117
[E117B0  |    640/50000 (  1%) ] Loss: 0.0127 top1= 99.6875
[E117B10 |   7040/50000 ( 14%) ] Loss: 0.0177 top1= 99.3750
[E117B20 |  13440/50000 ( 27%) ] Loss: 0.0154 top1= 99.3750
[E117B30 |  19840/50000 ( 40%) ] Loss: 0.0222 top1= 98.9062
[E117B40 |  26240/50000 ( 52%) ] Loss: 0.0049 top1=100.0000
[E117B50 |  32640/50000 ( 65%) ] Loss: 0.0112 top1= 99.6875
[E117B60 |  39040/50000 ( 78%) ] Loss: 0.0167 top1= 99.5312
[E117B70 |  45440/50000 ( 91%) ] Loss: 0.0105 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6742 top1= 72.9868


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.1597 top1= 44.9820


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.3278 top1= 47.9267

Train epoch 118
[E118B0  |    640/50000 (  1%) ] Loss: 0.0236 top1= 99.6875
[E118B10 |   7040/50000 ( 14%) ] Loss: 0.0266 top1= 99.2188
[E118B20 |  13440/50000 ( 27%) ] Loss: 0.0086 top1= 99.5312
[E118B30 |  19840/50000 ( 40%) ] Loss: 0.0096 top1= 99.8438
[E118B40 |  26240/50000 ( 52%) ] Loss: 0.0101 top1= 99.8438
[E118B50 |  32640/50000 ( 65%) ] Loss: 0.0155 top1= 99.5312
[E118B60 |  39040/50000 ( 78%) ] Loss: 0.0167 top1= 99.3750
[E118B70 |  45440/50000 ( 91%) ] Loss: 0.0151 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7330 top1= 72.8065


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.2802 top1= 45.1022


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=10.0972 top1= 48.1270

Train epoch 119
[E119B0  |    640/50000 (  1%) ] Loss: 0.0281 top1= 99.3750
[E119B10 |   7040/50000 ( 14%) ] Loss: 0.0227 top1= 99.2188
[E119B20 |  13440/50000 ( 27%) ] Loss: 0.0156 top1= 99.3750
[E119B30 |  19840/50000 ( 40%) ] Loss: 0.0195 top1= 99.6875
[E119B40 |  26240/50000 ( 52%) ] Loss: 0.0090 top1= 99.8438
[E119B50 |  32640/50000 ( 65%) ] Loss: 0.0116 top1= 99.5312
[E119B60 |  39040/50000 ( 78%) ] Loss: 0.0241 top1= 99.3750
[E119B70 |  45440/50000 ( 91%) ] Loss: 0.0119 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7759 top1= 72.6863


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=11.1950 top1= 45.1823


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.9504 top1= 48.5076

Train epoch 120
[E120B0  |    640/50000 (  1%) ] Loss: 0.0216 top1= 99.3750
[E120B10 |   7040/50000 ( 14%) ] Loss: 0.0107 top1= 99.6875
[E120B20 |  13440/50000 ( 27%) ] Loss: 0.0325 top1= 99.0625
[E120B30 |  19840/50000 ( 40%) ] Loss: 0.0262 top1= 99.0625
[E120B40 |  26240/50000 ( 52%) ] Loss: 0.0185 top1= 99.2188
[E120B50 |  32640/50000 ( 65%) ] Loss: 0.0132 top1= 99.8438
[E120B60 |  39040/50000 ( 78%) ] Loss: 0.0092 top1= 99.8438
[E120B70 |  45440/50000 ( 91%) ] Loss: 0.0097 top1= 99.6875

