==========
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='resnet101', pretrained_path='/home/ma-user/work/Projects/ReIDNets_Checkpoints_TransReID/ResNet101.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_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
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.496)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 21.767799377441406
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 606 clusters
Epoch: [0][20/200]	Time 0.382 (0.933)	Data 0.001 (0.104)	Loss 4.270 (2.943)
Epoch: [0][40/200]	Time 0.379 (0.696)	Data 0.001 (0.084)	Loss 3.042 (3.345)
Epoch: [0][60/200]	Time 0.380 (0.618)	Data 0.001 (0.080)	Loss 2.941 (3.264)
Epoch: [0][80/200]	Time 0.381 (0.582)	Data 0.000 (0.080)	Loss 3.471 (3.190)
Epoch: [0][100/200]	Time 0.384 (0.558)	Data 0.001 (0.077)	Loss 2.681 (3.112)
Epoch: [0][120/200]	Time 0.383 (0.542)	Data 0.000 (0.077)	Loss 2.838 (3.032)
Epoch: [0][140/200]	Time 0.381 (0.532)	Data 0.000 (0.077)	Loss 2.753 (2.963)
Epoch: [0][160/200]	Time 0.382 (0.523)	Data 0.000 (0.075)	Loss 1.763 (2.913)
Epoch: [0][180/200]	Time 0.379 (0.515)	Data 0.000 (0.074)	Loss 2.294 (2.861)
Epoch: [0][200/200]	Time 0.488 (0.517)	Data 0.001 (0.080)	Loss 2.103 (2.813)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.137 (0.178)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.487735271453857
==> Statistics for epoch 1: 611 clusters
Epoch: [1][20/200]	Time 1.852 (0.498)	Data 1.313 (0.104)	Loss 2.578 (0.847)
Epoch: [1][40/200]	Time 0.396 (0.485)	Data 0.001 (0.090)	Loss 2.201 (1.611)
Epoch: [1][60/200]	Time 0.386 (0.476)	Data 0.002 (0.082)	Loss 2.549 (1.877)
Epoch: [1][80/200]	Time 0.385 (0.471)	Data 0.001 (0.079)	Loss 2.429 (1.967)
Epoch: [1][100/200]	Time 0.387 (0.470)	Data 0.001 (0.076)	Loss 2.631 (2.019)
Epoch: [1][120/200]	Time 0.383 (0.469)	Data 0.001 (0.075)	Loss 2.104 (2.057)
Epoch: [1][140/200]	Time 0.385 (0.470)	Data 0.004 (0.075)	Loss 2.169 (2.073)
Epoch: [1][160/200]	Time 0.382 (0.470)	Data 0.001 (0.075)	Loss 2.071 (2.090)
Epoch: [1][180/200]	Time 0.386 (0.471)	Data 0.001 (0.076)	Loss 1.624 (2.083)
Epoch: [1][200/200]	Time 0.383 (0.469)	Data 0.001 (0.075)	Loss 1.894 (2.085)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.170)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.66223931312561
==> Statistics for epoch 2: 599 clusters
Epoch: [2][20/200]	Time 0.382 (0.501)	Data 0.001 (0.103)	Loss 1.660 (0.595)
Epoch: [2][40/200]	Time 0.381 (0.482)	Data 0.001 (0.084)	Loss 2.039 (1.306)
Epoch: [2][60/200]	Time 0.389 (0.474)	Data 0.001 (0.078)	Loss 2.222 (1.537)
Epoch: [2][80/200]	Time 0.486 (0.471)	Data 0.001 (0.076)	Loss 1.887 (1.661)
Epoch: [2][100/200]	Time 0.378 (0.469)	Data 0.001 (0.074)	Loss 2.197 (1.735)
Epoch: [2][120/200]	Time 0.383 (0.468)	Data 0.000 (0.073)	Loss 1.908 (1.764)
Epoch: [2][140/200]	Time 0.382 (0.468)	Data 0.000 (0.073)	Loss 2.050 (1.783)
Epoch: [2][160/200]	Time 0.379 (0.467)	Data 0.000 (0.073)	Loss 2.199 (1.805)
Epoch: [2][180/200]	Time 0.379 (0.466)	Data 0.000 (0.072)	Loss 1.681 (1.814)
Epoch: [2][200/200]	Time 0.390 (0.474)	Data 0.001 (0.079)	Loss 1.507 (1.820)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.169)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.44210910797119
==> Statistics for epoch 3: 596 clusters
Epoch: [3][20/200]	Time 0.383 (0.503)	Data 0.001 (0.109)	Loss 2.057 (0.601)
Epoch: [3][40/200]	Time 0.384 (0.481)	Data 0.001 (0.089)	Loss 1.878 (1.267)
Epoch: [3][60/200]	Time 0.385 (0.475)	Data 0.001 (0.082)	Loss 1.625 (1.510)
Epoch: [3][80/200]	Time 0.385 (0.471)	Data 0.001 (0.078)	Loss 1.973 (1.607)
Epoch: [3][100/200]	Time 0.389 (0.474)	Data 0.001 (0.080)	Loss 1.839 (1.642)
Epoch: [3][120/200]	Time 0.380 (0.472)	Data 0.000 (0.078)	Loss 1.690 (1.700)
Epoch: [3][140/200]	Time 0.381 (0.470)	Data 0.000 (0.076)	Loss 1.494 (1.703)
Epoch: [3][160/200]	Time 0.383 (0.469)	Data 0.000 (0.075)	Loss 1.212 (1.707)
Epoch: [3][180/200]	Time 0.382 (0.468)	Data 0.000 (0.075)	Loss 1.983 (1.716)
Epoch: [3][200/200]	Time 0.383 (0.476)	Data 0.001 (0.081)	Loss 1.872 (1.716)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.174)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.404477834701538
==> Statistics for epoch 4: 588 clusters
Epoch: [4][20/200]	Time 0.483 (0.518)	Data 0.001 (0.120)	Loss 1.563 (0.484)
Epoch: [4][40/200]	Time 0.391 (0.490)	Data 0.002 (0.093)	Loss 1.582 (1.075)
Epoch: [4][60/200]	Time 0.381 (0.485)	Data 0.001 (0.089)	Loss 1.804 (1.272)
Epoch: [4][80/200]	Time 0.382 (0.483)	Data 0.001 (0.087)	Loss 1.577 (1.396)
Epoch: [4][100/200]	Time 0.380 (0.482)	Data 0.001 (0.085)	Loss 1.652 (1.453)
Epoch: [4][120/200]	Time 0.381 (0.481)	Data 0.000 (0.085)	Loss 1.677 (1.485)
Epoch: [4][140/200]	Time 0.381 (0.480)	Data 0.000 (0.083)	Loss 1.569 (1.498)
Epoch: [4][160/200]	Time 0.387 (0.478)	Data 0.000 (0.083)	Loss 1.661 (1.516)
Epoch: [4][180/200]	Time 0.383 (0.477)	Data 0.000 (0.082)	Loss 1.459 (1.531)
Epoch: [4][200/200]	Time 0.385 (0.485)	Data 0.001 (0.089)	Loss 1.586 (1.540)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.176)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.263657808303833
==> Statistics for epoch 5: 585 clusters
Epoch: [5][20/200]	Time 0.386 (0.510)	Data 0.001 (0.110)	Loss 1.722 (0.429)
Epoch: [5][40/200]	Time 0.383 (0.494)	Data 0.001 (0.096)	Loss 1.389 (0.993)
Epoch: [5][60/200]	Time 0.381 (0.487)	Data 0.001 (0.091)	Loss 1.632 (1.212)
Epoch: [5][80/200]	Time 0.383 (0.483)	Data 0.001 (0.087)	Loss 1.557 (1.300)
Epoch: [5][100/200]	Time 0.382 (0.481)	Data 0.001 (0.085)	Loss 1.482 (1.338)
Epoch: [5][120/200]	Time 0.381 (0.480)	Data 0.000 (0.085)	Loss 1.518 (1.369)
Epoch: [5][140/200]	Time 0.382 (0.480)	Data 0.000 (0.084)	Loss 1.305 (1.400)
Epoch: [5][160/200]	Time 0.382 (0.480)	Data 0.000 (0.084)	Loss 1.422 (1.411)
Epoch: [5][180/200]	Time 0.380 (0.479)	Data 0.000 (0.084)	Loss 1.632 (1.430)
Epoch: [5][200/200]	Time 0.537 (0.488)	Data 0.001 (0.092)	Loss 1.588 (1.444)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.175)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.762505769729614
