==========
Args:Namespace(dataset='market1501', batch_size=256, workers=4, height=256, width=128, num_instances=8, eps=0.6, eps_gap=0.02, k1=30, k2=6, arch='resnet_ibn101a', pretrained_path='/home/ma-user/work/Projects/ReIDNets_Checkpoints_TransReID/ResNet101_IBN.pth', features=0, dropout=0, momentum=0.2, drop_path_rate=0.3, hw_ratio=2, self_norm=True, conv_stem=False, lr=0.00035, weight_decay=0.0005, optimizer='SGD', epochs=50, iters=200, step_size=20, seed=1, print_freq=20, eval_step=10, temp=0.05, data_dir='/home/ma-user/work/Projects/ReIDNet_Finetune/CContrast/data', logs_dir='log/market/resnet101_ibn_cion', pooling_type='gem', use_hard=True)
==========
==> Load unlabeled dataset
=> Market1501 loaded
Dataset statistics:
  ----------------------------------------
  subset   | # ids | # images | # cameras
  ----------------------------------------
  train    |   751 |    12936 |         6
  query    |   750 |     3368 |         6
  gallery  |   751 |    15913 |         6
  ----------------------------------------
pooling_type: gem
checkpoint loaded!
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.525)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 21.804372549057007
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 587 clusters
Epoch: [0][20/200]	Time 0.424 (0.866)	Data 0.001 (0.098)	Loss 3.948 (2.746)
Epoch: [0][40/200]	Time 0.505 (0.673)	Data 0.001 (0.079)	Loss 3.055 (3.303)
Epoch: [0][60/200]	Time 0.403 (0.608)	Data 0.001 (0.075)	Loss 2.980 (3.201)
Epoch: [0][80/200]	Time 0.422 (0.576)	Data 0.001 (0.072)	Loss 2.704 (3.128)
Epoch: [0][100/200]	Time 0.399 (0.558)	Data 0.001 (0.071)	Loss 2.847 (3.043)
Epoch: [0][120/200]	Time 0.401 (0.545)	Data 0.000 (0.070)	Loss 2.648 (2.960)
Epoch: [0][140/200]	Time 0.406 (0.536)	Data 0.000 (0.069)	Loss 2.356 (2.890)
Epoch: [0][160/200]	Time 0.414 (0.531)	Data 0.000 (0.069)	Loss 2.447 (2.829)
Epoch: [0][180/200]	Time 0.400 (0.525)	Data 0.000 (0.068)	Loss 2.351 (2.781)
Epoch: [0][200/200]	Time 0.421 (0.528)	Data 0.001 (0.074)	Loss 2.025 (2.731)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.196)	Data 0.000 (0.019)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.016430616378784
==> Statistics for epoch 1: 618 clusters
Epoch: [1][20/200]	Time 1.755 (0.520)	Data 1.313 (0.099)	Loss 2.231 (0.582)
Epoch: [1][40/200]	Time 0.403 (0.502)	Data 0.001 (0.082)	Loss 2.020 (1.376)
Epoch: [1][60/200]	Time 0.508 (0.495)	Data 0.001 (0.074)	Loss 1.984 (1.636)
Epoch: [1][80/200]	Time 0.411 (0.489)	Data 0.001 (0.071)	Loss 2.269 (1.765)
Epoch: [1][100/200]	Time 0.408 (0.489)	Data 0.001 (0.071)	Loss 2.618 (1.832)
Epoch: [1][120/200]	Time 0.408 (0.490)	Data 0.001 (0.070)	Loss 1.715 (1.837)
Epoch: [1][140/200]	Time 0.530 (0.489)	Data 0.001 (0.070)	Loss 1.852 (1.850)
Epoch: [1][160/200]	Time 0.408 (0.488)	Data 0.001 (0.069)	Loss 2.033 (1.877)
Epoch: [1][180/200]	Time 0.400 (0.488)	Data 0.001 (0.068)	Loss 2.151 (1.867)
Epoch: [1][200/200]	Time 0.517 (0.488)	Data 0.001 (0.068)	Loss 1.796 (1.867)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.185)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.122270345687866
==> Statistics for epoch 2: 584 clusters
Epoch: [2][20/200]	Time 0.419 (0.531)	Data 0.001 (0.106)	Loss 2.014 (0.554)
Epoch: [2][40/200]	Time 0.406 (0.515)	Data 0.001 (0.089)	Loss 1.896 (1.195)
Epoch: [2][60/200]	Time 0.522 (0.508)	Data 0.001 (0.083)	Loss 1.470 (1.393)
Epoch: [2][80/200]	Time 0.414 (0.501)	Data 0.001 (0.078)	Loss 1.751 (1.505)
Epoch: [2][100/200]	Time 0.394 (0.498)	Data 0.001 (0.076)	Loss 1.282 (1.561)
Epoch: [2][120/200]	Time 0.403 (0.495)	Data 0.000 (0.074)	Loss 1.443 (1.610)
Epoch: [2][140/200]	Time 0.400 (0.493)	Data 0.000 (0.072)	Loss 1.565 (1.638)
Epoch: [2][160/200]	Time 0.406 (0.491)	Data 0.000 (0.071)	Loss 2.529 (1.658)
Epoch: [2][180/200]	Time 0.411 (0.491)	Data 0.000 (0.070)	Loss 1.931 (1.655)
Epoch: [2][200/200]	Time 0.562 (0.497)	Data 0.001 (0.076)	Loss 1.465 (1.652)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.178)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.1749165058136
==> Statistics for epoch 3: 592 clusters
Epoch: [3][20/200]	Time 0.423 (0.515)	Data 0.001 (0.090)	Loss 1.467 (0.429)
Epoch: [3][40/200]	Time 0.398 (0.495)	Data 0.001 (0.075)	Loss 1.592 (1.004)
Epoch: [3][60/200]	Time 0.395 (0.490)	Data 0.001 (0.070)	Loss 1.962 (1.227)
Epoch: [3][80/200]	Time 0.391 (0.489)	Data 0.001 (0.069)	Loss 1.243 (1.337)
Epoch: [3][100/200]	Time 0.414 (0.485)	Data 0.001 (0.066)	Loss 1.959 (1.401)
Epoch: [3][120/200]	Time 0.404 (0.485)	Data 0.000 (0.066)	Loss 1.565 (1.434)
Epoch: [3][140/200]	Time 0.403 (0.485)	Data 0.000 (0.065)	Loss 1.392 (1.457)
Epoch: [3][160/200]	Time 0.411 (0.485)	Data 0.000 (0.064)	Loss 1.560 (1.476)
Epoch: [3][180/200]	Time 0.415 (0.486)	Data 0.000 (0.065)	Loss 1.824 (1.487)
Epoch: [3][200/200]	Time 0.410 (0.492)	Data 0.001 (0.071)	Loss 1.598 (1.487)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.231 (0.178)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.912503719329834
==> Statistics for epoch 4: 606 clusters
Epoch: [4][20/200]	Time 0.456 (0.521)	Data 0.001 (0.094)	Loss 1.660 (0.426)
Epoch: [4][40/200]	Time 0.391 (0.503)	Data 0.001 (0.078)	Loss 1.519 (0.935)
Epoch: [4][60/200]	Time 0.410 (0.496)	Data 0.001 (0.072)	Loss 1.666 (1.128)
Epoch: [4][80/200]	Time 0.394 (0.493)	Data 0.001 (0.070)	Loss 1.558 (1.221)
Epoch: [4][100/200]	Time 0.471 (0.491)	Data 0.001 (0.069)	Loss 1.542 (1.276)
Epoch: [4][120/200]	Time 0.395 (0.487)	Data 0.000 (0.067)	Loss 1.479 (1.316)
Epoch: [4][140/200]	Time 0.403 (0.487)	Data 0.000 (0.066)	Loss 1.593 (1.341)
Epoch: [4][160/200]	Time 0.392 (0.485)	Data 0.000 (0.065)	Loss 1.390 (1.350)
Epoch: [4][180/200]	Time 0.407 (0.485)	Data 0.000 (0.066)	Loss 1.196 (1.359)
Epoch: [4][200/200]	Time 0.415 (0.493)	Data 0.001 (0.072)	Loss 1.269 (1.365)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.175)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.278106927871704
==> Statistics for epoch 5: 602 clusters
Epoch: [5][20/200]	Time 0.414 (0.520)	Data 0.001 (0.099)	Loss 1.353 (0.363)
Epoch: [5][40/200]	Time 0.408 (0.504)	Data 0.001 (0.079)	Loss 1.256 (0.861)
Epoch: [5][60/200]	Time 0.521 (0.498)	Data 0.001 (0.073)	Loss 1.224 (1.019)