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7922 top1= 72.6062


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=10.9864 top1= 45.3225


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=9.7690 top1= 48.7280

Train epoch 121
[E121B0  |    640/50000 (  1%) ] Loss: 0.0342 top1= 98.9062
[E121B10 |   7040/50000 ( 14%) ] Loss: 0.0176 top1= 99.5312
[E121B20 |  13440/50000 ( 27%) ] Loss: 0.0129 top1= 99.6875
[E121B30 |  19840/50000 ( 40%) ] Loss: 0.0123 top1= 99.5312
[E121B40 |  26240/50000 ( 52%) ] Loss: 0.0212 top1= 99.3750
[E121B50 |  32640/50000 ( 65%) ] Loss: 0.0187 top1= 99.6875
[E121B60 |  39040/50000 ( 78%) ] Loss: 0.0211 top1= 99.3750
[E121B70 |  45440/50000 ( 91%) ] Loss: 0.0184 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7951 top1= 72.7063


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=9.2580 top1= 46.6947


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=7.8638 top1= 50.9515

Train epoch 122
[E122B0  |    640/50000 (  1%) ] Loss: 0.0195 top1= 99.5312
[E122B10 |   7040/50000 ( 14%) ] Loss: 0.0334 top1= 98.9062
[E122B20 |  13440/50000 ( 27%) ] Loss: 0.0187 top1= 99.2188
[E122B30 |  19840/50000 ( 40%) ] Loss: 0.0275 top1= 99.0625
[E122B40 |  26240/50000 ( 52%) ] Loss: 0.0223 top1= 99.0625
[E122B50 |  32640/50000 ( 65%) ] Loss: 0.0319 top1= 98.5938
[E122B60 |  39040/50000 ( 78%) ] Loss: 0.0210 top1= 99.3750
[E122B70 |  45440/50000 ( 91%) ] Loss: 0.0107 top1= 99.8438

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7757 top1= 72.7764


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.9967 top1= 48.9683


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=6.6025 top1= 53.3053

Train epoch 123
[E123B0  |    640/50000 (  1%) ] Loss: 0.0596 top1= 97.9688
[E123B10 |   7040/50000 ( 14%) ] Loss: 0.0373 top1= 98.7500
[E123B20 |  13440/50000 ( 27%) ] Loss: 0.0367 top1= 98.7500
[E123B30 |  19840/50000 ( 40%) ] Loss: 0.0369 top1= 98.9062
[E123B40 |  26240/50000 ( 52%) ] Loss: 0.0378 top1= 98.7500
[E123B50 |  32640/50000 ( 65%) ] Loss: 0.0351 top1= 98.9062
[E123B60 |  39040/50000 ( 78%) ] Loss: 0.0450 top1= 98.5938
[E123B70 |  45440/50000 ( 91%) ] Loss: 0.0248 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7634 top1= 72.6963


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=7.2122 top1= 50.8514


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.8984 top1= 55.0180

Train epoch 124
[E124B0  |    640/50000 (  1%) ] Loss: 0.0589 top1= 98.7500
[E124B10 |   7040/50000 ( 14%) ] Loss: 0.0443 top1= 98.4375
[E124B20 |  13440/50000 ( 27%) ] Loss: 0.0343 top1= 99.0625
[E124B30 |  19840/50000 ( 40%) ] Loss: 0.0346 top1= 99.2188
[E124B40 |  26240/50000 ( 52%) ] Loss: 0.0315 top1= 99.3750
[E124B50 |  32640/50000 ( 65%) ] Loss: 0.0461 top1= 98.4375
[E124B60 |  39040/50000 ( 78%) ] Loss: 0.0392 top1= 98.5938
[E124B70 |  45440/50000 ( 91%) ] Loss: 0.0348 top1= 99.2188

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.7342 top1= 72.8666


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.8887 top1= 51.1018


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.6790 top1= 55.4487

Train epoch 125
[E125B0  |    640/50000 (  1%) ] Loss: 0.0512 top1= 98.7500
[E125B10 |   7040/50000 ( 14%) ] Loss: 0.0584 top1= 97.9688
[E125B20 |  13440/50000 ( 27%) ] Loss: 0.0529 top1= 99.0625
[E125B30 |  19840/50000 ( 40%) ] Loss: 0.0505 top1= 98.7500
[E125B40 |  26240/50000 ( 52%) ] Loss: 0.0615 top1= 98.7500
[E125B50 |  32640/50000 ( 65%) ] Loss: 0.0502 top1= 98.1250
[E125B60 |  39040/50000 ( 78%) ] Loss: 0.0661 top1= 97.8125
[E125B70 |  45440/50000 ( 91%) ] Loss: 0.0422 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6945 top1= 73.0569