==> Statistics for epoch 6: 587 clusters
Epoch: [6][20/200]	Time 0.518 (0.524)	Data 0.001 (0.117)	Loss 1.214 (0.370)
Epoch: [6][40/200]	Time 0.378 (0.497)	Data 0.002 (0.100)	Loss 1.776 (0.900)
Epoch: [6][60/200]	Time 0.383 (0.488)	Data 0.001 (0.091)	Loss 1.472 (1.071)
Epoch: [6][80/200]	Time 0.384 (0.481)	Data 0.001 (0.085)	Loss 1.648 (1.165)
Epoch: [6][100/200]	Time 0.381 (0.477)	Data 0.001 (0.081)	Loss 1.246 (1.218)
Epoch: [6][120/200]	Time 0.384 (0.475)	Data 0.000 (0.080)	Loss 1.555 (1.251)
Epoch: [6][140/200]	Time 0.382 (0.473)	Data 0.000 (0.077)	Loss 1.410 (1.277)
Epoch: [6][160/200]	Time 0.381 (0.473)	Data 0.000 (0.077)	Loss 1.661 (1.298)
Epoch: [6][180/200]	Time 0.383 (0.472)	Data 0.000 (0.077)	Loss 1.285 (1.319)
Epoch: [6][200/200]	Time 0.387 (0.481)	Data 0.001 (0.085)	Loss 1.231 (1.331)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.180)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.545055389404297
==> Statistics for epoch 7: 604 clusters
Epoch: [7][20/200]	Time 0.394 (0.501)	Data 0.001 (0.107)	Loss 1.135 (0.368)
Epoch: [7][40/200]	Time 0.382 (0.479)	Data 0.001 (0.087)	Loss 1.546 (0.882)
Epoch: [7][60/200]	Time 0.385 (0.474)	Data 0.001 (0.082)	Loss 1.757 (1.062)
Epoch: [7][80/200]	Time 0.383 (0.471)	Data 0.001 (0.079)	Loss 1.414 (1.180)
Epoch: [7][100/200]	Time 0.460 (0.469)	Data 0.001 (0.076)	Loss 1.727 (1.218)
Epoch: [7][120/200]	Time 0.381 (0.468)	Data 0.000 (0.075)	Loss 1.527 (1.253)
Epoch: [7][140/200]	Time 0.380 (0.470)	Data 0.000 (0.076)	Loss 1.265 (1.271)
Epoch: [7][160/200]	Time 0.379 (0.470)	Data 0.000 (0.076)	Loss 1.986 (1.289)
Epoch: [7][180/200]	Time 0.380 (0.468)	Data 0.000 (0.075)	Loss 1.542 (1.295)
Epoch: [7][200/200]	Time 0.386 (0.476)	Data 0.001 (0.081)	Loss 1.216 (1.302)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.172)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.44887661933899
==> Statistics for epoch 8: 596 clusters
Epoch: [8][20/200]	Time 0.390 (0.508)	Data 0.001 (0.111)	Loss 1.067 (0.313)
Epoch: [8][40/200]	Time 0.386 (0.493)	Data 0.001 (0.097)	Loss 1.649 (0.767)
Epoch: [8][60/200]	Time 0.393 (0.487)	Data 0.001 (0.091)	Loss 1.372 (0.965)
Epoch: [8][80/200]	Time 0.386 (0.483)	Data 0.001 (0.088)	Loss 1.375 (1.033)
Epoch: [8][100/200]	Time 0.513 (0.483)	Data 0.001 (0.086)	Loss 1.108 (1.084)
Epoch: [8][120/200]	Time 0.387 (0.480)	Data 0.000 (0.084)	Loss 1.300 (1.114)
Epoch: [8][140/200]	Time 0.382 (0.479)	Data 0.000 (0.083)	Loss 1.448 (1.142)
Epoch: [8][160/200]	Time 0.382 (0.478)	Data 0.000 (0.082)	Loss 1.373 (1.156)
Epoch: [8][180/200]	Time 0.389 (0.478)	Data 0.000 (0.082)	Loss 1.272 (1.171)
Epoch: [8][200/200]	Time 0.388 (0.485)	Data 0.002 (0.089)	Loss 1.428 (1.179)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.175)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.27826976776123
==> Statistics for epoch 9: 595 clusters
Epoch: [9][20/200]	Time 0.381 (0.506)	Data 0.001 (0.111)	Loss 1.122 (0.295)
Epoch: [9][40/200]	Time 0.383 (0.486)	Data 0.001 (0.095)	Loss 1.586 (0.754)
Epoch: [9][60/200]	Time 0.384 (0.478)	Data 0.001 (0.086)	Loss 1.184 (0.906)
Epoch: [9][80/200]	Time 0.384 (0.475)	Data 0.001 (0.083)	Loss 1.437 (1.005)
Epoch: [9][100/200]	Time 0.381 (0.475)	Data 0.001 (0.083)	Loss 1.402 (1.056)
Epoch: [9][120/200]	Time 0.381 (0.472)	Data 0.000 (0.080)	Loss 1.497 (1.091)
Epoch: [9][140/200]	Time 0.383 (0.470)	Data 0.000 (0.078)	Loss 1.313 (1.117)
Epoch: [9][160/200]	Time 0.382 (0.469)	Data 0.000 (0.077)	Loss 1.391 (1.136)
Epoch: [9][180/200]	Time 0.381 (0.469)	Data 0.000 (0.077)	Loss 1.407 (1.156)
Epoch: [9][200/200]	Time 0.386 (0.475)	Data 0.001 (0.082)	Loss 1.838 (1.169)
Extract Features: [50/76]	Time 0.140 (0.176)	Data 0.000 (0.024)	
Mean AP: 90.2%

 * Finished epoch   9  model mAP: 90.2%  best: 90.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.172)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.946671962738037
==> Statistics for epoch 10: 599 clusters
Epoch: [10][20/200]	Time 0.393 (0.523)	Data 0.001 (0.116)	Loss 1.115 (0.308)
Epoch: [10][40/200]	Time 0.479 (0.494)	Data 0.001 (0.092)	Loss 1.308 (0.745)
Epoch: [10][60/200]	Time 0.383 (0.481)	Data 0.001 (0.084)	Loss 1.043 (0.893)
Epoch: [10][80/200]	Time 0.383 (0.476)	Data 0.001 (0.081)	Loss 1.289 (1.001)
Epoch: [10][100/200]	Time 0.384 (0.472)	Data 0.001 (0.078)	Loss 1.368 (1.058)
Epoch: [10][120/200]	Time 0.383 (0.470)	Data 0.000 (0.076)	Loss 1.198 (1.082)
Epoch: [10][140/200]	Time 0.382 (0.469)	Data 0.000 (0.075)	Loss 1.501 (1.109)
Epoch: [10][160/200]	Time 0.381 (0.469)	Data 0.000 (0.074)	Loss 1.397 (1.129)
Epoch: [10][180/200]	Time 0.398 (0.470)	Data 0.000 (0.074)	Loss 1.208 (1.145)
Epoch: [10][200/200]	Time 0.389 (0.476)	Data 0.001 (0.081)	Loss 1.727 (1.162)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.169)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.72742986679077
==> Statistics for epoch 11: 601 clusters
Epoch: [11][20/200]	Time 0.385 (0.505)	Data 0.001 (0.104)	Loss 1.099 (0.304)
Epoch: [11][40/200]	Time 0.379 (0.483)	Data 0.001 (0.086)	Loss 1.033 (0.737)
Epoch: [11][60/200]	Time 0.383 (0.475)	Data 0.001 (0.079)	Loss 1.225 (0.887)
Epoch: [11][80/200]	Time 0.384 (0.470)	Data 0.001 (0.076)	Loss 1.186 (0.952)
Epoch: [11][100/200]	Time 0.387 (0.468)	Data 0.001 (0.074)	Loss 1.294 (1.003)
Epoch: [11][120/200]	Time 0.464 (0.471)	Data 0.000 (0.075)	Loss 0.851 (1.025)
Epoch: [11][140/200]	Time 0.380 (0.469)	Data 0.000 (0.074)	Loss 1.349 (1.049)
Epoch: [11][160/200]	Time 0.383 (0.469)	Data 0.000 (0.074)	Loss 0.989 (1.072)
Epoch: [11][180/200]	Time 0.381 (0.468)	Data 0.000 (0.073)	Loss 1.319 (1.083)
Epoch: [11][200/200]	Time 0.386 (0.474)	Data 0.001 (0.079)	Loss 1.244 (1.093)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.170)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.594136953353882
==> Statistics for epoch 12: 605 clusters
Epoch: [12][20/200]	Time 0.393 (0.522)	Data 0.001 (0.125)	Loss 0.910 (0.318)
Epoch: [12][40/200]	Time 0.382 (0.496)	Data 0.001 (0.099)	Loss 1.265 (0.713)
Epoch: [12][60/200]	Time 0.391 (0.490)	Data 0.001 (0.093)	Loss 1.530 (0.886)
Epoch: [12][80/200]	Time 0.384 (0.487)	Data 0.001 (0.090)	Loss 1.171 (0.971)