Epoch: [5][80/200]	Time 0.393 (0.492)	Data 0.001 (0.071)	Loss 1.749 (1.100)
Epoch: [5][100/200]	Time 0.402 (0.490)	Data 0.001 (0.068)	Loss 1.153 (1.174)
Epoch: [5][120/200]	Time 0.399 (0.489)	Data 0.000 (0.068)	Loss 1.147 (1.213)
Epoch: [5][140/200]	Time 0.408 (0.487)	Data 0.000 (0.067)	Loss 1.434 (1.245)
Epoch: [5][160/200]	Time 0.402 (0.486)	Data 0.000 (0.066)	Loss 1.340 (1.256)
Epoch: [5][180/200]	Time 0.403 (0.485)	Data 0.000 (0.065)	Loss 1.490 (1.263)
Epoch: [5][200/200]	Time 0.413 (0.491)	Data 0.001 (0.071)	Loss 1.283 (1.271)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.177)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.87418842315674
==> Statistics for epoch 6: 606 clusters
Epoch: [6][20/200]	Time 0.409 (0.527)	Data 0.001 (0.097)	Loss 1.011 (0.300)
Epoch: [6][40/200]	Time 0.415 (0.503)	Data 0.001 (0.078)	Loss 1.057 (0.754)
Epoch: [6][60/200]	Time 0.415 (0.502)	Data 0.001 (0.077)	Loss 1.377 (0.920)
Epoch: [6][80/200]	Time 0.391 (0.498)	Data 0.001 (0.075)	Loss 1.016 (1.013)
Epoch: [6][100/200]	Time 0.392 (0.495)	Data 0.001 (0.073)	Loss 1.319 (1.065)
Epoch: [6][120/200]	Time 0.406 (0.494)	Data 0.000 (0.072)	Loss 1.033 (1.103)
Epoch: [6][140/200]	Time 0.400 (0.492)	Data 0.000 (0.071)	Loss 1.592 (1.124)
Epoch: [6][160/200]	Time 0.494 (0.492)	Data 0.000 (0.072)	Loss 1.348 (1.146)
Epoch: [6][180/200]	Time 0.412 (0.490)	Data 0.000 (0.071)	Loss 1.083 (1.160)
Epoch: [6][200/200]	Time 0.436 (0.498)	Data 0.001 (0.078)	Loss 1.412 (1.168)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.183)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.581111907958984
==> Statistics for epoch 7: 614 clusters
Epoch: [7][20/200]	Time 1.692 (0.512)	Data 1.235 (0.099)	Loss 1.063 (0.260)
Epoch: [7][40/200]	Time 0.552 (0.503)	Data 0.001 (0.082)	Loss 1.365 (0.767)
Epoch: [7][60/200]	Time 0.415 (0.497)	Data 0.001 (0.077)	Loss 1.103 (0.922)
Epoch: [7][80/200]	Time 0.411 (0.494)	Data 0.001 (0.075)	Loss 1.329 (1.003)
Epoch: [7][100/200]	Time 0.437 (0.494)	Data 0.001 (0.073)	Loss 1.071 (1.038)
Epoch: [7][120/200]	Time 0.400 (0.490)	Data 0.001 (0.072)	Loss 0.962 (1.058)
Epoch: [7][140/200]	Time 0.409 (0.489)	Data 0.001 (0.071)	Loss 1.308 (1.082)
Epoch: [7][160/200]	Time 0.399 (0.489)	Data 0.001 (0.070)	Loss 1.316 (1.101)
Epoch: [7][180/200]	Time 0.503 (0.490)	Data 0.001 (0.070)	Loss 1.212 (1.111)
Epoch: [7][200/200]	Time 0.392 (0.489)	Data 0.001 (0.070)	Loss 1.588 (1.126)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.184)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.46791911125183
==> Statistics for epoch 8: 614 clusters
Epoch: [8][20/200]	Time 1.859 (0.538)	Data 1.413 (0.110)	Loss 1.126 (0.229)
Epoch: [8][40/200]	Time 0.402 (0.509)	Data 0.001 (0.089)	Loss 1.031 (0.669)
Epoch: [8][60/200]	Time 0.410 (0.504)	Data 0.001 (0.084)	Loss 1.250 (0.844)
Epoch: [8][80/200]	Time 0.402 (0.502)	Data 0.001 (0.081)	Loss 1.234 (0.932)
Epoch: [8][100/200]	Time 0.413 (0.499)	Data 0.001 (0.078)	Loss 0.953 (0.972)
Epoch: [8][120/200]	Time 0.419 (0.499)	Data 0.001 (0.078)	Loss 0.867 (1.005)
Epoch: [8][140/200]	Time 0.418 (0.499)	Data 0.001 (0.077)	Loss 1.075 (1.029)
Epoch: [8][160/200]	Time 0.402 (0.497)	Data 0.001 (0.076)	Loss 1.434 (1.047)
Epoch: [8][180/200]	Time 0.500 (0.497)	Data 0.001 (0.076)	Loss 1.412 (1.060)
Epoch: [8][200/200]	Time 0.416 (0.497)	Data 0.001 (0.076)	Loss 0.949 (1.066)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.182)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.53443670272827
==> Statistics for epoch 9: 611 clusters
Epoch: [9][20/200]	Time 1.880 (0.523)	Data 1.325 (0.102)	Loss 1.331 (0.232)
Epoch: [9][40/200]	Time 0.417 (0.500)	Data 0.001 (0.081)	Loss 1.196 (0.663)
Epoch: [9][60/200]	Time 0.418 (0.496)	Data 0.001 (0.075)	Loss 1.034 (0.806)
Epoch: [9][80/200]	Time 0.407 (0.494)	Data 0.001 (0.072)	Loss 1.341 (0.887)
Epoch: [9][100/200]	Time 0.404 (0.492)	Data 0.001 (0.070)	Loss 1.241 (0.930)
Epoch: [9][120/200]	Time 0.411 (0.493)	Data 0.001 (0.070)	Loss 0.706 (0.962)
Epoch: [9][140/200]	Time 0.396 (0.491)	Data 0.001 (0.069)	Loss 1.228 (0.975)
Epoch: [9][160/200]	Time 0.409 (0.491)	Data 0.001 (0.068)	Loss 1.111 (0.996)
Epoch: [9][180/200]	Time 0.418 (0.491)	Data 0.001 (0.068)	Loss 1.191 (1.013)
Epoch: [9][200/200]	Time 0.398 (0.490)	Data 0.001 (0.068)	Loss 1.270 (1.019)
Extract Features: [50/76]	Time 0.156 (0.182)	Data 0.000 (0.023)	
Mean AP: 92.3%

 * Finished epoch   9  model mAP: 92.3%  best: 92.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.182)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.553534507751465
==> Statistics for epoch 10: 613 clusters
Epoch: [10][20/200]	Time 1.868 (0.558)	Data 1.380 (0.103)	Loss 0.912 (0.193)
Epoch: [10][40/200]	Time 0.540 (0.525)	Data 0.001 (0.084)	Loss 1.010 (0.607)
Epoch: [10][60/200]	Time 0.428 (0.511)	Data 0.001 (0.075)	Loss 1.642 (0.776)
Epoch: [10][80/200]	Time 0.409 (0.506)	Data 0.001 (0.074)	Loss 1.346 (0.856)
Epoch: [10][100/200]	Time 0.404 (0.504)	Data 0.001 (0.072)	Loss 1.091 (0.900)
Epoch: [10][120/200]	Time 0.396 (0.501)	Data 0.001 (0.071)	Loss 1.267 (0.935)
Epoch: [10][140/200]	Time 0.409 (0.500)	Data 0.000 (0.071)	Loss 1.034 (0.964)
Epoch: [10][160/200]	Time 0.408 (0.499)	Data 0.001 (0.070)	Loss 1.042 (0.972)
Epoch: [10][180/200]	Time 0.493 (0.499)	Data 0.000 (0.070)	Loss 1.194 (0.977)
Epoch: [10][200/200]	Time 0.421 (0.498)	Data 0.000 (0.070)	Loss 0.864 (0.980)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.147 (0.180)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.730428218841553
==> Statistics for epoch 11: 611 clusters
Epoch: [11][20/200]	Time 1.796 (0.541)	Data 1.352 (0.119)	Loss 0.860 (0.197)
Epoch: [11][40/200]	Time 0.409 (0.516)	Data 0.001 (0.098)	Loss 1.107 (0.602)
Epoch: [11][60/200]	Time 0.407 (0.510)	Data 0.001 (0.090)	Loss 0.978 (0.736)
Epoch: [11][80/200]	Time 0.517 (0.505)	Data 0.001 (0.085)	Loss 1.039 (0.821)
Epoch: [11][100/200]	Time 0.407 (0.502)	Data 0.001 (0.083)	Loss 0.697 (0.866)
Epoch: [11][120/200]	Time 0.396 (0.502)	Data 0.001 (0.082)	Loss 1.299 (0.895)
Epoch: [11][140/200]	Time 0.519 (0.501)	Data 0.001 (0.080)	Loss 1.087 (0.914)
Epoch: [11][160/200]	Time 0.399 (0.499)	Data 0.001 (0.079)	Loss 0.987 (0.919)