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=6.1137 top1= 53.1951


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.3009 top1= 56.2500

Train epoch 126
[E126B0  |    640/50000 (  1%) ] Loss: 0.0518 top1= 98.5938
[E126B10 |   7040/50000 ( 14%) ] Loss: 0.0723 top1= 97.9688
[E126B20 |  13440/50000 ( 27%) ] Loss: 0.0584 top1= 98.1250
[E126B30 |  19840/50000 ( 40%) ] Loss: 0.0919 top1= 98.5938
[E126B40 |  26240/50000 ( 52%) ] Loss: 0.0584 top1= 98.4375
[E126B50 |  32640/50000 ( 65%) ] Loss: 0.0467 top1= 98.4375
[E126B60 |  39040/50000 ( 78%) ] Loss: 0.0715 top1= 97.8125
[E126B70 |  45440/50000 ( 91%) ] Loss: 0.0592 top1= 98.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6630 top1= 72.9768


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8698 top1= 53.5657


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=5.1478 top1= 56.7007

Train epoch 127
[E127B0  |    640/50000 (  1%) ] Loss: 0.0584 top1= 98.5938
[E127B10 |   7040/50000 ( 14%) ] Loss: 0.0623 top1= 97.8125
[E127B20 |  13440/50000 ( 27%) ] Loss: 0.0460 top1= 98.4375
[E127B30 |  19840/50000 ( 40%) ] Loss: 0.0688 top1= 97.9688
[E127B40 |  26240/50000 ( 52%) ] Loss: 0.0428 top1= 98.4375
[E127B50 |  32640/50000 ( 65%) ] Loss: 0.0435 top1= 98.4375
[E127B60 |  39040/50000 ( 78%) ] Loss: 0.0622 top1= 97.5000
[E127B70 |  45440/50000 ( 91%) ] Loss: 0.0481 top1= 97.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6532 top1= 73.0669


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.8014 top1= 53.5156


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8973 top1= 57.4820

Train epoch 128
[E128B0  |    640/50000 (  1%) ] Loss: 0.0569 top1= 98.1250
[E128B10 |   7040/50000 ( 14%) ] Loss: 0.0711 top1= 97.1875
[E128B20 |  13440/50000 ( 27%) ] Loss: 0.0597 top1= 98.1250
[E128B30 |  19840/50000 ( 40%) ] Loss: 0.0855 top1= 97.3438
[E128B40 |  26240/50000 ( 52%) ] Loss: 0.0495 top1= 97.9688
[E128B50 |  32640/50000 ( 65%) ] Loss: 0.0335 top1= 98.7500
[E128B60 |  39040/50000 ( 78%) ] Loss: 0.0514 top1= 98.4375
[E128B70 |  45440/50000 ( 91%) ] Loss: 0.0446 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6446 top1= 73.0268


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.6196 top1= 53.8762


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.9513 top1= 57.1615

Train epoch 129
[E129B0  |    640/50000 (  1%) ] Loss: 0.0707 top1= 98.2812
[E129B10 |   7040/50000 ( 14%) ] Loss: 0.0409 top1= 98.5938
[E129B20 |  13440/50000 ( 27%) ] Loss: 0.0546 top1= 98.4375
[E129B30 |  19840/50000 ( 40%) ] Loss: 0.0722 top1= 97.1875
[E129B40 |  26240/50000 ( 52%) ] Loss: 0.0767 top1= 97.6562
[E129B50 |  32640/50000 ( 65%) ] Loss: 0.0630 top1= 97.8125
[E129B60 |  39040/50000 ( 78%) ] Loss: 0.0538 top1= 98.2812
[E129B70 |  45440/50000 ( 91%) ] Loss: 0.0354 top1= 99.3750