Epoch: [12][100/200]	Time 0.472 (0.485)	Data 0.001 (0.087)	Loss 1.037 (1.019)
Epoch: [12][120/200]	Time 0.381 (0.484)	Data 0.000 (0.086)	Loss 1.273 (1.040)
Epoch: [12][140/200]	Time 0.382 (0.481)	Data 0.000 (0.085)	Loss 1.340 (1.065)
Epoch: [12][160/200]	Time 0.384 (0.481)	Data 0.000 (0.084)	Loss 1.105 (1.071)
Epoch: [12][180/200]	Time 0.380 (0.480)	Data 0.000 (0.084)	Loss 1.369 (1.075)
Epoch: [12][200/200]	Time 0.409 (0.487)	Data 0.001 (0.090)	Loss 0.842 (1.080)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.175)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.276807069778442
==> Statistics for epoch 13: 601 clusters
Epoch: [13][20/200]	Time 0.387 (0.524)	Data 0.001 (0.119)	Loss 0.908 (0.294)
Epoch: [13][40/200]	Time 0.383 (0.498)	Data 0.001 (0.099)	Loss 1.422 (0.690)
Epoch: [13][60/200]	Time 0.382 (0.491)	Data 0.001 (0.094)	Loss 1.199 (0.822)
Epoch: [13][80/200]	Time 0.385 (0.487)	Data 0.001 (0.089)	Loss 1.386 (0.900)
Epoch: [13][100/200]	Time 0.383 (0.484)	Data 0.001 (0.087)	Loss 1.045 (0.937)
Epoch: [13][120/200]	Time 0.381 (0.481)	Data 0.000 (0.084)	Loss 1.070 (0.957)
Epoch: [13][140/200]	Time 0.384 (0.480)	Data 0.000 (0.083)	Loss 0.982 (0.970)
Epoch: [13][160/200]	Time 0.381 (0.479)	Data 0.000 (0.082)	Loss 1.231 (0.983)
Epoch: [13][180/200]	Time 0.386 (0.478)	Data 0.000 (0.082)	Loss 1.075 (0.998)
Epoch: [13][200/200]	Time 0.392 (0.485)	Data 0.001 (0.089)	Loss 1.003 (1.008)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.179)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.676990032196045
==> Statistics for epoch 14: 611 clusters
Epoch: [14][20/200]	Time 1.955 (0.518)	Data 1.525 (0.121)	Loss 1.281 (0.220)
Epoch: [14][40/200]	Time 0.379 (0.496)	Data 0.001 (0.101)	Loss 1.147 (0.646)
Epoch: [14][60/200]	Time 0.387 (0.488)	Data 0.001 (0.093)	Loss 0.947 (0.791)
Epoch: [14][80/200]	Time 0.383 (0.480)	Data 0.001 (0.086)	Loss 0.998 (0.889)
Epoch: [14][100/200]	Time 0.382 (0.475)	Data 0.001 (0.082)	Loss 0.868 (0.916)
Epoch: [14][120/200]	Time 0.385 (0.475)	Data 0.001 (0.081)	Loss 1.165 (0.949)
Epoch: [14][140/200]	Time 0.394 (0.475)	Data 0.001 (0.081)	Loss 1.051 (0.972)
Epoch: [14][160/200]	Time 0.397 (0.474)	Data 0.001 (0.080)	Loss 0.981 (0.983)
Epoch: [14][180/200]	Time 0.392 (0.475)	Data 0.001 (0.080)	Loss 0.923 (0.997)
Epoch: [14][200/200]	Time 0.385 (0.475)	Data 0.001 (0.080)	Loss 1.044 (1.004)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.174)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.356167793273926
==> Statistics for epoch 15: 615 clusters
Epoch: [15][20/200]	Time 1.957 (0.507)	Data 1.499 (0.115)	Loss 1.316 (0.228)
Epoch: [15][40/200]	Time 0.391 (0.490)	Data 0.001 (0.097)	Loss 0.981 (0.615)
Epoch: [15][60/200]	Time 0.383 (0.482)	Data 0.001 (0.089)	Loss 1.021 (0.764)
Epoch: [15][80/200]	Time 0.385 (0.480)	Data 0.001 (0.087)	Loss 1.022 (0.842)
Epoch: [15][100/200]	Time 0.380 (0.479)	Data 0.001 (0.086)	Loss 1.016 (0.892)
Epoch: [15][120/200]	Time 0.377 (0.477)	Data 0.001 (0.084)	Loss 1.447 (0.918)
Epoch: [15][140/200]	Time 0.386 (0.477)	Data 0.001 (0.083)	Loss 1.162 (0.939)
Epoch: [15][160/200]	Time 0.383 (0.476)	Data 0.001 (0.083)	Loss 0.690 (0.951)
Epoch: [15][180/200]	Time 0.383 (0.477)	Data 0.001 (0.083)	Loss 0.952 (0.968)
Epoch: [15][200/200]	Time 0.485 (0.477)	Data 0.002 (0.083)	Loss 0.887 (0.977)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.175)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.319772481918335
==> Statistics for epoch 16: 611 clusters
Epoch: [16][20/200]	Time 1.799 (0.504)	Data 1.373 (0.108)	Loss 1.133 (0.215)
Epoch: [16][40/200]	Time 0.385 (0.484)	Data 0.001 (0.089)	Loss 0.853 (0.624)
Epoch: [16][60/200]	Time 0.390 (0.476)	Data 0.000 (0.081)	Loss 1.197 (0.777)
Epoch: [16][80/200]	Time 0.389 (0.471)	Data 0.001 (0.078)	Loss 0.894 (0.837)
Epoch: [16][100/200]	Time 0.380 (0.470)	Data 0.000 (0.076)	Loss 1.151 (0.871)
Epoch: [16][120/200]	Time 0.383 (0.469)	Data 0.000 (0.075)	Loss 1.226 (0.907)
Epoch: [16][140/200]	Time 0.384 (0.468)	Data 0.001 (0.075)	Loss 1.179 (0.928)
Epoch: [16][160/200]	Time 0.386 (0.469)	Data 0.000 (0.076)	Loss 1.334 (0.941)
Epoch: [16][180/200]	Time 0.384 (0.468)	Data 0.001 (0.074)	Loss 0.972 (0.955)
Epoch: [16][200/200]	Time 0.479 (0.467)	Data 0.000 (0.073)	Loss 1.121 (0.962)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.169)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.51160955429077
==> Statistics for epoch 17: 612 clusters
Epoch: [17][20/200]	Time 2.152 (0.528)	Data 1.744 (0.123)	Loss 0.840 (0.179)
Epoch: [17][40/200]	Time 0.383 (0.503)	Data 0.001 (0.102)	Loss 1.335 (0.585)
Epoch: [17][60/200]	Time 0.382 (0.491)	Data 0.001 (0.093)	Loss 1.172 (0.726)
Epoch: [17][80/200]	Time 0.383 (0.486)	Data 0.001 (0.089)	Loss 0.974 (0.815)
Epoch: [17][100/200]	Time 0.382 (0.483)	Data 0.001 (0.086)	Loss 1.077 (0.861)
Epoch: [17][120/200]	Time 0.505 (0.481)	Data 0.001 (0.085)	Loss 1.269 (0.879)
Epoch: [17][140/200]	Time 0.386 (0.479)	Data 0.000 (0.083)	Loss 0.971 (0.887)
Epoch: [17][160/200]	Time 0.386 (0.479)	Data 0.000 (0.083)	Loss 1.060 (0.897)
Epoch: [17][180/200]	Time 0.389 (0.479)	Data 0.000 (0.082)	Loss 0.904 (0.911)
Epoch: [17][200/200]	Time 0.385 (0.477)	Data 0.001 (0.081)	Loss 0.873 (0.918)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.223 (0.179)	Data 0.000 (0.028)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.509323358535767
==> Statistics for epoch 18: 611 clusters
Epoch: [18][20/200]	Time 2.028 (0.514)	Data 1.621 (0.123)	Loss 1.227 (0.214)
Epoch: [18][40/200]	Time 0.388 (0.495)	Data 0.001 (0.100)	Loss 0.597 (0.603)
Epoch: [18][60/200]	Time 0.390 (0.485)	Data 0.001 (0.090)	Loss 1.264 (0.733)
Epoch: [18][80/200]	Time 0.385 (0.478)	Data 0.001 (0.084)	Loss 0.774 (0.814)
Epoch: [18][100/200]	Time 0.390 (0.476)	Data 0.001 (0.081)	Loss 0.857 (0.854)
Epoch: [18][120/200]	Time 0.384 (0.478)	Data 0.001 (0.083)	Loss 1.405 (0.899)
Epoch: [18][140/200]	Time 0.382 (0.478)	Data 0.001 (0.082)	Loss 1.015 (0.916)
Epoch: [18][160/200]	Time 0.384 (0.477)	Data 0.001 (0.081)	Loss 1.162 (0.921)
