==========
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='resnet152', pretrained_path='/home/ma-user/work/Projects/ReIDNet_Finetune/TransReID/log/bs_exp/ResNet152_MSMT17/64bs_lr0.0004_ep120_warm20_seed0/resnet152_120.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/msmt2market/resnet152_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 0.181 (0.560)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.742334604263306
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 556 clusters
Epoch: [0][20/200]	Time 0.502 (1.055)	Data 0.001 (0.099)	Loss 2.929 (4.016)
Epoch: [0][40/200]	Time 0.501 (0.821)	Data 0.001 (0.083)	Loss 2.582 (3.481)
Epoch: [0][60/200]	Time 0.505 (0.740)	Data 0.001 (0.075)	Loss 1.825 (3.063)
Epoch: [0][80/200]	Time 0.499 (0.700)	Data 0.000 (0.071)	Loss 2.043 (2.828)
Epoch: [0][100/200]	Time 0.498 (0.676)	Data 0.000 (0.069)	Loss 2.206 (2.663)
Epoch: [0][120/200]	Time 1.837 (0.671)	Data 1.281 (0.078)	Loss 1.715 (2.526)
Epoch: [0][140/200]	Time 0.500 (0.660)	Data 0.001 (0.076)	Loss 1.641 (2.433)
Epoch: [0][160/200]	Time 0.591 (0.651)	Data 0.001 (0.074)	Loss 1.608 (2.346)
Epoch: [0][180/200]	Time 0.517 (0.644)	Data 0.001 (0.073)	Loss 1.698 (2.280)
Epoch: [0][200/200]	Time 0.505 (0.639)	Data 0.000 (0.072)	Loss 1.175 (2.210)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.220)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.479233026504517
==> Statistics for epoch 1: 605 clusters
Epoch: [1][20/200]	Time 0.505 (0.648)	Data 0.001 (0.115)	Loss 1.647 (0.373)
Epoch: [1][40/200]	Time 0.502 (0.621)	Data 0.001 (0.091)	Loss 1.901 (0.967)
Epoch: [1][60/200]	Time 0.510 (0.609)	Data 0.001 (0.083)	Loss 1.564 (1.143)
Epoch: [1][80/200]	Time 0.593 (0.605)	Data 0.001 (0.079)	Loss 1.303 (1.227)
Epoch: [1][100/200]	Time 0.516 (0.608)	Data 0.001 (0.081)	Loss 1.819 (1.275)
Epoch: [1][120/200]	Time 0.503 (0.606)	Data 0.000 (0.079)	Loss 1.587 (1.306)
Epoch: [1][140/200]	Time 0.503 (0.603)	Data 0.000 (0.077)	Loss 1.713 (1.327)
Epoch: [1][160/200]	Time 0.504 (0.601)	Data 0.000 (0.075)	Loss 1.534 (1.337)
Epoch: [1][180/200]	Time 0.511 (0.600)	Data 0.000 (0.074)	Loss 1.407 (1.344)
Epoch: [1][200/200]	Time 0.636 (0.605)	Data 0.001 (0.080)	Loss 1.390 (1.346)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.258 (0.227)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.395079612731934
==> Statistics for epoch 2: 606 clusters
Epoch: [2][20/200]	Time 0.602 (0.639)	Data 0.001 (0.110)	Loss 1.010 (0.323)
Epoch: [2][40/200]	Time 0.513 (0.614)	Data 0.002 (0.090)	Loss 1.492 (0.831)
Epoch: [2][60/200]	Time 0.518 (0.612)	Data 0.001 (0.089)	Loss 1.063 (1.021)
Epoch: [2][80/200]	Time 0.580 (0.609)	Data 0.002 (0.084)	Loss 1.497 (1.090)
Epoch: [2][100/200]	Time 0.509 (0.607)	Data 0.001 (0.082)	Loss 1.392 (1.132)
Epoch: [2][120/200]	Time 0.517 (0.607)	Data 0.000 (0.081)	Loss 1.346 (1.163)