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6233 top1= 72.9667


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4908 top1= 54.1366


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8824 top1= 57.3718

Train epoch 130
[E130B0  |    640/50000 (  1%) ] Loss: 0.0864 top1= 97.5000
[E130B10 |   7040/50000 ( 14%) ] Loss: 0.0686 top1= 97.6562
[E130B20 |  13440/50000 ( 27%) ] Loss: 0.0402 top1= 98.7500
[E130B30 |  19840/50000 ( 40%) ] Loss: 0.0507 top1= 97.9688
[E130B40 |  26240/50000 ( 52%) ] Loss: 0.0507 top1= 98.1250
[E130B50 |  32640/50000 ( 65%) ] Loss: 0.0535 top1= 97.9688
[E130B60 |  39040/50000 ( 78%) ] Loss: 0.0808 top1= 97.9688
[E130B70 |  45440/50000 ( 91%) ] Loss: 0.0403 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6090 top1= 73.2372


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4231 top1= 54.2768


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8453 top1= 57.4619

Train epoch 131
[E131B0  |    640/50000 (  1%) ] Loss: 0.0826 top1= 97.8125
[E131B10 |   7040/50000 ( 14%) ] Loss: 0.0797 top1= 97.5000
[E131B20 |  13440/50000 ( 27%) ] Loss: 0.0394 top1= 99.2188
[E131B30 |  19840/50000 ( 40%) ] Loss: 0.0723 top1= 97.5000
[E131B40 |  26240/50000 ( 52%) ] Loss: 0.0594 top1= 98.9062
[E131B50 |  32640/50000 ( 65%) ] Loss: 0.0652 top1= 97.6562
[E131B60 |  39040/50000 ( 78%) ] Loss: 0.0566 top1= 98.4375
[E131B70 |  45440/50000 ( 91%) ] Loss: 0.0451 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.6086 top1= 73.1571


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.5066 top1= 53.7159


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8328 top1= 57.4319

Train epoch 132
[E132B0  |    640/50000 (  1%) ] Loss: 0.0582 top1= 98.4375
[E132B10 |   7040/50000 ( 14%) ] Loss: 0.0685 top1= 98.2812
[E132B20 |  13440/50000 ( 27%) ] Loss: 0.0714 top1= 98.2812
[E132B30 |  19840/50000 ( 40%) ] Loss: 0.0647 top1= 97.8125
[E132B40 |  26240/50000 ( 52%) ] Loss: 0.0635 top1= 97.8125
[E132B50 |  32640/50000 ( 65%) ] Loss: 0.0626 top1= 97.9688
[E132B60 |  39040/50000 ( 78%) ] Loss: 0.0679 top1= 98.1250
[E132B70 |  45440/50000 ( 91%) ] Loss: 0.0499 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5975 top1= 73.2572


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4887 top1= 53.8562


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8316 top1= 57.4419

Train epoch 133
[E133B0  |    640/50000 (  1%) ] Loss: 0.0506 top1= 98.1250
[E133B10 |   7040/50000 ( 14%) ] Loss: 0.0606 top1= 97.6562
[E133B20 |  13440/50000 ( 27%) ] Loss: 0.0473 top1= 98.2812
[E133B30 |  19840/50000 ( 40%) ] Loss: 0.0518 top1= 98.1250
[E133B40 |  26240/50000 ( 52%) ] Loss: 0.0593 top1= 98.1250
[E133B50 |  32640/50000 ( 65%) ] Loss: 0.0463 top1= 98.2812
[E133B60 |  39040/50000 ( 78%) ] Loss: 0.0365 top1= 99.0625
[E133B70 |  45440/50000 ( 91%) ] Loss: 0.0424 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5938 top1= 73.3373