Epoch: [11][180/200]	Time 0.403 (0.498)	Data 0.001 (0.078)	Loss 1.075 (0.934)
Epoch: [11][200/200]	Time 0.403 (0.497)	Data 0.001 (0.077)	Loss 0.765 (0.934)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.180)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.554524183273315
==> Statistics for epoch 12: 617 clusters
Epoch: [12][20/200]	Time 1.907 (0.542)	Data 1.412 (0.110)	Loss 1.167 (0.212)
Epoch: [12][40/200]	Time 0.423 (0.517)	Data 0.001 (0.091)	Loss 0.840 (0.568)
Epoch: [12][60/200]	Time 0.411 (0.510)	Data 0.000 (0.084)	Loss 1.041 (0.711)
Epoch: [12][80/200]	Time 0.400 (0.509)	Data 0.001 (0.083)	Loss 1.025 (0.790)
Epoch: [12][100/200]	Time 0.517 (0.508)	Data 0.001 (0.082)	Loss 0.830 (0.828)
Epoch: [12][120/200]	Time 0.416 (0.506)	Data 0.001 (0.081)	Loss 0.956 (0.851)
Epoch: [12][140/200]	Time 0.415 (0.504)	Data 0.001 (0.079)	Loss 1.264 (0.877)
Epoch: [12][160/200]	Time 0.404 (0.503)	Data 0.001 (0.078)	Loss 1.219 (0.896)
Epoch: [12][180/200]	Time 0.402 (0.503)	Data 0.001 (0.078)	Loss 0.858 (0.903)
Epoch: [12][200/200]	Time 0.419 (0.502)	Data 0.001 (0.078)	Loss 0.800 (0.906)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.236 (0.185)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.32970142364502
==> Statistics for epoch 13: 614 clusters
Epoch: [13][20/200]	Time 1.937 (0.524)	Data 1.475 (0.109)	Loss 0.924 (0.193)
Epoch: [13][40/200]	Time 0.418 (0.505)	Data 0.001 (0.087)	Loss 0.931 (0.553)
Epoch: [13][60/200]	Time 0.417 (0.500)	Data 0.001 (0.080)	Loss 1.081 (0.674)
Epoch: [13][80/200]	Time 0.402 (0.495)	Data 0.001 (0.076)	Loss 1.003 (0.752)
Epoch: [13][100/200]	Time 0.510 (0.494)	Data 0.001 (0.073)	Loss 1.049 (0.793)
Epoch: [13][120/200]	Time 0.410 (0.492)	Data 0.001 (0.072)	Loss 0.930 (0.825)
Epoch: [13][140/200]	Time 0.410 (0.493)	Data 0.001 (0.073)	Loss 0.966 (0.842)
Epoch: [13][160/200]	Time 0.410 (0.494)	Data 0.001 (0.073)	Loss 0.845 (0.859)
Epoch: [13][180/200]	Time 0.404 (0.494)	Data 0.001 (0.073)	Loss 0.932 (0.869)
Epoch: [13][200/200]	Time 0.392 (0.493)	Data 0.001 (0.072)	Loss 0.824 (0.878)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.181)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.425162076950073
==> Statistics for epoch 14: 610 clusters
Epoch: [14][20/200]	Time 1.964 (0.546)	Data 1.521 (0.118)	Loss 0.738 (0.165)
Epoch: [14][40/200]	Time 0.404 (0.522)	Data 0.001 (0.097)	Loss 1.074 (0.549)
Epoch: [14][60/200]	Time 0.406 (0.511)	Data 0.000 (0.089)	Loss 0.973 (0.677)
Epoch: [14][80/200]	Time 0.402 (0.507)	Data 0.000 (0.085)	Loss 1.125 (0.762)
Epoch: [14][100/200]	Time 0.412 (0.506)	Data 0.000 (0.082)	Loss 0.732 (0.797)
Epoch: [14][120/200]	Time 0.528 (0.504)	Data 0.000 (0.080)	Loss 0.987 (0.821)
Epoch: [14][140/200]	Time 0.407 (0.503)	Data 0.000 (0.080)	Loss 1.149 (0.838)
Epoch: [14][160/200]	Time 0.399 (0.502)	Data 0.000 (0.079)	Loss 1.101 (0.842)
Epoch: [14][180/200]	Time 0.554 (0.500)	Data 0.000 (0.077)	Loss 0.906 (0.857)
Epoch: [14][200/200]	Time 0.415 (0.499)	Data 0.000 (0.076)	Loss 0.923 (0.858)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.182)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.9878830909729
==> Statistics for epoch 15: 610 clusters
Epoch: [15][20/200]	Time 1.875 (0.528)	Data 1.415 (0.104)	Loss 0.875 (0.182)
Epoch: [15][40/200]	Time 0.419 (0.508)	Data 0.001 (0.086)	Loss 1.173 (0.562)
Epoch: [15][60/200]	Time 0.414 (0.503)	Data 0.001 (0.079)	Loss 1.135 (0.697)
Epoch: [15][80/200]	Time 0.414 (0.499)	Data 0.001 (0.076)	Loss 0.884 (0.759)
Epoch: [15][100/200]	Time 0.418 (0.495)	Data 0.001 (0.073)	Loss 0.800 (0.803)
Epoch: [15][120/200]	Time 0.414 (0.495)	Data 0.001 (0.073)	Loss 0.830 (0.821)
Epoch: [15][140/200]	Time 0.413 (0.494)	Data 0.001 (0.071)	Loss 0.944 (0.835)
Epoch: [15][160/200]	Time 0.518 (0.495)	Data 0.001 (0.072)	Loss 0.702 (0.846)
Epoch: [15][180/200]	Time 0.408 (0.495)	Data 0.001 (0.071)	Loss 0.884 (0.844)
Epoch: [15][200/200]	Time 0.407 (0.494)	Data 0.001 (0.071)	Loss 0.887 (0.844)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.182)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.244033813476562
==> Statistics for epoch 16: 614 clusters
Epoch: [16][20/200]	Time 1.796 (0.533)	Data 1.339 (0.106)	Loss 1.220 (0.194)
Epoch: [16][40/200]	Time 0.421 (0.511)	Data 0.001 (0.088)	Loss 0.784 (0.517)
Epoch: [16][60/200]	Time 0.411 (0.505)	Data 0.001 (0.082)	Loss 1.010 (0.638)
Epoch: [16][80/200]	Time 0.567 (0.504)	Data 0.002 (0.080)	Loss 0.821 (0.718)
Epoch: [16][100/200]	Time 0.399 (0.499)	Data 0.001 (0.076)	Loss 0.968 (0.756)
Epoch: [16][120/200]	Time 0.408 (0.498)	Data 0.001 (0.074)	Loss 0.735 (0.787)
Epoch: [16][140/200]	Time 0.526 (0.497)	Data 0.001 (0.072)	Loss 0.830 (0.806)
Epoch: [16][160/200]	Time 0.413 (0.495)	Data 0.001 (0.072)	Loss 1.038 (0.819)
Epoch: [16][180/200]	Time 0.401 (0.494)	Data 0.001 (0.071)	Loss 1.037 (0.827)
Epoch: [16][200/200]	Time 0.415 (0.493)	Data 0.001 (0.070)	Loss 0.784 (0.831)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.179)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.659132957458496
==> Statistics for epoch 17: 610 clusters
Epoch: [17][20/200]	Time 1.774 (0.526)	Data 1.312 (0.102)	Loss 0.896 (0.168)
Epoch: [17][40/200]	Time 0.411 (0.507)	Data 0.001 (0.084)	Loss 0.646 (0.513)
Epoch: [17][60/200]	Time 0.553 (0.504)	Data 0.001 (0.081)	Loss 0.622 (0.634)
Epoch: [17][80/200]	Time 0.413 (0.500)	Data 0.001 (0.077)	Loss 0.882 (0.688)
Epoch: [17][100/200]	Time 0.406 (0.500)	Data 0.001 (0.075)	Loss 0.673 (0.714)
Epoch: [17][120/200]	Time 0.396 (0.497)	Data 0.001 (0.073)	Loss 1.215 (0.743)
Epoch: [17][140/200]	Time 0.401 (0.496)	Data 0.001 (0.073)	Loss 1.027 (0.762)
Epoch: [17][160/200]	Time 0.399 (0.496)	Data 0.001 (0.072)	Loss 0.594 (0.773)
Epoch: [17][180/200]	Time 0.414 (0.495)	Data 0.001 (0.071)	Loss 0.794 (0.783)
Epoch: [17][200/200]	Time 0.411 (0.494)	Data 0.001 (0.070)	Loss 1.046 (0.789)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.181)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.707523822784424
==> Statistics for epoch 18: 613 clusters
Epoch: [18][20/200]	Time 1.779 (0.524)	Data 1.334 (0.101)	Loss 0.598 (0.158)
Epoch: [18][40/200]	Time 0.428 (0.508)	Data 0.001 (0.082)	Loss 0.608 (0.489)