Epoch: [2][140/200]	Time 0.506 (0.605)	Data 0.000 (0.080)	Loss 1.143 (1.181)
Epoch: [2][160/200]	Time 0.505 (0.604)	Data 0.000 (0.079)	Loss 1.028 (1.193)
Epoch: [2][180/200]	Time 0.595 (0.604)	Data 0.000 (0.079)	Loss 1.472 (1.200)
Epoch: [2][200/200]	Time 0.529 (0.611)	Data 0.001 (0.086)	Loss 1.066 (1.207)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.223)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.827301025390625
==> Statistics for epoch 3: 606 clusters
Epoch: [3][20/200]	Time 0.626 (0.645)	Data 0.001 (0.106)	Loss 1.122 (0.280)
Epoch: [3][40/200]	Time 0.504 (0.618)	Data 0.001 (0.086)	Loss 1.080 (0.714)
Epoch: [3][60/200]	Time 0.504 (0.615)	Data 0.001 (0.081)	Loss 1.442 (0.893)
Epoch: [3][80/200]	Time 0.510 (0.611)	Data 0.001 (0.079)	Loss 1.278 (0.981)
Epoch: [3][100/200]	Time 0.507 (0.608)	Data 0.001 (0.078)	Loss 1.095 (1.027)
Epoch: [3][120/200]	Time 0.500 (0.605)	Data 0.000 (0.076)	Loss 1.021 (1.046)
Epoch: [3][140/200]	Time 0.621 (0.605)	Data 0.000 (0.075)	Loss 0.997 (1.060)
Epoch: [3][160/200]	Time 0.504 (0.604)	Data 0.000 (0.075)	Loss 1.427 (1.068)
Epoch: [3][180/200]	Time 0.507 (0.606)	Data 0.000 (0.076)	Loss 1.273 (1.074)
Epoch: [3][200/200]	Time 0.619 (0.613)	Data 0.001 (0.083)	Loss 1.621 (1.085)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.225)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.215636730194092
==> Statistics for epoch 4: 611 clusters
Epoch: [4][20/200]	Time 2.107 (0.646)	Data 1.558 (0.117)	Loss 1.131 (0.221)
Epoch: [4][40/200]	Time 0.516 (0.621)	Data 0.001 (0.094)	Loss 1.294 (0.660)
Epoch: [4][60/200]	Time 0.633 (0.616)	Data 0.001 (0.087)	Loss 1.173 (0.810)
Epoch: [4][80/200]	Time 0.504 (0.612)	Data 0.002 (0.084)	Loss 0.973 (0.883)
Epoch: [4][100/200]	Time 0.504 (0.610)	Data 0.001 (0.082)	Loss 0.923 (0.935)
Epoch: [4][120/200]	Time 0.507 (0.609)	Data 0.002 (0.081)	Loss 0.819 (0.965)
Epoch: [4][140/200]	Time 0.523 (0.605)	Data 0.001 (0.079)	Loss 1.265 (0.981)
Epoch: [4][160/200]	Time 0.509 (0.607)	Data 0.001 (0.078)	Loss 1.196 (0.998)
Epoch: [4][180/200]	Time 0.594 (0.606)	Data 0.002 (0.078)	Loss 1.253 (1.010)
Epoch: [4][200/200]	Time 0.510 (0.606)	Data 0.001 (0.078)	Loss 1.158 (1.017)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.183 (0.224)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.93165874481201
==> Statistics for epoch 5: 618 clusters
Epoch: [5][20/200]	Time 2.126 (0.635)	Data 1.550 (0.110)	Loss 1.111 (0.210)
Epoch: [5][40/200]	Time 0.502 (0.622)	Data 0.001 (0.094)	Loss 1.047 (0.643)
Epoch: [5][60/200]	Time 0.504 (0.613)	Data 0.001 (0.086)	Loss 1.029 (0.789)
Epoch: [5][80/200]	Time 0.533 (0.609)	Data 0.001 (0.081)	Loss 0.975 (0.860)
Epoch: [5][100/200]	Time 0.620 (0.609)	Data 0.001 (0.079)	Loss 0.903 (0.893)
Epoch: [5][120/200]	Time 0.506 (0.610)	Data 0.001 (0.081)	Loss 1.040 (0.933)