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.6095 top1= 53.1751


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8301 top1= 57.1915

Train epoch 134
[E134B0  |    640/50000 (  1%) ] Loss: 0.0455 top1= 98.4375
[E134B10 |   7040/50000 ( 14%) ] Loss: 0.0615 top1= 97.8125
[E134B20 |  13440/50000 ( 27%) ] Loss: 0.0394 top1= 99.2188
[E134B30 |  19840/50000 ( 40%) ] Loss: 0.0422 top1= 98.1250
[E134B40 |  26240/50000 ( 52%) ] Loss: 0.0608 top1= 98.4375
[E134B50 |  32640/50000 ( 65%) ] Loss: 0.0644 top1= 97.6562
[E134B60 |  39040/50000 ( 78%) ] Loss: 0.0426 top1= 98.4375
[E134B70 |  45440/50000 ( 91%) ] Loss: 0.0583 top1= 98.2812

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5899 top1= 73.3774


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4140 top1= 53.9563


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8724 top1= 56.9812

Train epoch 135
[E135B0  |    640/50000 (  1%) ] Loss: 0.0408 top1= 98.5938
[E135B10 |   7040/50000 ( 14%) ] Loss: 0.0439 top1= 98.5938
[E135B20 |  13440/50000 ( 27%) ] Loss: 0.0343 top1= 99.0625
[E135B30 |  19840/50000 ( 40%) ] Loss: 0.0682 top1= 97.8125
[E135B40 |  26240/50000 ( 52%) ] Loss: 0.0622 top1= 97.9688
[E135B50 |  32640/50000 ( 65%) ] Loss: 0.0615 top1= 98.2812
[E135B60 |  39040/50000 ( 78%) ] Loss: 0.0722 top1= 98.1250
[E135B70 |  45440/50000 ( 91%) ] Loss: 0.0512 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5826 top1= 73.2772


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4567 top1= 53.6058


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7896 top1= 57.4119

Train epoch 136
[E136B0  |    640/50000 (  1%) ] Loss: 0.0681 top1= 97.9688
[E136B10 |   7040/50000 ( 14%) ] Loss: 0.0964 top1= 97.1875
[E136B20 |  13440/50000 ( 27%) ] Loss: 0.0559 top1= 98.4375
[E136B30 |  19840/50000 ( 40%) ] Loss: 0.0333 top1= 99.5312
[E136B40 |  26240/50000 ( 52%) ] Loss: 0.0410 top1= 98.5938
[E136B50 |  32640/50000 ( 65%) ] Loss: 0.0605 top1= 98.5938
[E136B60 |  39040/50000 ( 78%) ] Loss: 0.0610 top1= 98.4375
[E136B70 |  45440/50000 ( 91%) ] Loss: 0.0435 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5748 top1= 73.2272


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2382 top1= 54.2869


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7202 top1= 57.6923

Train epoch 137
[E137B0  |    640/50000 (  1%) ] Loss: 0.0674 top1= 97.9688
[E137B10 |   7040/50000 ( 14%) ] Loss: 0.0656 top1= 97.5000
[E137B20 |  13440/50000 ( 27%) ] Loss: 0.0330 top1= 98.7500
[E137B30 |  19840/50000 ( 40%) ] Loss: 0.0507 top1= 98.9062
[E137B40 |  26240/50000 ( 52%) ] Loss: 0.0556 top1= 97.8125
[E137B50 |  32640/50000 ( 65%) ] Loss: 0.0529 top1= 98.2812
[E137B60 |  39040/50000 ( 78%) ] Loss: 0.0499 top1= 99.0625
[E137B70 |  45440/50000 ( 91%) ] Loss: 0.0449 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5799 top1= 73.3373


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4925 top1= 53.2652


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5989 top1= 58.0629

Train epoch 138
[E138B0  |    640/50000 (  1%) ] Loss: 0.0541 top1= 98.1250
[E138B10 |   7040/50000 ( 14%) ] Loss: 0.0662 top1= 98.2812
[E138B20 |  13440/50000 ( 27%) ] Loss: 0.0583 top1= 97.8125
[E138B30 |  19840/50000 ( 40%) ] Loss: 0.0576 top1= 98.2812
[E138B40 |  26240/50000 ( 52%) ] Loss: 0.0615 top1= 97.8125
[E138B50 |  32640/50000 ( 65%) ] Loss: 0.0514 top1= 98.2812
[E138B60 |  39040/50000 ( 78%) ] Loss: 0.0746 top1= 97.8125
[E138B70 |  45440/50000 ( 91%) ] Loss: 0.0557 top1= 97.9688