Epoch: [18][60/200]	Time 0.541 (0.501)	Data 0.001 (0.075)	Loss 0.832 (0.606)
Epoch: [18][80/200]	Time 0.394 (0.496)	Data 0.001 (0.073)	Loss 0.689 (0.656)
Epoch: [18][100/200]	Time 0.400 (0.495)	Data 0.001 (0.071)	Loss 0.666 (0.690)
Epoch: [18][120/200]	Time 0.514 (0.493)	Data 0.001 (0.070)	Loss 0.928 (0.721)
Epoch: [18][140/200]	Time 0.401 (0.491)	Data 0.002 (0.069)	Loss 1.020 (0.733)
Epoch: [18][160/200]	Time 0.399 (0.491)	Data 0.001 (0.068)	Loss 0.643 (0.739)
Epoch: [18][180/200]	Time 0.408 (0.490)	Data 0.001 (0.068)	Loss 0.858 (0.754)
Epoch: [18][200/200]	Time 0.417 (0.490)	Data 0.001 (0.068)	Loss 1.053 (0.765)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.181)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.461902618408203
==> Statistics for epoch 19: 615 clusters
Epoch: [19][20/200]	Time 1.814 (0.529)	Data 1.385 (0.108)	Loss 0.767 (0.154)
Epoch: [19][40/200]	Time 0.410 (0.507)	Data 0.001 (0.087)	Loss 0.615 (0.480)
Epoch: [19][60/200]	Time 0.403 (0.503)	Data 0.001 (0.080)	Loss 0.768 (0.579)
Epoch: [19][80/200]	Time 0.397 (0.499)	Data 0.001 (0.077)	Loss 0.847 (0.629)
Epoch: [19][100/200]	Time 0.411 (0.496)	Data 0.001 (0.075)	Loss 0.844 (0.667)
Epoch: [19][120/200]	Time 0.404 (0.495)	Data 0.001 (0.074)	Loss 0.824 (0.688)
Epoch: [19][140/200]	Time 0.403 (0.493)	Data 0.001 (0.073)	Loss 0.869 (0.709)
Epoch: [19][160/200]	Time 0.485 (0.492)	Data 0.001 (0.071)	Loss 1.305 (0.725)
Epoch: [19][180/200]	Time 0.398 (0.491)	Data 0.001 (0.070)	Loss 0.858 (0.733)
Epoch: [19][200/200]	Time 0.404 (0.490)	Data 0.001 (0.070)	Loss 1.342 (0.740)
Extract Features: [50/76]	Time 0.239 (0.182)	Data 0.000 (0.023)	
Mean AP: 93.4%

 * Finished epoch  19  model mAP: 93.4%  best: 93.4% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.179)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.284464836120605
==> Statistics for epoch 20: 613 clusters
Epoch: [20][20/200]	Time 1.791 (0.523)	Data 1.210 (0.099)	Loss 0.858 (0.156)
Epoch: [20][40/200]	Time 0.432 (0.496)	Data 0.001 (0.081)	Loss 0.792 (0.464)
Epoch: [20][60/200]	Time 0.418 (0.493)	Data 0.001 (0.074)	Loss 0.838 (0.562)
Epoch: [20][80/200]	Time 0.396 (0.491)	Data 0.001 (0.071)	Loss 0.877 (0.616)
Epoch: [20][100/200]	Time 0.520 (0.489)	Data 0.001 (0.069)	Loss 0.651 (0.633)
Epoch: [20][120/200]	Time 0.397 (0.488)	Data 0.001 (0.069)	Loss 0.902 (0.660)
Epoch: [20][140/200]	Time 0.399 (0.488)	Data 0.001 (0.068)	Loss 0.700 (0.675)
Epoch: [20][160/200]	Time 0.407 (0.488)	Data 0.001 (0.068)	Loss 0.604 (0.686)
Epoch: [20][180/200]	Time 0.414 (0.488)	Data 0.001 (0.068)	Loss 1.022 (0.698)
Epoch: [20][200/200]	Time 0.408 (0.488)	Data 0.001 (0.068)	Loss 0.594 (0.708)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.228 (0.179)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.341634035110474
==> Statistics for epoch 21: 615 clusters
Epoch: [21][20/200]	Time 1.920 (0.521)	Data 1.448 (0.108)	Loss 0.618 (0.128)
Epoch: [21][40/200]	Time 0.415 (0.502)	Data 0.002 (0.086)	Loss 0.669 (0.406)
Epoch: [21][60/200]	Time 0.417 (0.497)	Data 0.001 (0.079)	Loss 0.924 (0.524)
Epoch: [21][80/200]	Time 0.410 (0.496)	Data 0.001 (0.077)	Loss 0.807 (0.572)
Epoch: [21][100/200]	Time 0.535 (0.495)	Data 0.001 (0.075)	Loss 0.550 (0.615)
Epoch: [21][120/200]	Time 0.392 (0.495)	Data 0.001 (0.074)	Loss 0.679 (0.631)
Epoch: [21][140/200]	Time 0.407 (0.494)	Data 0.001 (0.074)	Loss 0.811 (0.644)
Epoch: [21][160/200]	Time 0.412 (0.494)	Data 0.001 (0.073)	Loss 0.531 (0.657)
Epoch: [21][180/200]	Time 0.399 (0.493)	Data 0.001 (0.073)	Loss 0.603 (0.666)
Epoch: [21][200/200]	Time 0.395 (0.492)	Data 0.001 (0.072)	Loss 0.577 (0.668)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.177)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.277154445648193
==> Statistics for epoch 22: 609 clusters
Epoch: [22][20/200]	Time 1.677 (0.519)	Data 1.237 (0.101)	Loss 0.723 (0.139)
Epoch: [22][40/200]	Time 0.412 (0.502)	Data 0.001 (0.081)	Loss 0.451 (0.427)
Epoch: [22][60/200]	Time 0.515 (0.494)	Data 0.001 (0.074)	Loss 0.626 (0.538)
Epoch: [22][80/200]	Time 0.410 (0.489)	Data 0.001 (0.071)	Loss 0.788 (0.589)
Epoch: [22][100/200]	Time 0.399 (0.490)	Data 0.001 (0.071)	Loss 0.669 (0.621)
Epoch: [22][120/200]	Time 0.511 (0.490)	Data 0.001 (0.071)	Loss 0.926 (0.651)
Epoch: [22][140/200]	Time 0.414 (0.489)	Data 0.001 (0.070)	Loss 0.679 (0.666)
Epoch: [22][160/200]	Time 0.404 (0.488)	Data 0.001 (0.069)	Loss 0.801 (0.679)
Epoch: [22][180/200]	Time 0.410 (0.488)	Data 0.001 (0.068)	Loss 0.844 (0.690)
Epoch: [22][200/200]	Time 0.411 (0.487)	Data 0.001 (0.068)	Loss 0.960 (0.691)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.179)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.896573305130005
==> Statistics for epoch 23: 615 clusters
Epoch: [23][20/200]	Time 1.728 (0.518)	Data 1.283 (0.101)	Loss 0.608 (0.144)
Epoch: [23][40/200]	Time 0.407 (0.502)	Data 0.001 (0.083)	Loss 0.667 (0.432)
Epoch: [23][60/200]	Time 0.417 (0.496)	Data 0.001 (0.076)	Loss 0.645 (0.525)
Epoch: [23][80/200]	Time 0.500 (0.494)	Data 0.001 (0.074)	Loss 0.842 (0.580)
Epoch: [23][100/200]	Time 0.401 (0.491)	Data 0.001 (0.072)	Loss 0.816 (0.625)
Epoch: [23][120/200]	Time 0.400 (0.490)	Data 0.001 (0.071)	Loss 0.813 (0.643)
Epoch: [23][140/200]	Time 0.409 (0.488)	Data 0.001 (0.069)	Loss 0.595 (0.648)
Epoch: [23][160/200]	Time 0.409 (0.487)	Data 0.001 (0.069)	Loss 0.853 (0.659)
Epoch: [23][180/200]	Time 0.413 (0.487)	Data 0.001 (0.068)	Loss 0.648 (0.671)
Epoch: [23][200/200]	Time 0.409 (0.486)	Data 0.001 (0.067)	Loss 0.913 (0.677)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.177)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.549686431884766
==> Statistics for epoch 24: 615 clusters
Epoch: [24][20/200]	Time 1.902 (0.530)	Data 1.438 (0.104)	Loss 0.706 (0.142)
Epoch: [24][40/200]	Time 0.408 (0.511)	Data 0.001 (0.088)	Loss 0.676 (0.421)
Epoch: [24][60/200]	Time 0.414 (0.508)	Data 0.001 (0.082)	Loss 1.112 (0.536)
Epoch: [24][80/200]	Time 0.400 (0.504)	Data 0.002 (0.079)	Loss 0.894 (0.589)
Epoch: [24][100/200]	Time 0.418 (0.501)	Data 0.001 (0.077)	Loss 0.817 (0.625)
Epoch: [24][120/200]	Time 0.411 (0.498)	Data 0.001 (0.074)	Loss 0.678 (0.651)
Epoch: [24][140/200]	Time 0.405 (0.496)	Data 0.001 (0.073)	Loss 0.509 (0.664)