Epoch: [5][140/200]	Time 0.504 (0.608)	Data 0.001 (0.079)	Loss 1.005 (0.938)
Epoch: [5][160/200]	Time 0.509 (0.607)	Data 0.001 (0.078)	Loss 1.100 (0.959)
Epoch: [5][180/200]	Time 0.515 (0.607)	Data 0.001 (0.078)	Loss 1.320 (0.964)
Epoch: [5][200/200]	Time 0.608 (0.608)	Data 0.001 (0.079)	Loss 1.164 (0.969)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.224)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.2910258769989
==> Statistics for epoch 6: 616 clusters
Epoch: [6][20/200]	Time 2.035 (0.661)	Data 1.456 (0.121)	Loss 0.842 (0.186)
Epoch: [6][40/200]	Time 0.620 (0.628)	Data 0.001 (0.094)	Loss 1.201 (0.578)
Epoch: [6][60/200]	Time 0.510 (0.620)	Data 0.001 (0.089)	Loss 0.911 (0.710)
Epoch: [6][80/200]	Time 0.505 (0.613)	Data 0.001 (0.084)	Loss 1.040 (0.781)
Epoch: [6][100/200]	Time 0.505 (0.608)	Data 0.001 (0.080)	Loss 1.194 (0.830)
Epoch: [6][120/200]	Time 0.502 (0.604)	Data 0.001 (0.078)	Loss 0.962 (0.852)
Epoch: [6][140/200]	Time 0.505 (0.602)	Data 0.001 (0.076)	Loss 0.761 (0.874)
Epoch: [6][160/200]	Time 0.505 (0.600)	Data 0.001 (0.074)	Loss 0.791 (0.879)
Epoch: [6][180/200]	Time 0.587 (0.598)	Data 0.001 (0.073)	Loss 1.200 (0.889)
Epoch: [6][200/200]	Time 0.508 (0.598)	Data 0.001 (0.073)	Loss 1.074 (0.898)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.183 (0.217)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.07100820541382
==> Statistics for epoch 7: 619 clusters
Epoch: [7][20/200]	Time 2.056 (0.650)	Data 1.451 (0.115)	Loss 1.121 (0.191)
Epoch: [7][40/200]	Time 0.601 (0.631)	Data 0.001 (0.099)	Loss 0.737 (0.551)
Epoch: [7][60/200]	Time 0.510 (0.622)	Data 0.001 (0.092)	Loss 1.115 (0.692)
Epoch: [7][80/200]	Time 0.510 (0.615)	Data 0.001 (0.085)	Loss 1.013 (0.757)
Epoch: [7][100/200]	Time 0.526 (0.611)	Data 0.001 (0.082)	Loss 0.962 (0.790)
Epoch: [7][120/200]	Time 0.515 (0.608)	Data 0.001 (0.080)	Loss 1.053 (0.813)
Epoch: [7][140/200]	Time 0.519 (0.606)	Data 0.001 (0.079)	Loss 0.856 (0.828)
Epoch: [7][160/200]	Time 0.594 (0.605)	Data 0.001 (0.078)	Loss 1.162 (0.838)
Epoch: [7][180/200]	Time 0.502 (0.604)	Data 0.001 (0.077)	Loss 0.875 (0.846)
Epoch: [7][200/200]	Time 0.596 (0.604)	Data 0.001 (0.076)	Loss 0.824 (0.849)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.220)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.20759105682373
==> Statistics for epoch 8: 614 clusters
Epoch: [8][20/200]	Time 1.848 (0.624)	Data 1.283 (0.109)	Loss 0.786 (0.158)
Epoch: [8][40/200]	Time 0.501 (0.606)	Data 0.001 (0.090)	Loss 1.069 (0.520)
Epoch: [8][60/200]	Time 0.512 (0.608)	Data 0.001 (0.087)	Loss 0.646 (0.629)
Epoch: [8][80/200]	Time 0.503 (0.608)	Data 0.001 (0.086)	Loss 0.874 (0.698)
Epoch: [8][100/200]	Time 0.509 (0.606)	Data 0.001 (0.083)	Loss 0.949 (0.745)
Epoch: [8][120/200]	Time 0.509 (0.605)	Data 0.001 (0.080)	Loss 0.920 (0.775)
Epoch: [8][140/200]	Time 0.515 (0.605)	Data 0.001 (0.080)	Loss 0.705 (0.792)