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5831 top1= 73.2572


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.4346 top1= 53.4355


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6489 top1= 57.8025

Train epoch 139
[E139B0  |    640/50000 (  1%) ] Loss: 0.0828 top1= 97.5000
[E139B10 |   7040/50000 ( 14%) ] Loss: 0.1061 top1= 96.8750
[E139B20 |  13440/50000 ( 27%) ] Loss: 0.0505 top1= 98.2812
[E139B30 |  19840/50000 ( 40%) ] Loss: 0.0460 top1= 98.2812
[E139B40 |  26240/50000 ( 52%) ] Loss: 0.0476 top1= 98.4375
[E139B50 |  32640/50000 ( 65%) ] Loss: 0.0377 top1= 99.2188
[E139B60 |  39040/50000 ( 78%) ] Loss: 0.0430 top1= 98.5938
[E139B70 |  45440/50000 ( 91%) ] Loss: 0.0346 top1= 99.0625

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5738 top1= 73.4475


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3900 top1= 53.5156


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6751 top1= 57.7123

Train epoch 140
[E140B0  |    640/50000 (  1%) ] Loss: 0.0432 top1= 98.5938
[E140B10 |   7040/50000 ( 14%) ] Loss: 0.0547 top1= 98.1250
[E140B20 |  13440/50000 ( 27%) ] Loss: 0.0427 top1= 98.5938
[E140B30 |  19840/50000 ( 40%) ] Loss: 0.0511 top1= 98.2812
[E140B40 |  26240/50000 ( 52%) ] Loss: 0.0451 top1= 98.9062
[E140B50 |  32640/50000 ( 65%) ] Loss: 0.0531 top1= 98.7500
[E140B60 |  39040/50000 ( 78%) ] Loss: 0.0420 top1= 98.5938
[E140B70 |  45440/50000 ( 91%) ] Loss: 0.0534 top1= 98.5938

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5756 top1= 73.5577


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2939 top1= 53.9062


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5026 top1= 58.3233

Train epoch 141
[E141B0  |    640/50000 (  1%) ] Loss: 0.0538 top1= 97.9688
[E141B10 |   7040/50000 ( 14%) ] Loss: 0.0516 top1= 97.9688
[E141B20 |  13440/50000 ( 27%) ] Loss: 0.0680 top1= 98.7500
[E141B30 |  19840/50000 ( 40%) ] Loss: 0.0980 top1= 97.6562
[E141B40 |  26240/50000 ( 52%) ] Loss: 0.0683 top1= 98.1250
[E141B50 |  32640/50000 ( 65%) ] Loss: 0.0586 top1= 98.2812
[E141B60 |  39040/50000 ( 78%) ] Loss: 0.0791 top1= 97.5000
[E141B70 |  45440/50000 ( 91%) ] Loss: 0.0272 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5735 top1= 73.6579


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2543 top1= 54.0665


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.5232 top1= 58.3233

Train epoch 142
[E142B0  |    640/50000 (  1%) ] Loss: 0.0382 top1= 98.7500
[E142B10 |   7040/50000 ( 14%) ] Loss: 0.0642 top1= 97.9688
[E142B20 |  13440/50000 ( 27%) ] Loss: 0.0533 top1= 99.2188
[E142B30 |  19840/50000 ( 40%) ] Loss: 0.0512 top1= 97.8125
[E142B40 |  26240/50000 ( 52%) ] Loss: 0.0552 top1= 98.1250
[E142B50 |  32640/50000 ( 65%) ] Loss: 0.0537 top1= 98.4375
[E142B60 |  39040/50000 ( 78%) ] Loss: 0.0430 top1= 98.7500
[E142B70 |  45440/50000 ( 91%) ] Loss: 0.0482 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5757 top1= 73.3373