Epoch: [24][160/200]	Time 0.490 (0.496)	Data 0.001 (0.073)	Loss 0.981 (0.670)
Epoch: [24][180/200]	Time 0.408 (0.494)	Data 0.006 (0.071)	Loss 0.695 (0.675)
Epoch: [24][200/200]	Time 0.399 (0.494)	Data 0.001 (0.071)	Loss 0.471 (0.685)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.182)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.279381036758423
==> Statistics for epoch 25: 614 clusters
Epoch: [25][20/200]	Time 1.745 (0.523)	Data 1.258 (0.099)	Loss 0.448 (0.122)
Epoch: [25][40/200]	Time 0.409 (0.514)	Data 0.001 (0.090)	Loss 0.844 (0.436)
Epoch: [25][60/200]	Time 0.537 (0.507)	Data 0.001 (0.084)	Loss 0.757 (0.533)
Epoch: [25][80/200]	Time 0.412 (0.502)	Data 0.001 (0.079)	Loss 0.863 (0.577)
Epoch: [25][100/200]	Time 0.406 (0.499)	Data 0.001 (0.077)	Loss 0.891 (0.605)
Epoch: [25][120/200]	Time 0.426 (0.496)	Data 0.001 (0.075)	Loss 0.943 (0.628)
Epoch: [25][140/200]	Time 0.397 (0.494)	Data 0.001 (0.073)	Loss 1.006 (0.641)
Epoch: [25][160/200]	Time 0.400 (0.494)	Data 0.001 (0.073)	Loss 0.753 (0.654)
Epoch: [25][180/200]	Time 0.405 (0.493)	Data 0.001 (0.072)	Loss 0.675 (0.666)
Epoch: [25][200/200]	Time 0.401 (0.494)	Data 0.001 (0.073)	Loss 0.674 (0.674)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.188)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.108891010284424
==> Statistics for epoch 26: 615 clusters
Epoch: [26][20/200]	Time 2.002 (0.536)	Data 1.562 (0.113)	Loss 0.656 (0.138)
Epoch: [26][40/200]	Time 0.408 (0.507)	Data 0.001 (0.090)	Loss 0.723 (0.443)
Epoch: [26][60/200]	Time 0.415 (0.499)	Data 0.001 (0.080)	Loss 0.772 (0.551)
Epoch: [26][80/200]	Time 0.401 (0.497)	Data 0.001 (0.078)	Loss 0.899 (0.603)
Epoch: [26][100/200]	Time 0.402 (0.497)	Data 0.001 (0.078)	Loss 0.856 (0.641)
Epoch: [26][120/200]	Time 0.413 (0.495)	Data 0.001 (0.076)	Loss 0.877 (0.659)
Epoch: [26][140/200]	Time 0.416 (0.494)	Data 0.001 (0.074)	Loss 0.611 (0.667)
Epoch: [26][160/200]	Time 0.393 (0.493)	Data 0.001 (0.073)	Loss 0.483 (0.676)
Epoch: [26][180/200]	Time 0.493 (0.492)	Data 0.001 (0.071)	Loss 0.607 (0.685)
Epoch: [26][200/200]	Time 0.416 (0.491)	Data 0.001 (0.071)	Loss 0.664 (0.690)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.179)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.588603973388672
==> Statistics for epoch 27: 614 clusters
Epoch: [27][20/200]	Time 1.776 (0.545)	Data 1.340 (0.118)	Loss 0.584 (0.127)
Epoch: [27][40/200]	Time 0.406 (0.521)	Data 0.001 (0.097)	Loss 0.649 (0.402)
Epoch: [27][60/200]	Time 0.407 (0.510)	Data 0.001 (0.089)	Loss 0.624 (0.501)
Epoch: [27][80/200]	Time 0.404 (0.507)	Data 0.001 (0.086)	Loss 0.809 (0.563)
Epoch: [27][100/200]	Time 0.512 (0.504)	Data 0.001 (0.082)	Loss 0.751 (0.588)
Epoch: [27][120/200]	Time 0.415 (0.502)	Data 0.001 (0.081)	Loss 0.913 (0.624)
Epoch: [27][140/200]	Time 0.404 (0.502)	Data 0.001 (0.080)	Loss 0.917 (0.645)
Epoch: [27][160/200]	Time 0.539 (0.501)	Data 0.001 (0.079)	Loss 0.613 (0.660)
Epoch: [27][180/200]	Time 0.405 (0.500)	Data 0.001 (0.078)	Loss 0.978 (0.672)
Epoch: [27][200/200]	Time 0.412 (0.499)	Data 0.001 (0.078)	Loss 0.591 (0.673)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.184)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.539620637893677
==> Statistics for epoch 28: 613 clusters
Epoch: [28][20/200]	Time 1.912 (0.536)	Data 1.473 (0.112)	Loss 0.877 (0.133)
Epoch: [28][40/200]	Time 0.405 (0.516)	Data 0.001 (0.093)	Loss 0.695 (0.400)
Epoch: [28][60/200]	Time 0.400 (0.508)	Data 0.001 (0.087)	Loss 0.780 (0.508)
Epoch: [28][80/200]	Time 0.411 (0.506)	Data 0.001 (0.084)	Loss 0.795 (0.565)
Epoch: [28][100/200]	Time 0.413 (0.505)	Data 0.001 (0.082)	Loss 0.725 (0.607)
Epoch: [28][120/200]	Time 0.526 (0.502)	Data 0.001 (0.081)	Loss 0.820 (0.632)
Epoch: [28][140/200]	Time 0.397 (0.501)	Data 0.001 (0.080)	Loss 0.850 (0.644)
Epoch: [28][160/200]	Time 0.421 (0.501)	Data 0.001 (0.079)	Loss 1.018 (0.659)
Epoch: [28][180/200]	Time 0.528 (0.500)	Data 0.001 (0.077)	Loss 1.011 (0.670)
Epoch: [28][200/200]	Time 0.410 (0.499)	Data 0.001 (0.077)	Loss 0.544 (0.671)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.183)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.76088047027588
==> Statistics for epoch 29: 611 clusters
Epoch: [29][20/200]	Time 1.813 (0.528)	Data 1.360 (0.104)	Loss 0.789 (0.136)
Epoch: [29][40/200]	Time 0.545 (0.510)	Data 0.000 (0.086)	Loss 0.866 (0.434)
Epoch: [29][60/200]	Time 0.399 (0.503)	Data 0.000 (0.083)	Loss 0.875 (0.530)
Epoch: [29][80/200]	Time 0.406 (0.500)	Data 0.000 (0.079)	Loss 0.614 (0.579)
Epoch: [29][100/200]	Time 0.406 (0.499)	Data 0.000 (0.078)	Loss 0.878 (0.592)
Epoch: [29][120/200]	Time 0.411 (0.498)	Data 0.000 (0.077)	Loss 0.968 (0.615)
Epoch: [29][140/200]	Time 0.422 (0.499)	Data 0.000 (0.076)	Loss 0.650 (0.628)
Epoch: [29][160/200]	Time 0.497 (0.499)	Data 0.000 (0.076)	Loss 0.655 (0.637)
Epoch: [29][180/200]	Time 0.401 (0.498)	Data 0.000 (0.076)	Loss 0.965 (0.657)
Epoch: [29][200/200]	Time 0.409 (0.497)	Data 0.000 (0.075)	Loss 0.647 (0.663)
Extract Features: [50/76]	Time 0.142 (0.183)	Data 0.000 (0.026)	
Mean AP: 93.7%

 * Finished epoch  29  model mAP: 93.7%  best: 93.7% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.180)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.42333149909973
==> Statistics for epoch 30: 611 clusters
Epoch: [30][20/200]	Time 1.834 (0.532)	Data 1.371 (0.109)	Loss 0.836 (0.130)
Epoch: [30][40/200]	Time 0.428 (0.514)	Data 0.000 (0.091)	Loss 0.878 (0.399)
Epoch: [30][60/200]	Time 0.417 (0.507)	Data 0.000 (0.083)	Loss 0.585 (0.488)
Epoch: [30][80/200]	Time 0.525 (0.507)	Data 0.001 (0.082)	Loss 0.810 (0.552)
Epoch: [30][100/200]	Time 0.408 (0.503)	Data 0.000 (0.079)	Loss 0.534 (0.578)
Epoch: [30][120/200]	Time 0.412 (0.503)	Data 0.000 (0.078)	Loss 0.611 (0.610)
Epoch: [30][140/200]	Time 0.529 (0.502)	Data 0.000 (0.077)	Loss 0.699 (0.627)
Epoch: [30][160/200]	Time 0.413 (0.501)	Data 0.001 (0.076)	Loss 0.612 (0.645)
Epoch: [30][180/200]	Time 0.397 (0.500)	Data 0.000 (0.075)	Loss 0.867 (0.651)
Epoch: [30][200/200]	Time 0.415 (0.500)	Data 0.000 (0.075)	Loss 0.822 (0.659)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.184)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.76608896255493