Epoch: [8][160/200]	Time 0.509 (0.604)	Data 0.002 (0.079)	Loss 1.194 (0.808)
Epoch: [8][180/200]	Time 0.504 (0.604)	Data 0.000 (0.079)	Loss 1.070 (0.824)
Epoch: [8][200/200]	Time 0.510 (0.604)	Data 0.000 (0.079)	Loss 0.652 (0.822)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.218)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.23555850982666
==> Statistics for epoch 9: 617 clusters
Epoch: [9][20/200]	Time 2.076 (0.645)	Data 1.505 (0.111)	Loss 0.764 (0.153)
Epoch: [9][40/200]	Time 0.504 (0.619)	Data 0.001 (0.090)	Loss 0.990 (0.486)
Epoch: [9][60/200]	Time 0.501 (0.610)	Data 0.001 (0.083)	Loss 0.873 (0.609)
Epoch: [9][80/200]	Time 0.503 (0.606)	Data 0.002 (0.081)	Loss 1.146 (0.683)
Epoch: [9][100/200]	Time 0.511 (0.604)	Data 0.001 (0.078)	Loss 0.834 (0.720)
Epoch: [9][120/200]	Time 0.506 (0.600)	Data 0.001 (0.076)	Loss 0.816 (0.735)
Epoch: [9][140/200]	Time 0.630 (0.600)	Data 0.001 (0.074)	Loss 0.836 (0.756)
Epoch: [9][160/200]	Time 0.519 (0.598)	Data 0.001 (0.073)	Loss 0.547 (0.769)
Epoch: [9][180/200]	Time 0.504 (0.597)	Data 0.001 (0.072)	Loss 0.722 (0.777)
Epoch: [9][200/200]	Time 0.589 (0.599)	Data 0.001 (0.074)	Loss 0.751 (0.782)
Extract Features: [50/76]	Time 0.184 (0.217)	Data 0.000 (0.022)	
Mean AP: 92.8%

 * Finished epoch   9  model mAP: 92.8%  best: 92.8% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.179 (0.226)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.79106569290161
==> Statistics for epoch 10: 611 clusters
Epoch: [10][20/200]	Time 1.999 (0.706)	Data 1.428 (0.112)	Loss 0.729 (0.151)
Epoch: [10][40/200]	Time 0.506 (0.648)	Data 0.001 (0.089)	Loss 0.682 (0.476)
Epoch: [10][60/200]	Time 0.505 (0.629)	Data 0.001 (0.082)	Loss 0.793 (0.595)
Epoch: [10][80/200]	Time 0.506 (0.619)	Data 0.001 (0.078)	Loss 0.916 (0.660)
Epoch: [10][100/200]	Time 0.499 (0.613)	Data 0.001 (0.077)	Loss 0.682 (0.702)
Epoch: [10][120/200]	Time 0.589 (0.610)	Data 0.001 (0.075)	Loss 0.964 (0.720)
Epoch: [10][140/200]	Time 0.502 (0.607)	Data 0.001 (0.074)	Loss 0.798 (0.727)
Epoch: [10][160/200]	Time 0.502 (0.608)	Data 0.001 (0.075)	Loss 0.807 (0.735)
Epoch: [10][180/200]	Time 0.527 (0.606)	Data 0.001 (0.074)	Loss 0.707 (0.745)
Epoch: [10][200/200]	Time 0.508 (0.605)	Data 0.001 (0.074)	Loss 0.740 (0.748)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.216)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.10217809677124
==> Statistics for epoch 11: 615 clusters
Epoch: [11][20/200]	Time 1.920 (0.632)	Data 1.344 (0.110)	Loss 0.896 (0.161)
Epoch: [11][40/200]	Time 0.508 (0.611)	Data 0.001 (0.088)	Loss 0.830 (0.490)
Epoch: [11][60/200]	Time 0.504 (0.610)	Data 0.001 (0.085)	Loss 0.670 (0.583)
Epoch: [11][80/200]	Time 0.514 (0.607)	Data 0.002 (0.083)	Loss 0.725 (0.646)
Epoch: [11][100/200]	Time 0.507 (0.605)	Data 0.001 (0.081)	Loss 0.657 (0.693)