Epoch: [18][180/200]	Time 0.381 (0.477)	Data 0.001 (0.081)	Loss 1.025 (0.935)
Epoch: [18][200/200]	Time 0.382 (0.477)	Data 0.001 (0.081)	Loss 1.034 (0.943)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.176)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.52089214324951
==> Statistics for epoch 19: 609 clusters
Epoch: [19][20/200]	Time 2.155 (0.521)	Data 1.613 (0.116)	Loss 0.823 (0.179)
Epoch: [19][40/200]	Time 0.382 (0.497)	Data 0.001 (0.099)	Loss 1.124 (0.542)
Epoch: [19][60/200]	Time 0.381 (0.489)	Data 0.001 (0.092)	Loss 1.141 (0.673)
Epoch: [19][80/200]	Time 0.391 (0.482)	Data 0.001 (0.087)	Loss 0.861 (0.737)
Epoch: [19][100/200]	Time 0.381 (0.479)	Data 0.001 (0.085)	Loss 1.229 (0.784)
Epoch: [19][120/200]	Time 0.381 (0.478)	Data 0.001 (0.083)	Loss 0.892 (0.823)
Epoch: [19][140/200]	Time 0.383 (0.477)	Data 0.001 (0.082)	Loss 0.953 (0.845)
Epoch: [19][160/200]	Time 0.381 (0.477)	Data 0.001 (0.081)	Loss 0.857 (0.859)
Epoch: [19][180/200]	Time 0.384 (0.476)	Data 0.001 (0.081)	Loss 1.006 (0.871)
Epoch: [19][200/200]	Time 0.513 (0.476)	Data 0.002 (0.080)	Loss 1.108 (0.881)
Extract Features: [50/76]	Time 0.141 (0.175)	Data 0.001 (0.026)	
Mean AP: 92.6%

 * Finished epoch  19  model mAP: 92.6%  best: 92.6% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.183)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.23557162284851
==> Statistics for epoch 20: 616 clusters
Epoch: [20][20/200]	Time 1.763 (0.501)	Data 1.347 (0.107)	Loss 0.684 (0.170)
Epoch: [20][40/200]	Time 0.383 (0.483)	Data 0.001 (0.089)	Loss 0.888 (0.533)
Epoch: [20][60/200]	Time 0.386 (0.480)	Data 0.001 (0.084)	Loss 0.951 (0.670)
Epoch: [20][80/200]	Time 0.383 (0.477)	Data 0.001 (0.081)	Loss 0.880 (0.725)
Epoch: [20][100/200]	Time 0.513 (0.480)	Data 0.001 (0.083)	Loss 0.699 (0.763)
Epoch: [20][120/200]	Time 0.384 (0.478)	Data 0.001 (0.081)	Loss 0.637 (0.783)
Epoch: [20][140/200]	Time 0.380 (0.476)	Data 0.001 (0.080)	Loss 1.069 (0.806)
Epoch: [20][160/200]	Time 0.386 (0.475)	Data 0.001 (0.078)	Loss 0.635 (0.823)
Epoch: [20][180/200]	Time 0.387 (0.473)	Data 0.001 (0.078)	Loss 0.872 (0.838)
Epoch: [20][200/200]	Time 0.381 (0.472)	Data 0.001 (0.077)	Loss 0.894 (0.850)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.171)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.5198392868042
==> Statistics for epoch 21: 613 clusters
Epoch: [21][20/200]	Time 1.960 (0.512)	Data 1.518 (0.116)	Loss 1.010 (0.170)
Epoch: [21][40/200]	Time 0.387 (0.491)	Data 0.001 (0.095)	Loss 0.932 (0.503)
Epoch: [21][60/200]	Time 0.385 (0.481)	Data 0.001 (0.086)	Loss 0.960 (0.643)
Epoch: [21][80/200]	Time 0.384 (0.479)	Data 0.001 (0.083)	Loss 0.783 (0.722)
Epoch: [21][100/200]	Time 0.388 (0.478)	Data 0.001 (0.083)	Loss 0.964 (0.759)
Epoch: [21][120/200]	Time 0.382 (0.477)	Data 0.002 (0.083)	Loss 1.414 (0.794)
Epoch: [21][140/200]	Time 0.385 (0.475)	Data 0.002 (0.081)	Loss 0.800 (0.820)
Epoch: [21][160/200]	Time 0.384 (0.473)	Data 0.001 (0.079)	Loss 0.925 (0.833)
Epoch: [21][180/200]	Time 0.383 (0.473)	Data 0.001 (0.079)	Loss 1.073 (0.843)
Epoch: [21][200/200]	Time 0.384 (0.473)	Data 0.001 (0.078)	Loss 0.697 (0.847)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.170)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.651070833206177
==> Statistics for epoch 22: 614 clusters
Epoch: [22][20/200]	Time 2.117 (0.516)	Data 1.694 (0.118)	Loss 0.748 (0.170)
Epoch: [22][40/200]	Time 0.492 (0.497)	Data 0.001 (0.100)	Loss 0.751 (0.532)
Epoch: [22][60/200]	Time 0.382 (0.490)	Data 0.001 (0.094)	Loss 1.043 (0.667)
Epoch: [22][80/200]	Time 0.507 (0.485)	Data 0.001 (0.089)	Loss 0.936 (0.721)
Epoch: [22][100/200]	Time 0.388 (0.484)	Data 0.002 (0.087)	Loss 1.110 (0.768)
Epoch: [22][120/200]	Time 0.391 (0.482)	Data 0.001 (0.085)	Loss 1.023 (0.789)
Epoch: [22][140/200]	Time 0.384 (0.482)	Data 0.001 (0.085)	Loss 1.133 (0.805)
Epoch: [22][160/200]	Time 0.383 (0.481)	Data 0.001 (0.084)	Loss 0.951 (0.822)
Epoch: [22][180/200]	Time 0.393 (0.480)	Data 0.001 (0.083)	Loss 1.168 (0.833)
Epoch: [22][200/200]	Time 0.383 (0.480)	Data 0.001 (0.083)	Loss 1.153 (0.842)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.173)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.182090997695923
==> Statistics for epoch 23: 615 clusters
Epoch: [23][20/200]	Time 2.198 (0.518)	Data 1.770 (0.124)	Loss 0.779 (0.161)
Epoch: [23][40/200]	Time 0.517 (0.495)	Data 0.001 (0.100)	Loss 1.068 (0.524)
Epoch: [23][60/200]	Time 0.380 (0.488)	Data 0.001 (0.093)	Loss 0.872 (0.644)
Epoch: [23][80/200]	Time 0.381 (0.485)	Data 0.001 (0.090)	Loss 0.924 (0.710)
Epoch: [23][100/200]	Time 0.382 (0.483)	Data 0.001 (0.089)	Loss 0.862 (0.749)
Epoch: [23][120/200]	Time 0.494 (0.481)	Data 0.000 (0.086)	Loss 0.926 (0.772)
Epoch: [23][140/200]	Time 0.382 (0.480)	Data 0.000 (0.085)	Loss 0.864 (0.780)
Epoch: [23][160/200]	Time 0.387 (0.479)	Data 0.000 (0.084)	Loss 0.719 (0.786)
Epoch: [23][180/200]	Time 0.383 (0.478)	Data 0.000 (0.083)	Loss 0.648 (0.802)
Epoch: [23][200/200]	Time 0.387 (0.478)	Data 0.001 (0.083)	Loss 0.944 (0.801)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.174)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.10768747329712
==> Statistics for epoch 24: 613 clusters
Epoch: [24][20/200]	Time 1.983 (0.510)	Data 1.561 (0.118)	Loss 0.923 (0.159)
Epoch: [24][40/200]	Time 0.380 (0.492)	Data 0.001 (0.099)	Loss 0.863 (0.477)
Epoch: [24][60/200]	Time 0.382 (0.487)	Data 0.001 (0.094)	Loss 0.951 (0.606)
Epoch: [24][80/200]	Time 0.383 (0.485)	Data 0.001 (0.091)	Loss 0.972 (0.680)
Epoch: [24][100/200]	Time 0.385 (0.482)	Data 0.001 (0.089)	Loss 0.861 (0.713)
Epoch: [24][120/200]	Time 0.384 (0.481)	Data 0.000 (0.088)	Loss 0.701 (0.743)
Epoch: [24][140/200]	Time 0.383 (0.479)	Data 0.000 (0.086)	Loss 1.087 (0.767)
Epoch: [24][160/200]	Time 0.382 (0.479)	Data 0.000 (0.085)	Loss 0.660 (0.775)
Epoch: [24][180/200]	Time 0.500 (0.478)	Data 0.000 (0.084)	Loss 0.844 (0.783)
Epoch: [24][200/200]	Time 0.495 (0.477)	Data 0.000 (0.083)	Loss 0.988 (0.800)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.174)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.19994306564331