==> Statistics for epoch 31: 614 clusters
Epoch: [31][20/200]	Time 1.865 (0.532)	Data 1.406 (0.110)	Loss 0.610 (0.134)
Epoch: [31][40/200]	Time 0.405 (0.516)	Data 0.001 (0.092)	Loss 0.613 (0.414)
Epoch: [31][60/200]	Time 0.514 (0.509)	Data 0.001 (0.084)	Loss 0.795 (0.520)
Epoch: [31][80/200]	Time 0.400 (0.503)	Data 0.001 (0.080)	Loss 0.620 (0.574)
Epoch: [31][100/200]	Time 0.412 (0.499)	Data 0.001 (0.076)	Loss 1.024 (0.609)
Epoch: [31][120/200]	Time 0.514 (0.498)	Data 0.001 (0.076)	Loss 0.546 (0.624)
Epoch: [31][140/200]	Time 0.409 (0.497)	Data 0.001 (0.075)	Loss 0.841 (0.640)
Epoch: [31][160/200]	Time 0.423 (0.497)	Data 0.001 (0.076)	Loss 0.875 (0.656)
Epoch: [31][180/200]	Time 0.407 (0.495)	Data 0.001 (0.074)	Loss 0.745 (0.663)
Epoch: [31][200/200]	Time 0.392 (0.493)	Data 0.001 (0.073)	Loss 0.667 (0.671)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.180)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.051193952560425
==> Statistics for epoch 32: 614 clusters
Epoch: [32][20/200]	Time 1.745 (0.528)	Data 1.301 (0.104)	Loss 0.489 (0.116)
Epoch: [32][40/200]	Time 0.421 (0.504)	Data 0.001 (0.083)	Loss 0.808 (0.419)
Epoch: [32][60/200]	Time 0.404 (0.503)	Data 0.001 (0.080)	Loss 0.778 (0.525)
Epoch: [32][80/200]	Time 0.534 (0.503)	Data 0.001 (0.080)	Loss 0.692 (0.577)
Epoch: [32][100/200]	Time 0.391 (0.499)	Data 0.001 (0.076)	Loss 1.042 (0.607)
Epoch: [32][120/200]	Time 0.405 (0.499)	Data 0.001 (0.077)	Loss 0.660 (0.621)
Epoch: [32][140/200]	Time 0.413 (0.497)	Data 0.001 (0.074)	Loss 0.849 (0.633)
Epoch: [32][160/200]	Time 0.404 (0.496)	Data 0.001 (0.073)	Loss 0.574 (0.645)
Epoch: [32][180/200]	Time 0.413 (0.496)	Data 0.001 (0.073)	Loss 0.802 (0.653)
Epoch: [32][200/200]	Time 0.498 (0.496)	Data 0.001 (0.073)	Loss 0.460 (0.655)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.183)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.362398147583008
==> Statistics for epoch 33: 611 clusters
Epoch: [33][20/200]	Time 1.795 (0.531)	Data 1.310 (0.099)	Loss 0.934 (0.142)
Epoch: [33][40/200]	Time 0.515 (0.512)	Data 0.001 (0.084)	Loss 0.706 (0.407)
Epoch: [33][60/200]	Time 0.427 (0.500)	Data 0.001 (0.076)	Loss 0.803 (0.509)
Epoch: [33][80/200]	Time 0.408 (0.498)	Data 0.001 (0.073)	Loss 0.794 (0.557)
Epoch: [33][100/200]	Time 0.407 (0.496)	Data 0.001 (0.072)	Loss 0.693 (0.590)
Epoch: [33][120/200]	Time 0.411 (0.493)	Data 0.001 (0.070)	Loss 0.397 (0.622)
Epoch: [33][140/200]	Time 0.399 (0.494)	Data 0.001 (0.070)	Loss 0.660 (0.634)
Epoch: [33][160/200]	Time 0.412 (0.495)	Data 0.001 (0.070)	Loss 0.821 (0.645)
Epoch: [33][180/200]	Time 0.412 (0.493)	Data 0.001 (0.070)	Loss 0.576 (0.655)
Epoch: [33][200/200]	Time 0.401 (0.493)	Data 0.001 (0.069)	Loss 0.900 (0.661)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.178)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.164122104644775
==> Statistics for epoch 34: 613 clusters
Epoch: [34][20/200]	Time 1.715 (0.525)	Data 1.254 (0.104)	Loss 0.654 (0.126)
Epoch: [34][40/200]	Time 0.395 (0.506)	Data 0.001 (0.087)	Loss 0.749 (0.413)
Epoch: [34][60/200]	Time 0.526 (0.499)	Data 0.001 (0.079)	Loss 0.923 (0.524)
Epoch: [34][80/200]	Time 0.414 (0.494)	Data 0.001 (0.075)	Loss 0.674 (0.578)
Epoch: [34][100/200]	Time 0.403 (0.495)	Data 0.001 (0.074)	Loss 0.674 (0.608)
Epoch: [34][120/200]	Time 0.407 (0.494)	Data 0.001 (0.072)	Loss 0.838 (0.634)
Epoch: [34][140/200]	Time 0.406 (0.493)	Data 0.001 (0.072)	Loss 0.836 (0.650)
Epoch: [34][160/200]	Time 0.410 (0.492)	Data 0.001 (0.071)	Loss 0.631 (0.654)
Epoch: [34][180/200]	Time 0.413 (0.492)	Data 0.001 (0.071)	Loss 0.695 (0.661)
Epoch: [34][200/200]	Time 0.408 (0.492)	Data 0.001 (0.071)	Loss 0.488 (0.667)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.179)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.703001499176025
==> Statistics for epoch 35: 611 clusters
Epoch: [35][20/200]	Time 1.730 (0.520)	Data 1.286 (0.099)	Loss 0.618 (0.116)
Epoch: [35][40/200]	Time 0.406 (0.504)	Data 0.001 (0.084)	Loss 0.538 (0.368)
Epoch: [35][60/200]	Time 0.412 (0.505)	Data 0.001 (0.082)	Loss 0.460 (0.476)
Epoch: [35][80/200]	Time 0.404 (0.501)	Data 0.001 (0.078)	Loss 0.725 (0.534)
Epoch: [35][100/200]	Time 0.518 (0.499)	Data 0.001 (0.075)	Loss 0.644 (0.571)
Epoch: [35][120/200]	Time 0.416 (0.496)	Data 0.001 (0.074)	Loss 0.747 (0.591)
Epoch: [35][140/200]	Time 0.399 (0.494)	Data 0.001 (0.072)	Loss 0.666 (0.615)
Epoch: [35][160/200]	Time 0.411 (0.494)	Data 0.001 (0.071)	Loss 0.740 (0.626)
Epoch: [35][180/200]	Time 0.395 (0.493)	Data 0.001 (0.071)	Loss 0.839 (0.641)
Epoch: [35][200/200]	Time 0.406 (0.492)	Data 0.001 (0.071)	Loss 0.501 (0.645)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.179)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.856761932373047
==> Statistics for epoch 36: 614 clusters
Epoch: [36][20/200]	Time 1.714 (0.517)	Data 1.257 (0.101)	Loss 0.419 (0.108)
Epoch: [36][40/200]	Time 0.417 (0.497)	Data 0.001 (0.085)	Loss 0.529 (0.381)
Epoch: [36][60/200]	Time 0.400 (0.493)	Data 0.001 (0.080)	Loss 0.809 (0.482)
Epoch: [36][80/200]	Time 0.398 (0.493)	Data 0.001 (0.079)	Loss 0.557 (0.539)
Epoch: [36][100/200]	Time 0.419 (0.496)	Data 0.001 (0.079)	Loss 0.660 (0.576)
Epoch: [36][120/200]	Time 0.408 (0.495)	Data 0.001 (0.079)	Loss 0.692 (0.606)
Epoch: [36][140/200]	Time 0.518 (0.495)	Data 0.001 (0.077)	Loss 0.648 (0.623)
Epoch: [36][160/200]	Time 0.407 (0.495)	Data 0.001 (0.076)	Loss 0.760 (0.634)
Epoch: [36][180/200]	Time 0.412 (0.495)	Data 0.001 (0.076)	Loss 0.815 (0.641)
Epoch: [36][200/200]	Time 0.410 (0.494)	Data 0.001 (0.076)	Loss 0.975 (0.650)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.143 (0.185)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.6505069732666
==> Statistics for epoch 37: 612 clusters
Epoch: [37][20/200]	Time 1.641 (0.520)	Data 1.181 (0.099)	Loss 0.849 (0.138)
Epoch: [37][40/200]	Time 0.420 (0.508)	Data 0.001 (0.087)	Loss 0.687 (0.419)
Epoch: [37][60/200]	Time 0.401 (0.504)	Data 0.001 (0.081)	Loss 0.887 (0.527)
Epoch: [37][80/200]	Time 0.517 (0.502)	Data 0.001 (0.079)	Loss 0.521 (0.569)
Epoch: [37][100/200]	Time 0.394 (0.499)	Data 0.001 (0.079)	Loss 0.601 (0.598)