Epoch: [11][120/200]	Time 0.505 (0.603)	Data 0.001 (0.079)	Loss 0.986 (0.709)
Epoch: [11][140/200]	Time 0.506 (0.601)	Data 0.001 (0.077)	Loss 0.877 (0.720)
Epoch: [11][160/200]	Time 0.515 (0.599)	Data 0.002 (0.076)	Loss 1.115 (0.731)
Epoch: [11][180/200]	Time 0.509 (0.601)	Data 0.001 (0.076)	Loss 0.703 (0.742)
Epoch: [11][200/200]	Time 0.602 (0.601)	Data 0.001 (0.075)	Loss 0.403 (0.738)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.186 (0.222)	Data 0.001 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.335532665252686
==> Statistics for epoch 12: 618 clusters
Epoch: [12][20/200]	Time 1.894 (0.639)	Data 1.334 (0.105)	Loss 0.646 (0.148)
Epoch: [12][40/200]	Time 0.511 (0.620)	Data 0.002 (0.088)	Loss 0.929 (0.448)
Epoch: [12][60/200]	Time 0.509 (0.613)	Data 0.001 (0.083)	Loss 0.717 (0.559)
Epoch: [12][80/200]	Time 0.511 (0.610)	Data 0.001 (0.081)	Loss 1.009 (0.624)
Epoch: [12][100/200]	Time 0.519 (0.608)	Data 0.001 (0.079)	Loss 0.886 (0.655)
Epoch: [12][120/200]	Time 0.600 (0.606)	Data 0.001 (0.077)	Loss 0.620 (0.677)
Epoch: [12][140/200]	Time 0.511 (0.605)	Data 0.001 (0.077)	Loss 0.532 (0.686)
Epoch: [12][160/200]	Time 0.505 (0.605)	Data 0.001 (0.077)	Loss 0.648 (0.692)
Epoch: [12][180/200]	Time 0.510 (0.605)	Data 0.001 (0.077)	Loss 0.962 (0.694)
Epoch: [12][200/200]	Time 0.506 (0.605)	Data 0.001 (0.076)	Loss 0.708 (0.698)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.222)	Data 0.000 (0.028)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.424603939056396
==> Statistics for epoch 13: 615 clusters
Epoch: [13][20/200]	Time 2.084 (0.648)	Data 1.505 (0.115)	Loss 0.561 (0.135)
Epoch: [13][40/200]	Time 0.504 (0.624)	Data 0.001 (0.096)	Loss 0.888 (0.458)
Epoch: [13][60/200]	Time 0.506 (0.616)	Data 0.001 (0.090)	Loss 0.789 (0.547)
Epoch: [13][80/200]	Time 0.508 (0.612)	Data 0.001 (0.088)	Loss 0.747 (0.589)
Epoch: [13][100/200]	Time 0.502 (0.610)	Data 0.001 (0.085)	Loss 0.711 (0.622)
Epoch: [13][120/200]	Time 0.503 (0.609)	Data 0.001 (0.084)	Loss 0.703 (0.648)
Epoch: [13][140/200]	Time 0.515 (0.609)	Data 0.001 (0.084)	Loss 0.846 (0.666)
Epoch: [13][160/200]	Time 0.509 (0.608)	Data 0.001 (0.083)	Loss 0.670 (0.675)
Epoch: [13][180/200]	Time 0.506 (0.608)	Data 0.001 (0.083)	Loss 0.923 (0.675)
Epoch: [13][200/200]	Time 0.528 (0.608)	Data 0.001 (0.082)	Loss 0.780 (0.682)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.184 (0.221)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.171497106552124
==> Statistics for epoch 14: 614 clusters
Epoch: [14][20/200]	Time 1.913 (0.637)	Data 1.341 (0.111)	Loss 0.795 (0.145)
Epoch: [14][40/200]	Time 0.516 (0.619)	Data 0.001 (0.096)	Loss 0.479 (0.412)
Epoch: [14][60/200]	Time 0.511 (0.611)	Data 0.001 (0.088)	Loss 0.651 (0.497)
Epoch: [14][80/200]	Time 0.511 (0.611)	Data 0.001 (0.086)	Loss 0.584 (0.544)
Epoch: [14][100/200]	Time 0.622 (0.609)	Data 0.001 (0.083)	Loss 0.496 (0.575)