==> Statistics for epoch 25: 613 clusters
Epoch: [25][20/200]	Time 1.992 (0.516)	Data 1.568 (0.120)	Loss 0.855 (0.173)
Epoch: [25][40/200]	Time 0.382 (0.496)	Data 0.001 (0.101)	Loss 0.743 (0.490)
Epoch: [25][60/200]	Time 0.382 (0.488)	Data 0.001 (0.093)	Loss 0.728 (0.617)
Epoch: [25][80/200]	Time 0.489 (0.485)	Data 0.001 (0.089)	Loss 0.692 (0.701)
Epoch: [25][100/200]	Time 0.497 (0.483)	Data 0.001 (0.087)	Loss 0.814 (0.746)
Epoch: [25][120/200]	Time 0.381 (0.480)	Data 0.001 (0.085)	Loss 0.906 (0.773)
Epoch: [25][140/200]	Time 0.382 (0.479)	Data 0.001 (0.084)	Loss 0.691 (0.787)
Epoch: [25][160/200]	Time 0.489 (0.480)	Data 0.002 (0.084)	Loss 0.817 (0.797)
Epoch: [25][180/200]	Time 0.475 (0.478)	Data 0.001 (0.082)	Loss 1.285 (0.814)
Epoch: [25][200/200]	Time 0.494 (0.476)	Data 0.001 (0.081)	Loss 0.708 (0.816)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.173)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.679495811462402
==> Statistics for epoch 26: 613 clusters
Epoch: [26][20/200]	Time 1.913 (0.521)	Data 1.490 (0.123)	Loss 0.966 (0.172)
Epoch: [26][40/200]	Time 0.388 (0.495)	Data 0.001 (0.099)	Loss 0.768 (0.512)
Epoch: [26][60/200]	Time 0.384 (0.486)	Data 0.001 (0.090)	Loss 0.796 (0.628)
Epoch: [26][80/200]	Time 0.384 (0.483)	Data 0.001 (0.088)	Loss 1.004 (0.687)
Epoch: [26][100/200]	Time 0.392 (0.484)	Data 0.001 (0.088)	Loss 0.758 (0.718)
Epoch: [26][120/200]	Time 0.385 (0.482)	Data 0.001 (0.087)	Loss 1.165 (0.744)
Epoch: [26][140/200]	Time 0.389 (0.480)	Data 0.002 (0.084)	Loss 0.852 (0.758)
Epoch: [26][160/200]	Time 0.386 (0.479)	Data 0.001 (0.083)	Loss 0.991 (0.765)
Epoch: [26][180/200]	Time 0.381 (0.478)	Data 0.001 (0.082)	Loss 0.812 (0.778)
Epoch: [26][200/200]	Time 0.384 (0.478)	Data 0.001 (0.082)	Loss 0.966 (0.795)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.178)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.047632217407227
==> Statistics for epoch 27: 611 clusters
Epoch: [27][20/200]	Time 2.104 (0.517)	Data 1.658 (0.119)	Loss 0.751 (0.156)
Epoch: [27][40/200]	Time 0.391 (0.494)	Data 0.001 (0.097)	Loss 0.702 (0.485)
Epoch: [27][60/200]	Time 0.388 (0.486)	Data 0.001 (0.091)	Loss 0.724 (0.620)
Epoch: [27][80/200]	Time 0.391 (0.483)	Data 0.001 (0.089)	Loss 0.761 (0.677)
Epoch: [27][100/200]	Time 0.380 (0.482)	Data 0.001 (0.086)	Loss 0.735 (0.712)
Epoch: [27][120/200]	Time 0.385 (0.481)	Data 0.001 (0.085)	Loss 0.487 (0.740)
Epoch: [27][140/200]	Time 0.385 (0.480)	Data 0.001 (0.084)	Loss 0.753 (0.754)
Epoch: [27][160/200]	Time 0.389 (0.479)	Data 0.001 (0.084)	Loss 0.787 (0.775)
Epoch: [27][180/200]	Time 0.390 (0.479)	Data 0.001 (0.083)	Loss 0.848 (0.786)
Epoch: [27][200/200]	Time 0.388 (0.478)	Data 0.001 (0.083)	Loss 0.690 (0.796)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.172)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.350334405899048
==> Statistics for epoch 28: 612 clusters
Epoch: [28][20/200]	Time 1.782 (0.506)	Data 1.341 (0.106)	Loss 0.693 (0.156)
Epoch: [28][40/200]	Time 0.392 (0.488)	Data 0.001 (0.092)	Loss 0.860 (0.525)
Epoch: [28][60/200]	Time 0.383 (0.478)	Data 0.001 (0.085)	Loss 1.208 (0.666)
Epoch: [28][80/200]	Time 0.399 (0.476)	Data 0.001 (0.084)	Loss 1.115 (0.718)
Epoch: [28][100/200]	Time 0.391 (0.474)	Data 0.001 (0.082)	Loss 0.805 (0.744)
Epoch: [28][120/200]	Time 0.381 (0.471)	Data 0.001 (0.079)	Loss 0.859 (0.776)
Epoch: [28][140/200]	Time 0.382 (0.470)	Data 0.001 (0.077)	Loss 1.042 (0.790)
Epoch: [28][160/200]	Time 0.385 (0.470)	Data 0.001 (0.077)	Loss 0.725 (0.802)
Epoch: [28][180/200]	Time 0.383 (0.470)	Data 0.001 (0.077)	Loss 0.877 (0.810)
Epoch: [28][200/200]	Time 0.382 (0.469)	Data 0.001 (0.076)	Loss 0.860 (0.817)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.169)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.27895450592041
==> Statistics for epoch 29: 612 clusters
Epoch: [29][20/200]	Time 1.957 (0.519)	Data 1.532 (0.120)	Loss 0.778 (0.159)
Epoch: [29][40/200]	Time 0.386 (0.493)	Data 0.001 (0.096)	Loss 0.753 (0.499)
Epoch: [29][60/200]	Time 0.389 (0.487)	Data 0.001 (0.091)	Loss 0.977 (0.627)
Epoch: [29][80/200]	Time 0.387 (0.484)	Data 0.001 (0.089)	Loss 0.903 (0.680)
Epoch: [29][100/200]	Time 0.387 (0.483)	Data 0.001 (0.087)	Loss 0.733 (0.707)
Epoch: [29][120/200]	Time 0.515 (0.482)	Data 0.001 (0.085)	Loss 0.963 (0.738)
Epoch: [29][140/200]	Time 0.381 (0.480)	Data 0.001 (0.084)	Loss 0.745 (0.759)
Epoch: [29][160/200]	Time 0.385 (0.478)	Data 0.001 (0.082)	Loss 0.979 (0.773)
Epoch: [29][180/200]	Time 0.383 (0.478)	Data 0.001 (0.082)	Loss 0.738 (0.779)
Epoch: [29][200/200]	Time 0.382 (0.477)	Data 0.001 (0.082)	Loss 0.900 (0.784)
Extract Features: [50/76]	Time 0.139 (0.177)	Data 0.000 (0.027)	
Mean AP: 93.2%

 * Finished epoch  29  model mAP: 93.2%  best: 93.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.175)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.0740225315094
==> Statistics for epoch 30: 613 clusters
Epoch: [30][20/200]	Time 2.034 (0.514)	Data 1.612 (0.118)	Loss 0.947 (0.172)
Epoch: [30][40/200]	Time 0.425 (0.490)	Data 0.001 (0.096)	Loss 0.778 (0.521)
Epoch: [30][60/200]	Time 0.384 (0.481)	Data 0.000 (0.086)	Loss 1.243 (0.637)
Epoch: [30][80/200]	Time 0.391 (0.478)	Data 0.002 (0.081)	Loss 1.192 (0.705)
Epoch: [30][100/200]	Time 0.384 (0.477)	Data 0.001 (0.080)	Loss 1.262 (0.734)
Epoch: [30][120/200]	Time 0.480 (0.475)	Data 0.001 (0.079)	Loss 0.909 (0.757)
Epoch: [30][140/200]	Time 0.383 (0.475)	Data 0.001 (0.078)	Loss 0.985 (0.773)
Epoch: [30][160/200]	Time 0.384 (0.474)	Data 0.001 (0.078)	Loss 1.010 (0.786)
Epoch: [30][180/200]	Time 0.385 (0.474)	Data 0.001 (0.078)	Loss 0.697 (0.796)
Epoch: [30][200/200]	Time 0.453 (0.473)	Data 0.001 (0.077)	Loss 1.092 (0.807)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.175)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.132311820983887
==> Statistics for epoch 31: 614 clusters
Epoch: [31][20/200]	Time 1.798 (0.512)	Data 1.344 (0.115)	Loss 0.551 (0.141)
Epoch: [31][40/200]	Time 0.388 (0.488)	Data 0.001 (0.091)	Loss 1.093 (0.483)
Epoch: [31][60/200]	Time 0.384 (0.483)	Data 0.001 (0.087)	Loss 1.125 (0.609)