Epoch: [37][120/200]	Time 0.415 (0.498)	Data 0.001 (0.078)	Loss 0.966 (0.622)
Epoch: [37][140/200]	Time 0.408 (0.496)	Data 0.001 (0.076)	Loss 0.727 (0.637)
Epoch: [37][160/200]	Time 0.412 (0.495)	Data 0.001 (0.075)	Loss 0.579 (0.649)
Epoch: [37][180/200]	Time 0.403 (0.494)	Data 0.001 (0.074)	Loss 0.789 (0.653)
Epoch: [37][200/200]	Time 0.407 (0.494)	Data 0.001 (0.073)	Loss 0.906 (0.662)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.145 (0.180)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.48750352859497
==> Statistics for epoch 38: 609 clusters
Epoch: [38][20/200]	Time 1.731 (0.538)	Data 1.253 (0.107)	Loss 0.710 (0.129)
Epoch: [38][40/200]	Time 0.412 (0.519)	Data 0.001 (0.091)	Loss 0.641 (0.389)
Epoch: [38][60/200]	Time 0.402 (0.510)	Data 0.001 (0.084)	Loss 0.848 (0.486)
Epoch: [38][80/200]	Time 0.409 (0.505)	Data 0.001 (0.080)	Loss 0.441 (0.540)
Epoch: [38][100/200]	Time 0.408 (0.501)	Data 0.001 (0.078)	Loss 0.938 (0.576)
Epoch: [38][120/200]	Time 0.548 (0.502)	Data 0.001 (0.078)	Loss 0.531 (0.605)
Epoch: [38][140/200]	Time 0.400 (0.501)	Data 0.001 (0.078)	Loss 0.521 (0.619)
Epoch: [38][160/200]	Time 0.398 (0.500)	Data 0.001 (0.077)	Loss 0.700 (0.632)
Epoch: [38][180/200]	Time 0.407 (0.499)	Data 0.001 (0.077)	Loss 0.774 (0.642)
Epoch: [38][200/200]	Time 0.406 (0.498)	Data 0.001 (0.076)	Loss 0.598 (0.647)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.184)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.805599689483643
==> Statistics for epoch 39: 609 clusters
Epoch: [39][20/200]	Time 1.832 (0.529)	Data 1.376 (0.103)	Loss 0.632 (0.114)
Epoch: [39][40/200]	Time 0.420 (0.503)	Data 0.001 (0.082)	Loss 0.633 (0.375)
Epoch: [39][60/200]	Time 0.417 (0.503)	Data 0.001 (0.080)	Loss 0.620 (0.471)
Epoch: [39][80/200]	Time 0.411 (0.501)	Data 0.001 (0.077)	Loss 0.562 (0.520)
Epoch: [39][100/200]	Time 0.566 (0.501)	Data 0.001 (0.078)	Loss 0.724 (0.552)
Epoch: [39][120/200]	Time 0.394 (0.499)	Data 0.001 (0.077)	Loss 0.710 (0.572)
Epoch: [39][140/200]	Time 0.410 (0.498)	Data 0.001 (0.076)	Loss 0.848 (0.594)
Epoch: [39][160/200]	Time 0.411 (0.498)	Data 0.001 (0.076)	Loss 0.598 (0.606)
Epoch: [39][180/200]	Time 0.424 (0.497)	Data 0.001 (0.075)	Loss 0.546 (0.617)
Epoch: [39][200/200]	Time 0.398 (0.497)	Data 0.001 (0.074)	Loss 0.583 (0.625)
Extract Features: [50/76]	Time 0.150 (0.181)	Data 0.000 (0.023)	
Mean AP: 93.7%

 * Finished epoch  39  model mAP: 93.7%  best: 93.7% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.181)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.93198037147522
==> Statistics for epoch 40: 612 clusters
Epoch: [40][20/200]	Time 1.854 (0.530)	Data 1.407 (0.108)	Loss 0.684 (0.126)
Epoch: [40][40/200]	Time 0.405 (0.508)	Data 0.001 (0.090)	Loss 0.807 (0.394)
Epoch: [40][60/200]	Time 0.409 (0.502)	Data 0.003 (0.082)	Loss 0.812 (0.512)
Epoch: [40][80/200]	Time 0.511 (0.499)	Data 0.001 (0.078)	Loss 0.522 (0.558)
Epoch: [40][100/200]	Time 0.409 (0.495)	Data 0.001 (0.075)	Loss 0.858 (0.584)
Epoch: [40][120/200]	Time 0.407 (0.495)	Data 0.001 (0.074)	Loss 0.621 (0.604)
Epoch: [40][140/200]	Time 0.514 (0.495)	Data 0.001 (0.073)	Loss 0.753 (0.615)
Epoch: [40][160/200]	Time 0.400 (0.495)	Data 0.000 (0.074)	Loss 0.614 (0.620)
Epoch: [40][180/200]	Time 0.405 (0.494)	Data 0.000 (0.072)	Loss 0.643 (0.628)
Epoch: [40][200/200]	Time 0.411 (0.494)	Data 0.000 (0.072)	Loss 0.912 (0.637)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.179)	Data 0.000 (0.020)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.48392391204834
==> Statistics for epoch 41: 611 clusters
Epoch: [41][20/200]	Time 1.731 (0.524)	Data 1.264 (0.106)	Loss 0.687 (0.127)
Epoch: [41][40/200]	Time 0.399 (0.508)	Data 0.001 (0.089)	Loss 0.691 (0.403)
Epoch: [41][60/200]	Time 0.414 (0.503)	Data 0.001 (0.082)	Loss 0.786 (0.509)
Epoch: [41][80/200]	Time 0.418 (0.501)	Data 0.001 (0.081)	Loss 0.597 (0.550)
Epoch: [41][100/200]	Time 0.517 (0.499)	Data 0.001 (0.078)	Loss 0.767 (0.572)
Epoch: [41][120/200]	Time 0.392 (0.497)	Data 0.001 (0.077)	Loss 0.464 (0.591)
Epoch: [41][140/200]	Time 0.417 (0.496)	Data 0.002 (0.076)	Loss 0.688 (0.609)
Epoch: [41][160/200]	Time 0.413 (0.496)	Data 0.001 (0.074)	Loss 0.737 (0.628)
Epoch: [41][180/200]	Time 0.408 (0.496)	Data 0.001 (0.074)	Loss 0.972 (0.637)
Epoch: [41][200/200]	Time 0.398 (0.496)	Data 0.001 (0.074)	Loss 0.739 (0.642)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.241 (0.186)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.190723657608032
==> Statistics for epoch 42: 610 clusters
Epoch: [42][20/200]	Time 1.695 (0.535)	Data 1.231 (0.113)	Loss 0.636 (0.119)
Epoch: [42][40/200]	Time 0.407 (0.516)	Data 0.001 (0.091)	Loss 0.605 (0.387)
Epoch: [42][60/200]	Time 0.415 (0.508)	Data 0.001 (0.082)	Loss 0.829 (0.501)
Epoch: [42][80/200]	Time 0.523 (0.502)	Data 0.001 (0.077)	Loss 0.608 (0.548)
Epoch: [42][100/200]	Time 0.414 (0.500)	Data 0.001 (0.077)	Loss 0.582 (0.579)
Epoch: [42][120/200]	Time 0.419 (0.498)	Data 0.001 (0.074)	Loss 0.525 (0.607)
Epoch: [42][140/200]	Time 0.410 (0.496)	Data 0.001 (0.072)	Loss 0.521 (0.617)
Epoch: [42][160/200]	Time 0.415 (0.495)	Data 0.001 (0.071)	Loss 0.552 (0.625)
Epoch: [42][180/200]	Time 0.441 (0.494)	Data 0.001 (0.070)	Loss 0.641 (0.632)
Epoch: [42][200/200]	Time 0.489 (0.495)	Data 0.001 (0.071)	Loss 0.774 (0.644)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.177)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.51796293258667
==> Statistics for epoch 43: 612 clusters
Epoch: [43][20/200]	Time 1.716 (0.517)	Data 1.234 (0.094)	Loss 0.707 (0.119)
Epoch: [43][40/200]	Time 0.411 (0.504)	Data 0.002 (0.081)	Loss 0.661 (0.384)
Epoch: [43][60/200]	Time 0.399 (0.502)	Data 0.001 (0.078)	Loss 0.663 (0.499)
Epoch: [43][80/200]	Time 0.499 (0.499)	Data 0.001 (0.076)	Loss 0.558 (0.550)
Epoch: [43][100/200]	Time 0.400 (0.496)	Data 0.001 (0.074)	Loss 0.753 (0.572)
Epoch: [43][120/200]	Time 0.396 (0.496)	Data 0.001 (0.073)	Loss 0.892 (0.592)
Epoch: [43][140/200]	Time 0.557 (0.496)	Data 0.002 (0.072)	Loss 0.534 (0.610)
Epoch: [43][160/200]	Time 0.402 (0.495)	Data 0.001 (0.071)	Loss 0.762 (0.625)
Epoch: [43][180/200]	Time 0.392 (0.495)	Data 0.001 (0.071)	Loss 0.625 (0.628)