Epoch: [14][120/200]	Time 0.505 (0.608)	Data 0.001 (0.083)	Loss 0.448 (0.601)
Epoch: [14][140/200]	Time 0.509 (0.607)	Data 0.001 (0.082)	Loss 0.845 (0.619)
Epoch: [14][160/200]	Time 0.508 (0.605)	Data 0.001 (0.081)	Loss 0.738 (0.637)
Epoch: [14][180/200]	Time 0.608 (0.605)	Data 0.002 (0.081)	Loss 0.740 (0.643)
Epoch: [14][200/200]	Time 0.509 (0.605)	Data 0.001 (0.080)	Loss 0.590 (0.643)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.221)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.04647707939148
==> Statistics for epoch 15: 618 clusters
Epoch: [15][20/200]	Time 1.873 (0.639)	Data 1.309 (0.108)	Loss 0.538 (0.125)
Epoch: [15][40/200]	Time 0.508 (0.615)	Data 0.001 (0.088)	Loss 0.725 (0.402)
Epoch: [15][60/200]	Time 0.501 (0.612)	Data 0.001 (0.080)	Loss 0.770 (0.499)
Epoch: [15][80/200]	Time 0.508 (0.605)	Data 0.000 (0.076)	Loss 0.647 (0.556)
Epoch: [15][100/200]	Time 0.508 (0.603)	Data 0.000 (0.075)	Loss 0.642 (0.587)
Epoch: [15][120/200]	Time 0.515 (0.602)	Data 0.001 (0.074)	Loss 0.681 (0.603)
Epoch: [15][140/200]	Time 0.528 (0.602)	Data 0.000 (0.075)	Loss 0.892 (0.620)
Epoch: [15][160/200]	Time 0.500 (0.601)	Data 0.000 (0.074)	Loss 0.746 (0.629)
Epoch: [15][180/200]	Time 0.511 (0.600)	Data 0.000 (0.074)	Loss 0.677 (0.635)
Epoch: [15][200/200]	Time 0.512 (0.600)	Data 0.000 (0.073)	Loss 0.937 (0.639)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.255 (0.218)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.2403564453125
==> Statistics for epoch 16: 614 clusters
Epoch: [16][20/200]	Time 1.847 (0.639)	Data 1.278 (0.106)	Loss 0.537 (0.125)
Epoch: [16][40/200]	Time 0.503 (0.616)	Data 0.001 (0.086)	Loss 0.449 (0.393)
Epoch: [16][60/200]	Time 0.503 (0.606)	Data 0.001 (0.080)	Loss 0.668 (0.481)
Epoch: [16][80/200]	Time 0.507 (0.601)	Data 0.001 (0.077)	Loss 0.679 (0.542)
Epoch: [16][100/200]	Time 0.503 (0.598)	Data 0.001 (0.074)	Loss 0.490 (0.566)
Epoch: [16][120/200]	Time 0.600 (0.598)	Data 0.001 (0.073)	Loss 0.655 (0.577)
Epoch: [16][140/200]	Time 0.509 (0.595)	Data 0.001 (0.072)	Loss 0.885 (0.586)
Epoch: [16][160/200]	Time 0.517 (0.596)	Data 0.001 (0.070)	Loss 0.470 (0.590)
Epoch: [16][180/200]	Time 0.503 (0.596)	Data 0.001 (0.070)	Loss 0.619 (0.597)
Epoch: [16][200/200]	Time 0.516 (0.596)	Data 0.001 (0.070)	Loss 0.546 (0.598)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.218)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.40554904937744
==> Statistics for epoch 17: 613 clusters
Epoch: [17][20/200]	Time 1.999 (0.630)	Data 1.321 (0.101)	Loss 0.634 (0.121)
Epoch: [17][40/200]	Time 0.505 (0.609)	Data 0.000 (0.082)	Loss 0.536 (0.375)
Epoch: [17][60/200]	Time 0.501 (0.601)	Data 0.000 (0.077)	Loss 0.851 (0.469)
Epoch: [17][80/200]	Time 0.503 (0.597)	Data 0.000 (0.075)	Loss 0.434 (0.518)
Epoch: [17][100/200]	Time 0.513 (0.597)	Data 0.000 (0.074)	Loss 0.926 (0.557)