Epoch: [31][80/200]	Time 0.381 (0.479)	Data 0.001 (0.083)	Loss 0.675 (0.683)
Epoch: [31][100/200]	Time 0.382 (0.476)	Data 0.001 (0.081)	Loss 0.840 (0.735)
Epoch: [31][120/200]	Time 0.385 (0.473)	Data 0.001 (0.079)	Loss 1.092 (0.761)
Epoch: [31][140/200]	Time 0.386 (0.472)	Data 0.001 (0.078)	Loss 0.787 (0.772)
Epoch: [31][160/200]	Time 0.384 (0.470)	Data 0.001 (0.076)	Loss 0.779 (0.785)
Epoch: [31][180/200]	Time 0.486 (0.470)	Data 0.001 (0.075)	Loss 0.826 (0.792)
Epoch: [31][200/200]	Time 0.479 (0.469)	Data 0.001 (0.075)	Loss 0.888 (0.797)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.168)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.29181671142578
==> Statistics for epoch 32: 611 clusters
Epoch: [32][20/200]	Time 1.768 (0.498)	Data 1.336 (0.102)	Loss 0.794 (0.161)
Epoch: [32][40/200]	Time 0.393 (0.479)	Data 0.001 (0.085)	Loss 1.153 (0.518)
Epoch: [32][60/200]	Time 0.385 (0.473)	Data 0.001 (0.080)	Loss 1.233 (0.644)
Epoch: [32][80/200]	Time 0.382 (0.471)	Data 0.001 (0.078)	Loss 0.976 (0.725)
Epoch: [32][100/200]	Time 0.388 (0.468)	Data 0.001 (0.075)	Loss 0.774 (0.759)
Epoch: [32][120/200]	Time 0.388 (0.467)	Data 0.001 (0.074)	Loss 0.685 (0.774)
Epoch: [32][140/200]	Time 0.388 (0.465)	Data 0.001 (0.073)	Loss 0.785 (0.781)
Epoch: [32][160/200]	Time 0.390 (0.465)	Data 0.001 (0.072)	Loss 0.777 (0.789)
Epoch: [32][180/200]	Time 0.384 (0.464)	Data 0.001 (0.072)	Loss 0.962 (0.796)
Epoch: [32][200/200]	Time 0.383 (0.464)	Data 0.001 (0.071)	Loss 0.768 (0.800)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.172)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.383086681365967
==> Statistics for epoch 33: 608 clusters
Epoch: [33][20/200]	Time 1.833 (0.500)	Data 1.410 (0.112)	Loss 0.937 (0.167)
Epoch: [33][40/200]	Time 0.400 (0.479)	Data 0.001 (0.088)	Loss 0.705 (0.488)
Epoch: [33][60/200]	Time 0.382 (0.473)	Data 0.001 (0.081)	Loss 1.027 (0.590)
Epoch: [33][80/200]	Time 0.383 (0.471)	Data 0.001 (0.079)	Loss 0.867 (0.656)
Epoch: [33][100/200]	Time 0.384 (0.472)	Data 0.001 (0.078)	Loss 1.070 (0.694)
Epoch: [33][120/200]	Time 0.498 (0.471)	Data 0.001 (0.077)	Loss 0.832 (0.717)
Epoch: [33][140/200]	Time 0.382 (0.470)	Data 0.000 (0.076)	Loss 0.699 (0.741)
Epoch: [33][160/200]	Time 0.383 (0.469)	Data 0.000 (0.075)	Loss 1.074 (0.763)
Epoch: [33][180/200]	Time 0.392 (0.469)	Data 0.000 (0.075)	Loss 0.573 (0.777)
Epoch: [33][200/200]	Time 0.383 (0.469)	Data 0.000 (0.074)	Loss 0.680 (0.786)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.171)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.871989250183105
==> Statistics for epoch 34: 611 clusters
Epoch: [34][20/200]	Time 1.971 (0.511)	Data 1.555 (0.113)	Loss 1.075 (0.173)
Epoch: [34][40/200]	Time 0.384 (0.492)	Data 0.001 (0.096)	Loss 0.834 (0.501)
Epoch: [34][60/200]	Time 0.385 (0.484)	Data 0.001 (0.089)	Loss 0.740 (0.610)
Epoch: [34][80/200]	Time 0.382 (0.481)	Data 0.001 (0.086)	Loss 0.636 (0.685)
Epoch: [34][100/200]	Time 0.382 (0.478)	Data 0.001 (0.083)	Loss 1.149 (0.718)
Epoch: [34][120/200]	Time 0.381 (0.478)	Data 0.001 (0.083)	Loss 1.141 (0.746)
Epoch: [34][140/200]	Time 0.383 (0.478)	Data 0.001 (0.083)	Loss 0.907 (0.762)
Epoch: [34][160/200]	Time 0.384 (0.478)	Data 0.001 (0.082)	Loss 0.803 (0.776)
Epoch: [34][180/200]	Time 0.384 (0.478)	Data 0.001 (0.083)	Loss 0.845 (0.788)
Epoch: [34][200/200]	Time 0.387 (0.478)	Data 0.001 (0.082)	Loss 0.499 (0.796)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.174)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.17927050590515
==> Statistics for epoch 35: 610 clusters
Epoch: [35][20/200]	Time 2.130 (0.519)	Data 1.587 (0.120)	Loss 0.637 (0.137)
Epoch: [35][40/200]	Time 0.520 (0.495)	Data 0.001 (0.097)	Loss 0.876 (0.491)
Epoch: [35][60/200]	Time 0.381 (0.489)	Data 0.000 (0.091)	Loss 1.090 (0.599)
Epoch: [35][80/200]	Time 0.389 (0.483)	Data 0.001 (0.087)	Loss 0.881 (0.667)
Epoch: [35][100/200]	Time 0.383 (0.479)	Data 0.001 (0.083)	Loss 0.610 (0.701)
Epoch: [35][120/200]	Time 0.387 (0.478)	Data 0.001 (0.083)	Loss 0.740 (0.725)
Epoch: [35][140/200]	Time 0.383 (0.476)	Data 0.001 (0.082)	Loss 0.863 (0.735)
Epoch: [35][160/200]	Time 0.486 (0.476)	Data 0.001 (0.081)	Loss 0.940 (0.753)
Epoch: [35][180/200]	Time 0.383 (0.476)	Data 0.000 (0.080)	Loss 1.008 (0.769)
Epoch: [35][200/200]	Time 0.457 (0.475)	Data 0.000 (0.080)	Loss 0.809 (0.775)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.172)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.5985209941864
==> Statistics for epoch 36: 610 clusters
Epoch: [36][20/200]	Time 1.796 (0.499)	Data 1.363 (0.104)	Loss 0.666 (0.143)
Epoch: [36][40/200]	Time 0.381 (0.481)	Data 0.001 (0.087)	Loss 0.729 (0.452)
Epoch: [36][60/200]	Time 0.388 (0.474)	Data 0.001 (0.081)	Loss 0.558 (0.579)
Epoch: [36][80/200]	Time 0.385 (0.470)	Data 0.001 (0.078)	Loss 0.792 (0.650)
Epoch: [36][100/200]	Time 0.381 (0.469)	Data 0.001 (0.076)	Loss 0.780 (0.688)
Epoch: [36][120/200]	Time 0.381 (0.468)	Data 0.001 (0.075)	Loss 0.806 (0.725)
Epoch: [36][140/200]	Time 0.389 (0.466)	Data 0.001 (0.074)	Loss 0.835 (0.732)
Epoch: [36][160/200]	Time 0.397 (0.466)	Data 0.001 (0.073)	Loss 0.990 (0.743)
Epoch: [36][180/200]	Time 0.381 (0.466)	Data 0.001 (0.072)	Loss 0.715 (0.755)
Epoch: [36][200/200]	Time 0.495 (0.466)	Data 0.001 (0.072)	Loss 0.734 (0.767)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.178)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.36927628517151
==> Statistics for epoch 37: 610 clusters
Epoch: [37][20/200]	Time 1.771 (0.497)	Data 1.332 (0.104)	Loss 0.817 (0.150)
Epoch: [37][40/200]	Time 0.380 (0.479)	Data 0.001 (0.088)	Loss 0.827 (0.470)
Epoch: [37][60/200]	Time 0.380 (0.473)	Data 0.001 (0.083)	Loss 0.974 (0.604)
Epoch: [37][80/200]	Time 0.381 (0.470)	Data 0.001 (0.078)	Loss 1.321 (0.672)
Epoch: [37][100/200]	Time 0.388 (0.468)	Data 0.001 (0.076)	Loss 0.904 (0.705)
Epoch: [37][120/200]	Time 0.383 (0.467)	Data 0.001 (0.075)	Loss 0.983 (0.732)
Epoch: [37][140/200]	Time 0.379 (0.466)	Data 0.001 (0.074)	Loss 1.048 (0.756)
Epoch: [37][160/200]	Time 0.384 (0.465)	Data 0.001 (0.073)	Loss 0.934 (0.765)