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.3179 top1= 53.7861


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7751 top1= 57.5621

Train epoch 143
[E143B0  |    640/50000 (  1%) ] Loss: 0.0661 top1= 98.2812
[E143B10 |   7040/50000 ( 14%) ] Loss: 0.0804 top1= 97.1875
[E143B20 |  13440/50000 ( 27%) ] Loss: 0.0495 top1= 98.4375
[E143B30 |  19840/50000 ( 40%) ] Loss: 0.0606 top1= 97.8125
[E143B40 |  26240/50000 ( 52%) ] Loss: 0.0447 top1= 98.7500
[E143B50 |  32640/50000 ( 65%) ] Loss: 0.0313 top1= 99.0625
[E143B60 |  39040/50000 ( 78%) ] Loss: 0.0541 top1= 98.7500
[E143B70 |  45440/50000 ( 91%) ] Loss: 0.0381 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5627 top1= 73.5677


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1575 top1= 54.3169


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7416 top1= 57.6522

Train epoch 144
[E144B0  |    640/50000 (  1%) ] Loss: 0.0547 top1= 97.6562
[E144B10 |   7040/50000 ( 14%) ] Loss: 0.0587 top1= 97.9688
[E144B20 |  13440/50000 ( 27%) ] Loss: 0.0697 top1= 98.5938
[E144B30 |  19840/50000 ( 40%) ] Loss: 0.0731 top1= 97.5000
[E144B40 |  26240/50000 ( 52%) ] Loss: 0.0454 top1= 98.7500
[E144B50 |  32640/50000 ( 65%) ] Loss: 0.0532 top1= 97.6562
[E144B60 |  39040/50000 ( 78%) ] Loss: 0.0424 top1= 98.7500
[E144B70 |  45440/50000 ( 91%) ] Loss: 0.0611 top1= 98.4375

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5601 top1= 73.5677


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.1357 top1= 54.2568


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6470 top1= 58.0128

Train epoch 145
[E145B0  |    640/50000 (  1%) ] Loss: 0.0472 top1= 98.4375
[E145B10 |   7040/50000 ( 14%) ] Loss: 0.0730 top1= 97.6562
[E145B20 |  13440/50000 ( 27%) ] Loss: 0.0353 top1= 98.7500
[E145B30 |  19840/50000 ( 40%) ] Loss: 0.0706 top1= 97.9688
[E145B40 |  26240/50000 ( 52%) ] Loss: 0.0702 top1= 97.3438
[E145B50 |  32640/50000 ( 65%) ] Loss: 0.0455 top1= 98.2812
[E145B60 |  39040/50000 ( 78%) ] Loss: 0.0331 top1= 99.0625
[E145B70 |  45440/50000 ( 91%) ] Loss: 0.0667 top1= 97.6562

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5633 top1= 73.5377


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.0342 top1= 54.9079


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6446 top1= 58.0729

Train epoch 146
[E146B0  |    640/50000 (  1%) ] Loss: 0.0473 top1= 98.2812
[E146B10 |   7040/50000 ( 14%) ] Loss: 0.0670 top1= 97.8125
[E146B20 |  13440/50000 ( 27%) ] Loss: 0.0552 top1= 97.6562
[E146B30 |  19840/50000 ( 40%) ] Loss: 0.0450 top1= 98.2812
[E146B40 |  26240/50000 ( 52%) ] Loss: 0.0677 top1= 97.6562
[E146B50 |  32640/50000 ( 65%) ] Loss: 0.0363 top1= 98.7500
[E146B60 |  39040/50000 ( 78%) ] Loss: 0.0732 top1= 97.6562
[E146B70 |  45440/50000 ( 91%) ] Loss: 0.0325 top1= 98.7500