Epoch: [17][120/200]	Time 0.512 (0.596)	Data 0.000 (0.073)	Loss 0.735 (0.561)
Epoch: [17][140/200]	Time 0.501 (0.598)	Data 0.000 (0.072)	Loss 0.648 (0.580)
Epoch: [17][160/200]	Time 0.507 (0.597)	Data 0.001 (0.071)	Loss 0.767 (0.590)
Epoch: [17][180/200]	Time 0.504 (0.596)	Data 0.000 (0.071)	Loss 0.527 (0.592)
Epoch: [17][200/200]	Time 0.512 (0.596)	Data 0.000 (0.071)	Loss 0.667 (0.599)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.216)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.208446264266968
==> Statistics for epoch 18: 612 clusters
Epoch: [18][20/200]	Time 1.810 (0.633)	Data 1.224 (0.099)	Loss 0.743 (0.115)
Epoch: [18][40/200]	Time 0.513 (0.608)	Data 0.001 (0.083)	Loss 0.430 (0.356)
Epoch: [18][60/200]	Time 0.637 (0.609)	Data 0.001 (0.081)	Loss 0.755 (0.467)
Epoch: [18][80/200]	Time 0.509 (0.605)	Data 0.001 (0.077)	Loss 0.659 (0.518)
Epoch: [18][100/200]	Time 0.509 (0.602)	Data 0.001 (0.075)	Loss 0.848 (0.547)
Epoch: [18][120/200]	Time 0.506 (0.599)	Data 0.001 (0.074)	Loss 0.590 (0.561)
Epoch: [18][140/200]	Time 0.508 (0.599)	Data 0.001 (0.073)	Loss 0.701 (0.566)
Epoch: [18][160/200]	Time 0.521 (0.597)	Data 0.006 (0.072)	Loss 0.413 (0.575)
Epoch: [18][180/200]	Time 0.510 (0.599)	Data 0.001 (0.074)	Loss 0.611 (0.580)
Epoch: [18][200/200]	Time 0.647 (0.599)	Data 0.002 (0.073)	Loss 0.674 (0.582)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.217)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.24428677558899
==> Statistics for epoch 19: 612 clusters
Epoch: [19][20/200]	Time 2.107 (0.646)	Data 1.471 (0.111)	Loss 0.605 (0.109)
Epoch: [19][40/200]	Time 0.499 (0.614)	Data 0.001 (0.087)	Loss 0.577 (0.350)
Epoch: [19][60/200]	Time 0.503 (0.605)	Data 0.001 (0.080)	Loss 0.480 (0.431)
Epoch: [19][80/200]	Time 0.501 (0.603)	Data 0.001 (0.078)	Loss 0.602 (0.468)
Epoch: [19][100/200]	Time 0.591 (0.600)	Data 0.001 (0.076)	Loss 0.624 (0.492)
Epoch: [19][120/200]	Time 0.501 (0.597)	Data 0.001 (0.074)	Loss 0.724 (0.515)
Epoch: [19][140/200]	Time 0.504 (0.596)	Data 0.001 (0.072)	Loss 0.559 (0.530)
Epoch: [19][160/200]	Time 0.504 (0.595)	Data 0.001 (0.071)	Loss 0.561 (0.539)
Epoch: [19][180/200]	Time 0.514 (0.595)	Data 0.001 (0.071)	Loss 0.431 (0.545)
Epoch: [19][200/200]	Time 0.520 (0.595)	Data 0.001 (0.070)	Loss 0.613 (0.555)
Extract Features: [50/76]	Time 0.181 (0.224)	Data 0.000 (0.026)	
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 0.182 (0.215)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.005631923675537
==> Statistics for epoch 20: 614 clusters
Epoch: [20][20/200]	Time 1.883 (0.633)	Data 1.346 (0.108)	Loss 0.715 (0.114)
Epoch: [20][40/200]	Time 0.509 (0.614)	Data 0.001 (0.087)	Loss 0.647 (0.347)
Epoch: [20][60/200]	Time 0.505 (0.605)	Data 0.001 (0.080)	Loss 0.567 (0.422)
Epoch: [20][80/200]	Time 0.505 (0.601)	Data 0.001 (0.077)	Loss 0.439 (0.464)