Epoch: [37][180/200]	Time 0.470 (0.465)	Data 0.001 (0.073)	Loss 1.406 (0.781)
Epoch: [37][200/200]	Time 0.384 (0.465)	Data 0.001 (0.072)	Loss 1.151 (0.790)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.171)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.559592723846436
==> Statistics for epoch 38: 612 clusters
Epoch: [38][20/200]	Time 1.999 (0.530)	Data 1.561 (0.125)	Loss 0.936 (0.160)
Epoch: [38][40/200]	Time 0.387 (0.504)	Data 0.001 (0.104)	Loss 0.796 (0.484)
Epoch: [38][60/200]	Time 0.390 (0.495)	Data 0.001 (0.095)	Loss 0.859 (0.607)
Epoch: [38][80/200]	Time 0.519 (0.490)	Data 0.001 (0.092)	Loss 0.828 (0.679)
Epoch: [38][100/200]	Time 0.388 (0.487)	Data 0.001 (0.088)	Loss 0.707 (0.711)
Epoch: [38][120/200]	Time 0.381 (0.485)	Data 0.001 (0.087)	Loss 0.710 (0.728)
Epoch: [38][140/200]	Time 0.383 (0.481)	Data 0.001 (0.084)	Loss 0.611 (0.747)
Epoch: [38][160/200]	Time 0.392 (0.481)	Data 0.001 (0.084)	Loss 1.092 (0.754)
Epoch: [38][180/200]	Time 0.384 (0.480)	Data 0.001 (0.083)	Loss 1.024 (0.765)
Epoch: [38][200/200]	Time 0.387 (0.480)	Data 0.001 (0.082)	Loss 0.865 (0.775)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.174)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.272294282913208
==> Statistics for epoch 39: 612 clusters
Epoch: [39][20/200]	Time 2.100 (0.517)	Data 1.673 (0.119)	Loss 0.683 (0.147)
Epoch: [39][40/200]	Time 0.386 (0.495)	Data 0.001 (0.100)	Loss 0.894 (0.477)
Epoch: [39][60/200]	Time 0.384 (0.486)	Data 0.001 (0.093)	Loss 0.824 (0.596)
Epoch: [39][80/200]	Time 0.384 (0.482)	Data 0.001 (0.088)	Loss 0.703 (0.662)
Epoch: [39][100/200]	Time 0.382 (0.479)	Data 0.000 (0.085)	Loss 1.009 (0.697)
Epoch: [39][120/200]	Time 0.478 (0.478)	Data 0.001 (0.082)	Loss 0.836 (0.721)
Epoch: [39][140/200]	Time 0.394 (0.477)	Data 0.000 (0.081)	Loss 0.933 (0.741)
Epoch: [39][160/200]	Time 0.386 (0.475)	Data 0.000 (0.080)	Loss 0.850 (0.751)
Epoch: [39][180/200]	Time 0.380 (0.475)	Data 0.001 (0.080)	Loss 1.076 (0.766)
Epoch: [39][200/200]	Time 0.382 (0.475)	Data 0.000 (0.080)	Loss 0.852 (0.770)
Extract Features: [50/76]	Time 0.141 (0.175)	Data 0.000 (0.025)	
Mean AP: 93.2%

 * Finished epoch  39  model mAP: 93.2%  best: 93.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.177)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.868385314941406
==> Statistics for epoch 40: 611 clusters
Epoch: [40][20/200]	Time 1.787 (0.496)	Data 1.314 (0.105)	Loss 0.913 (0.163)
Epoch: [40][40/200]	Time 0.397 (0.476)	Data 0.003 (0.084)	Loss 0.998 (0.501)
Epoch: [40][60/200]	Time 0.386 (0.475)	Data 0.001 (0.081)	Loss 0.730 (0.610)
Epoch: [40][80/200]	Time 0.387 (0.472)	Data 0.001 (0.078)	Loss 1.010 (0.684)
Epoch: [40][100/200]	Time 0.382 (0.470)	Data 0.001 (0.076)	Loss 1.124 (0.710)
Epoch: [40][120/200]	Time 0.382 (0.469)	Data 0.001 (0.075)	Loss 0.971 (0.737)
Epoch: [40][140/200]	Time 0.387 (0.469)	Data 0.001 (0.075)	Loss 0.745 (0.750)
Epoch: [40][160/200]	Time 0.387 (0.467)	Data 0.001 (0.073)	Loss 0.698 (0.768)
Epoch: [40][180/200]	Time 0.383 (0.468)	Data 0.001 (0.075)	Loss 0.909 (0.776)
Epoch: [40][200/200]	Time 0.383 (0.467)	Data 0.001 (0.074)	Loss 0.734 (0.774)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.173)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.08944845199585
==> Statistics for epoch 41: 609 clusters
Epoch: [41][20/200]	Time 1.881 (0.511)	Data 1.447 (0.112)	Loss 0.532 (0.124)
Epoch: [41][40/200]	Time 0.389 (0.493)	Data 0.001 (0.094)	Loss 0.596 (0.450)
Epoch: [41][60/200]	Time 0.383 (0.486)	Data 0.001 (0.088)	Loss 0.702 (0.580)
Epoch: [41][80/200]	Time 0.385 (0.483)	Data 0.001 (0.086)	Loss 0.841 (0.654)
Epoch: [41][100/200]	Time 0.381 (0.481)	Data 0.001 (0.085)	Loss 0.772 (0.706)
Epoch: [41][120/200]	Time 0.510 (0.481)	Data 0.001 (0.083)	Loss 0.683 (0.728)
Epoch: [41][140/200]	Time 0.481 (0.479)	Data 0.001 (0.082)	Loss 0.791 (0.751)
Epoch: [41][160/200]	Time 0.485 (0.478)	Data 0.001 (0.081)	Loss 0.991 (0.765)
Epoch: [41][180/200]	Time 0.385 (0.478)	Data 0.001 (0.082)	Loss 0.632 (0.780)
Epoch: [41][200/200]	Time 0.385 (0.477)	Data 0.001 (0.081)	Loss 0.959 (0.785)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.174)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.263917207717896
==> Statistics for epoch 42: 610 clusters
Epoch: [42][20/200]	Time 2.002 (0.514)	Data 1.597 (0.115)	Loss 0.655 (0.150)
Epoch: [42][40/200]	Time 0.388 (0.492)	Data 0.001 (0.096)	Loss 0.776 (0.494)
Epoch: [42][60/200]	Time 0.496 (0.488)	Data 0.001 (0.091)	Loss 0.696 (0.601)
Epoch: [42][80/200]	Time 0.389 (0.485)	Data 0.001 (0.087)	Loss 0.949 (0.661)
Epoch: [42][100/200]	Time 0.384 (0.481)	Data 0.001 (0.084)	Loss 1.112 (0.706)
Epoch: [42][120/200]	Time 0.381 (0.479)	Data 0.001 (0.083)	Loss 0.832 (0.741)
Epoch: [42][140/200]	Time 0.382 (0.477)	Data 0.001 (0.081)	Loss 0.709 (0.757)
Epoch: [42][160/200]	Time 0.384 (0.474)	Data 0.001 (0.079)	Loss 0.692 (0.772)
Epoch: [42][180/200]	Time 0.390 (0.473)	Data 0.001 (0.077)	Loss 0.651 (0.781)
Epoch: [42][200/200]	Time 0.385 (0.473)	Data 0.001 (0.077)	Loss 1.204 (0.790)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.170)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.929924964904785
==> Statistics for epoch 43: 612 clusters
Epoch: [43][20/200]	Time 1.787 (0.494)	Data 1.239 (0.096)	Loss 0.876 (0.157)
Epoch: [43][40/200]	Time 0.498 (0.478)	Data 0.001 (0.082)	Loss 0.847 (0.485)
Epoch: [43][60/200]	Time 0.389 (0.473)	Data 0.001 (0.078)	Loss 0.830 (0.594)
Epoch: [43][80/200]	Time 0.380 (0.471)	Data 0.001 (0.077)	Loss 0.683 (0.656)
Epoch: [43][100/200]	Time 0.382 (0.470)	Data 0.001 (0.075)	Loss 0.790 (0.688)
Epoch: [43][120/200]	Time 0.382 (0.469)	Data 0.001 (0.074)	Loss 0.984 (0.715)
Epoch: [43][140/200]	Time 0.380 (0.467)	Data 0.001 (0.073)	Loss 0.873 (0.726)
Epoch: [43][160/200]	Time 0.382 (0.468)	Data 0.001 (0.073)	Loss 0.906 (0.734)
Epoch: [43][180/200]	Time 0.387 (0.467)	Data 0.001 (0.073)	Loss 0.788 (0.745)
Epoch: [43][200/200]	Time 0.383 (0.468)	Data 0.001 (0.074)	Loss 0.629 (0.760)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.170)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.360512018203735