Epoch: [43][200/200]	Time 0.415 (0.495)	Data 0.001 (0.071)	Loss 0.810 (0.640)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.178)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.23676633834839
==> Statistics for epoch 44: 612 clusters
Epoch: [44][20/200]	Time 1.939 (0.527)	Data 1.503 (0.117)	Loss 0.589 (0.111)
Epoch: [44][40/200]	Time 0.418 (0.513)	Data 0.001 (0.095)	Loss 0.726 (0.382)
Epoch: [44][60/200]	Time 0.393 (0.507)	Data 0.001 (0.087)	Loss 0.587 (0.486)
Epoch: [44][80/200]	Time 0.426 (0.500)	Data 0.006 (0.081)	Loss 0.998 (0.554)
Epoch: [44][100/200]	Time 0.407 (0.501)	Data 0.001 (0.080)	Loss 0.620 (0.571)
Epoch: [44][120/200]	Time 0.404 (0.499)	Data 0.001 (0.077)	Loss 0.617 (0.592)
Epoch: [44][140/200]	Time 0.613 (0.499)	Data 0.001 (0.077)	Loss 0.834 (0.607)
Epoch: [44][160/200]	Time 0.401 (0.498)	Data 0.001 (0.075)	Loss 0.744 (0.624)
Epoch: [44][180/200]	Time 0.402 (0.497)	Data 0.001 (0.074)	Loss 0.708 (0.631)
Epoch: [44][200/200]	Time 0.494 (0.495)	Data 0.001 (0.073)	Loss 0.497 (0.643)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.145 (0.183)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.23867392539978
==> Statistics for epoch 45: 613 clusters
Epoch: [45][20/200]	Time 1.924 (0.541)	Data 1.458 (0.113)	Loss 0.800 (0.136)
Epoch: [45][40/200]	Time 0.405 (0.519)	Data 0.001 (0.093)	Loss 0.617 (0.426)
Epoch: [45][60/200]	Time 0.405 (0.511)	Data 0.001 (0.084)	Loss 0.666 (0.509)
Epoch: [45][80/200]	Time 0.517 (0.505)	Data 0.001 (0.079)	Loss 0.838 (0.552)
Epoch: [45][100/200]	Time 0.429 (0.499)	Data 0.001 (0.076)	Loss 0.569 (0.573)
Epoch: [45][120/200]	Time 0.408 (0.500)	Data 0.001 (0.076)	Loss 0.717 (0.603)
Epoch: [45][140/200]	Time 0.415 (0.500)	Data 0.001 (0.075)	Loss 0.732 (0.615)
Epoch: [45][160/200]	Time 0.398 (0.498)	Data 0.001 (0.075)	Loss 0.688 (0.634)
Epoch: [45][180/200]	Time 0.413 (0.499)	Data 0.001 (0.075)	Loss 0.587 (0.643)
Epoch: [45][200/200]	Time 0.399 (0.498)	Data 0.001 (0.075)	Loss 0.803 (0.650)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.146 (0.178)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.681782722473145
==> Statistics for epoch 46: 610 clusters
Epoch: [46][20/200]	Time 1.898 (0.530)	Data 1.402 (0.108)	Loss 0.656 (0.133)
Epoch: [46][40/200]	Time 0.408 (0.515)	Data 0.001 (0.090)	Loss 0.461 (0.403)
Epoch: [46][60/200]	Time 0.433 (0.508)	Data 0.001 (0.081)	Loss 0.603 (0.508)
Epoch: [46][80/200]	Time 0.525 (0.501)	Data 0.001 (0.077)	Loss 0.646 (0.563)
Epoch: [46][100/200]	Time 0.413 (0.499)	Data 0.001 (0.076)	Loss 0.826 (0.580)
Epoch: [46][120/200]	Time 0.401 (0.499)	Data 0.001 (0.075)	Loss 0.611 (0.601)
Epoch: [46][140/200]	Time 0.423 (0.498)	Data 0.001 (0.075)	Loss 0.603 (0.607)
Epoch: [46][160/200]	Time 0.399 (0.497)	Data 0.001 (0.074)	Loss 0.854 (0.615)
Epoch: [46][180/200]	Time 0.397 (0.498)	Data 0.001 (0.075)	Loss 0.621 (0.617)
Epoch: [46][200/200]	Time 0.400 (0.497)	Data 0.001 (0.075)	Loss 0.686 (0.621)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.178)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.482229709625244
==> Statistics for epoch 47: 613 clusters
Epoch: [47][20/200]	Time 1.874 (0.532)	Data 1.412 (0.109)	Loss 0.603 (0.113)
Epoch: [47][40/200]	Time 0.531 (0.512)	Data 0.001 (0.085)	Loss 0.741 (0.388)
Epoch: [47][60/200]	Time 0.393 (0.505)	Data 0.001 (0.082)	Loss 0.722 (0.485)
Epoch: [47][80/200]	Time 0.412 (0.500)	Data 0.001 (0.077)	Loss 0.443 (0.546)
Epoch: [47][100/200]	Time 0.409 (0.497)	Data 0.001 (0.075)	Loss 0.812 (0.574)
Epoch: [47][120/200]	Time 0.419 (0.495)	Data 0.001 (0.073)	Loss 0.906 (0.597)
Epoch: [47][140/200]	Time 0.416 (0.495)	Data 0.001 (0.073)	Loss 0.602 (0.617)
Epoch: [47][160/200]	Time 0.411 (0.494)	Data 0.001 (0.072)	Loss 0.752 (0.622)
Epoch: [47][180/200]	Time 0.479 (0.494)	Data 0.001 (0.071)	Loss 0.700 (0.630)
Epoch: [47][200/200]	Time 0.412 (0.493)	Data 0.001 (0.070)	Loss 0.707 (0.636)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.150 (0.180)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.49197220802307
==> Statistics for epoch 48: 613 clusters
Epoch: [48][20/200]	Time 1.959 (0.538)	Data 1.339 (0.105)	Loss 0.796 (0.137)
Epoch: [48][40/200]	Time 0.413 (0.506)	Data 0.001 (0.084)	Loss 0.539 (0.399)
Epoch: [48][60/200]	Time 0.405 (0.505)	Data 0.001 (0.082)	Loss 0.557 (0.486)
Epoch: [48][80/200]	Time 0.410 (0.500)	Data 0.001 (0.077)	Loss 0.833 (0.544)
Epoch: [48][100/200]	Time 0.526 (0.498)	Data 0.001 (0.075)	Loss 0.680 (0.575)
Epoch: [48][120/200]	Time 0.416 (0.496)	Data 0.001 (0.074)	Loss 0.824 (0.600)
Epoch: [48][140/200]	Time 0.408 (0.495)	Data 0.001 (0.073)	Loss 0.969 (0.620)
Epoch: [48][160/200]	Time 0.414 (0.494)	Data 0.002 (0.072)	Loss 0.646 (0.629)
Epoch: [48][180/200]	Time 0.409 (0.493)	Data 0.001 (0.071)	Loss 0.445 (0.634)
Epoch: [48][200/200]	Time 0.398 (0.492)	Data 0.001 (0.070)	Loss 0.672 (0.640)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.179)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.450905323028564
==> Statistics for epoch 49: 611 clusters
Epoch: [49][20/200]	Time 1.846 (0.535)	Data 1.384 (0.107)	Loss 0.795 (0.144)
Epoch: [49][40/200]	Time 0.414 (0.514)	Data 0.001 (0.086)	Loss 0.700 (0.414)
Epoch: [49][60/200]	Time 0.535 (0.509)	Data 0.001 (0.082)	Loss 0.594 (0.509)
Epoch: [49][80/200]	Time 0.415 (0.502)	Data 0.001 (0.077)	Loss 0.648 (0.564)
Epoch: [49][100/200]	Time 0.420 (0.501)	Data 0.001 (0.076)	Loss 0.625 (0.593)
Epoch: [49][120/200]	Time 0.435 (0.500)	Data 0.001 (0.074)	Loss 0.714 (0.605)
Epoch: [49][140/200]	Time 0.417 (0.499)	Data 0.001 (0.073)	Loss 0.805 (0.616)
Epoch: [49][160/200]	Time 0.401 (0.497)	Data 0.001 (0.072)	Loss 0.687 (0.631)
Epoch: [49][180/200]	Time 0.406 (0.495)	Data 0.001 (0.071)	Loss 1.049 (0.639)
Epoch: [49][200/200]	Time 0.412 (0.495)	Data 0.001 (0.071)	Loss 0.585 (0.646)
Extract Features: [50/76]	Time 0.147 (0.179)	Data 0.000 (0.020)	
Mean AP: 93.8%

 * Finished epoch  49  model mAP: 93.8%  best: 93.8% *

==> Test with the best model:
=> Loaded checkpoint 'log/market/resnet101_ibn_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.143 (0.180)	Data 0.000 (0.021)	
Mean AP: 93.8%
CMC Scores:
  top-1          96.7%
  top-5          99.0%
  top-10         99.3%
Total running time:  1:51:52.403148