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5765 top1= 73.3874


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2861 top1= 54.0565


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.8742 top1= 57.3117

Train epoch 147
[E147B0  |    640/50000 (  1%) ] Loss: 0.0708 top1= 97.3438
[E147B10 |   7040/50000 ( 14%) ] Loss: 0.0751 top1= 97.8125
[E147B20 |  13440/50000 ( 27%) ] Loss: 0.0445 top1= 99.0625
[E147B30 |  19840/50000 ( 40%) ] Loss: 0.0576 top1= 98.2812
[E147B40 |  26240/50000 ( 52%) ] Loss: 0.0357 top1= 98.7500
[E147B50 |  32640/50000 ( 65%) ] Loss: 0.0464 top1= 98.1250
[E147B60 |  39040/50000 ( 78%) ] Loss: 0.0499 top1= 98.1250
[E147B70 |  45440/50000 ( 91%) ] Loss: 0.0417 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5700 top1= 73.6078


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2834 top1= 53.8361


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.7914 top1= 57.5521

Train epoch 148
[E148B0  |    640/50000 (  1%) ] Loss: 0.0965 top1= 98.1250
[E148B10 |   7040/50000 ( 14%) ] Loss: 0.0660 top1= 97.8125
[E148B20 |  13440/50000 ( 27%) ] Loss: 0.0638 top1= 97.9688
[E148B30 |  19840/50000 ( 40%) ] Loss: 0.0628 top1= 97.5000
[E148B40 |  26240/50000 ( 52%) ] Loss: 0.0416 top1= 98.1250
[E148B50 |  32640/50000 ( 65%) ] Loss: 0.0489 top1= 98.5938
[E148B60 |  39040/50000 ( 78%) ] Loss: 0.0612 top1= 98.1250
[E148B70 |  45440/50000 ( 91%) ] Loss: 0.0392 top1= 98.9062

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5817 top1= 73.4575


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2541 top1= 53.9463


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.4967 top1= 58.5637

Train epoch 149
[E149B0  |    640/50000 (  1%) ] Loss: 0.0495 top1= 98.5938
[E149B10 |   7040/50000 ( 14%) ] Loss: 0.0768 top1= 97.5000
[E149B20 |  13440/50000 ( 27%) ] Loss: 0.0682 top1= 98.4375
[E149B30 |  19840/50000 ( 40%) ] Loss: 0.0456 top1= 98.5938
[E149B40 |  26240/50000 ( 52%) ] Loss: 0.0446 top1= 98.2812
[E149B50 |  32640/50000 ( 65%) ] Loss: 0.0404 top1= 99.0625
[E149B60 |  39040/50000 ( 78%) ] Loss: 0.0466 top1= 98.2812
[E149B70 |  45440/50000 ( 91%) ] Loss: 0.0542 top1= 98.1250

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5678 top1= 73.5577


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2416 top1= 53.9764


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6543 top1= 57.9527

Train epoch 150
[E150B0  |    640/50000 (  1%) ] Loss: 0.0649 top1= 98.4375
[E150B10 |   7040/50000 ( 14%) ] Loss: 0.0488 top1= 98.2812
[E150B20 |  13440/50000 ( 27%) ] Loss: 0.0312 top1= 98.9062
[E150B30 |  19840/50000 ( 40%) ] Loss: 0.0594 top1= 97.9688
[E150B40 |  26240/50000 ( 52%) ] Loss: 0.0435 top1= 97.9688
[E150B50 |  32640/50000 ( 65%) ] Loss: 0.0611 top1= 98.7500
[E150B60 |  39040/50000 ( 78%) ] Loss: 0.0412 top1= 98.7500
[E150B70 |  45440/50000 ( 91%) ] Loss: 0.0349 top1= 99.5312

=> Averaged model (Global Average Validation Accuracy) | Eval Loss=1.5897 top1= 73.2372


=> Averaged model (Clique1 Average Validation Accuracy) | Eval Loss=5.2982 top1= 53.9163


=> Averaged model (Clique2 Average Validation Accuracy) | Eval Loss=4.6765 top1= 57.8826