==> Statistics for epoch 44: 611 clusters
Epoch: [44][20/200]	Time 1.870 (0.506)	Data 1.445 (0.114)	Loss 0.766 (0.149)
Epoch: [44][40/200]	Time 0.379 (0.484)	Data 0.001 (0.093)	Loss 0.643 (0.486)
Epoch: [44][60/200]	Time 0.387 (0.475)	Data 0.001 (0.085)	Loss 0.883 (0.614)
Epoch: [44][80/200]	Time 0.389 (0.473)	Data 0.001 (0.082)	Loss 0.584 (0.668)
Epoch: [44][100/200]	Time 0.381 (0.471)	Data 0.001 (0.078)	Loss 0.822 (0.706)
Epoch: [44][120/200]	Time 0.381 (0.468)	Data 0.001 (0.076)	Loss 0.852 (0.729)
Epoch: [44][140/200]	Time 0.381 (0.468)	Data 0.001 (0.075)	Loss 0.770 (0.743)
Epoch: [44][160/200]	Time 0.382 (0.468)	Data 0.001 (0.076)	Loss 0.969 (0.760)
Epoch: [44][180/200]	Time 0.385 (0.467)	Data 0.001 (0.075)	Loss 0.767 (0.769)
Epoch: [44][200/200]	Time 0.477 (0.467)	Data 0.001 (0.074)	Loss 0.988 (0.777)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.141 (0.170)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.272587776184082
==> Statistics for epoch 45: 614 clusters
Epoch: [45][20/200]	Time 1.904 (0.511)	Data 1.343 (0.108)	Loss 0.811 (0.154)
Epoch: [45][40/200]	Time 0.504 (0.491)	Data 0.001 (0.090)	Loss 0.788 (0.487)
Epoch: [45][60/200]	Time 0.502 (0.484)	Data 0.001 (0.085)	Loss 0.781 (0.595)
Epoch: [45][80/200]	Time 0.382 (0.481)	Data 0.002 (0.082)	Loss 0.719 (0.669)
Epoch: [45][100/200]	Time 0.385 (0.478)	Data 0.001 (0.080)	Loss 0.891 (0.710)
Epoch: [45][120/200]	Time 0.384 (0.477)	Data 0.002 (0.080)	Loss 1.078 (0.719)
Epoch: [45][140/200]	Time 0.389 (0.476)	Data 0.001 (0.079)	Loss 0.831 (0.735)
Epoch: [45][160/200]	Time 0.389 (0.476)	Data 0.001 (0.079)	Loss 0.780 (0.748)
Epoch: [45][180/200]	Time 0.383 (0.475)	Data 0.001 (0.078)	Loss 0.754 (0.757)
Epoch: [45][200/200]	Time 0.384 (0.475)	Data 0.001 (0.078)	Loss 0.897 (0.769)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.178)	Data 0.000 (0.026)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.421788454055786
==> Statistics for epoch 46: 610 clusters
Epoch: [46][20/200]	Time 1.774 (0.498)	Data 1.361 (0.103)	Loss 0.794 (0.144)
Epoch: [46][40/200]	Time 0.388 (0.486)	Data 0.001 (0.094)	Loss 1.072 (0.485)
Epoch: [46][60/200]	Time 0.384 (0.476)	Data 0.001 (0.084)	Loss 0.911 (0.602)
Epoch: [46][80/200]	Time 0.383 (0.471)	Data 0.001 (0.079)	Loss 0.642 (0.658)
Epoch: [46][100/200]	Time 0.384 (0.471)	Data 0.001 (0.080)	Loss 0.723 (0.703)
Epoch: [46][120/200]	Time 0.384 (0.471)	Data 0.001 (0.080)	Loss 1.012 (0.725)
Epoch: [46][140/200]	Time 0.382 (0.470)	Data 0.001 (0.078)	Loss 1.049 (0.745)
Epoch: [46][160/200]	Time 0.384 (0.470)	Data 0.001 (0.078)	Loss 0.846 (0.755)
Epoch: [46][180/200]	Time 0.384 (0.470)	Data 0.001 (0.078)	Loss 0.804 (0.764)
Epoch: [46][200/200]	Time 0.463 (0.472)	Data 0.001 (0.079)	Loss 0.863 (0.773)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.142 (0.172)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.596879482269287
==> Statistics for epoch 47: 611 clusters
Epoch: [47][20/200]	Time 1.803 (0.502)	Data 1.393 (0.107)	Loss 1.073 (0.173)
Epoch: [47][40/200]	Time 0.393 (0.483)	Data 0.001 (0.087)	Loss 0.808 (0.490)
Epoch: [47][60/200]	Time 0.486 (0.477)	Data 0.001 (0.082)	Loss 1.169 (0.587)
Epoch: [47][80/200]	Time 0.383 (0.473)	Data 0.001 (0.077)	Loss 0.870 (0.669)
Epoch: [47][100/200]	Time 0.478 (0.470)	Data 0.001 (0.075)	Loss 0.614 (0.713)
Epoch: [47][120/200]	Time 0.383 (0.469)	Data 0.001 (0.074)	Loss 0.775 (0.737)
Epoch: [47][140/200]	Time 0.384 (0.469)	Data 0.001 (0.075)	Loss 0.754 (0.769)
Epoch: [47][160/200]	Time 0.386 (0.469)	Data 0.001 (0.074)	Loss 0.788 (0.780)
Epoch: [47][180/200]	Time 0.382 (0.468)	Data 0.001 (0.074)	Loss 0.718 (0.783)
Epoch: [47][200/200]	Time 0.386 (0.467)	Data 0.001 (0.073)	Loss 0.817 (0.789)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.168)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.271188497543335
==> Statistics for epoch 48: 610 clusters
Epoch: [48][20/200]	Time 1.748 (0.493)	Data 1.341 (0.099)	Loss 0.626 (0.132)
Epoch: [48][40/200]	Time 0.381 (0.477)	Data 0.001 (0.083)	Loss 0.792 (0.455)
Epoch: [48][60/200]	Time 0.478 (0.473)	Data 0.001 (0.078)	Loss 0.766 (0.562)
Epoch: [48][80/200]	Time 0.382 (0.468)	Data 0.001 (0.076)	Loss 0.837 (0.648)
Epoch: [48][100/200]	Time 0.381 (0.467)	Data 0.001 (0.074)	Loss 1.223 (0.691)
Epoch: [48][120/200]	Time 0.382 (0.466)	Data 0.001 (0.073)	Loss 0.883 (0.715)
Epoch: [48][140/200]	Time 0.383 (0.464)	Data 0.001 (0.072)	Loss 0.707 (0.733)
Epoch: [48][160/200]	Time 0.382 (0.463)	Data 0.001 (0.071)	Loss 0.759 (0.749)
Epoch: [48][180/200]	Time 0.382 (0.463)	Data 0.001 (0.072)	Loss 0.788 (0.754)
Epoch: [48][200/200]	Time 0.385 (0.464)	Data 0.001 (0.072)	Loss 0.642 (0.760)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.140 (0.170)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.482875108718872
==> Statistics for epoch 49: 613 clusters
Epoch: [49][20/200]	Time 1.754 (0.504)	Data 1.314 (0.107)	Loss 0.857 (0.160)
Epoch: [49][40/200]	Time 0.385 (0.481)	Data 0.001 (0.088)	Loss 0.893 (0.468)
Epoch: [49][60/200]	Time 0.385 (0.476)	Data 0.001 (0.083)	Loss 0.784 (0.568)
Epoch: [49][80/200]	Time 0.381 (0.472)	Data 0.001 (0.079)	Loss 0.503 (0.646)
Epoch: [49][100/200]	Time 0.387 (0.472)	Data 0.001 (0.079)	Loss 0.649 (0.681)
Epoch: [49][120/200]	Time 0.382 (0.471)	Data 0.001 (0.077)	Loss 0.967 (0.711)
Epoch: [49][140/200]	Time 0.382 (0.471)	Data 0.001 (0.077)	Loss 0.720 (0.737)
Epoch: [49][160/200]	Time 0.388 (0.471)	Data 0.001 (0.076)	Loss 0.717 (0.749)
Epoch: [49][180/200]	Time 0.383 (0.472)	Data 0.001 (0.077)	Loss 1.000 (0.760)
Epoch: [49][200/200]	Time 0.383 (0.471)	Data 0.001 (0.076)	Loss 0.944 (0.762)
Extract Features: [50/76]	Time 0.236 (0.172)	Data 0.000 (0.022)	
Mean AP: 93.2%

 * Finished epoch  49  model mAP: 93.2%  best: 93.2%

==> Test with the best model:
=> Loaded checkpoint 'log/market/resnet101_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.139 (0.172)	Data 0.000 (0.024)	
Mean AP: 93.2%
CMC Scores:
  top-1          97.0%
  top-5          98.9%
  top-10         99.3%
Total running time:  1:48:11.121555