Epoch: [20][100/200]	Time 0.514 (0.597)	Data 0.002 (0.075)	Loss 0.724 (0.485)
Epoch: [20][120/200]	Time 0.510 (0.598)	Data 0.001 (0.075)	Loss 0.692 (0.501)
Epoch: [20][140/200]	Time 0.511 (0.598)	Data 0.001 (0.073)	Loss 0.384 (0.512)
Epoch: [20][160/200]	Time 0.504 (0.598)	Data 0.001 (0.073)	Loss 0.625 (0.520)
Epoch: [20][180/200]	Time 0.502 (0.597)	Data 0.001 (0.072)	Loss 0.611 (0.528)
Epoch: [20][200/200]	Time 0.593 (0.597)	Data 0.001 (0.071)	Loss 0.775 (0.535)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.219)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.107186317443848
==> Statistics for epoch 21: 612 clusters
Epoch: [21][20/200]	Time 1.931 (0.635)	Data 1.233 (0.100)	Loss 0.455 (0.103)
Epoch: [21][40/200]	Time 0.502 (0.609)	Data 0.001 (0.081)	Loss 0.612 (0.342)
Epoch: [21][60/200]	Time 0.512 (0.609)	Data 0.001 (0.081)	Loss 0.595 (0.422)
Epoch: [21][80/200]	Time 0.502 (0.603)	Data 0.001 (0.078)	Loss 0.542 (0.461)
Epoch: [21][100/200]	Time 0.500 (0.601)	Data 0.001 (0.075)	Loss 0.809 (0.487)
Epoch: [21][120/200]	Time 0.606 (0.600)	Data 0.001 (0.074)	Loss 0.474 (0.500)
Epoch: [21][140/200]	Time 0.506 (0.599)	Data 0.001 (0.072)	Loss 0.531 (0.514)
Epoch: [21][160/200]	Time 0.511 (0.598)	Data 0.001 (0.072)	Loss 0.463 (0.521)
Epoch: [21][180/200]	Time 0.506 (0.597)	Data 0.002 (0.071)	Loss 0.473 (0.531)
Epoch: [21][200/200]	Time 0.513 (0.597)	Data 0.001 (0.072)	Loss 0.569 (0.539)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.183 (0.217)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.478306770324707
==> Statistics for epoch 22: 612 clusters
Epoch: [22][20/200]	Time 1.961 (0.645)	Data 1.403 (0.112)	Loss 0.445 (0.103)
Epoch: [22][40/200]	Time 0.520 (0.629)	Data 0.002 (0.095)	Loss 0.767 (0.336)
Epoch: [22][60/200]	Time 0.517 (0.618)	Data 0.001 (0.085)	Loss 0.594 (0.407)
Epoch: [22][80/200]	Time 0.512 (0.615)	Data 0.002 (0.082)	Loss 0.494 (0.450)
Epoch: [22][100/200]	Time 0.503 (0.612)	Data 0.001 (0.081)	Loss 0.385 (0.485)
Epoch: [22][120/200]	Time 0.514 (0.610)	Data 0.000 (0.079)	Loss 0.529 (0.496)
Epoch: [22][140/200]	Time 0.521 (0.608)	Data 0.000 (0.078)	Loss 0.599 (0.508)
Epoch: [22][160/200]	Time 0.625 (0.606)	Data 0.000 (0.077)	Loss 0.413 (0.510)
Epoch: [22][180/200]	Time 0.513 (0.604)	Data 0.000 (0.076)	Loss 0.678 (0.515)
Epoch: [22][200/200]	Time 0.508 (0.603)	Data 0.000 (0.075)	Loss 0.566 (0.518)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.184 (0.225)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.34435224533081
==> Statistics for epoch 23: 614 clusters
Epoch: [23][20/200]	Time 1.831 (0.632)	Data 1.253 (0.100)	Loss 0.654 (0.113)
Epoch: [23][40/200]	Time 0.666 (0.614)	Data 0.000 (0.083)	Loss 0.559 (0.328)
Epoch: [23][60/200]	Time 0.634 (0.611)	Data 0.001 (0.080)	Loss 0.377 (0.405)
Epoch: [23][80/200]	Time 0.513 (0.608)	Data 0.001 (0.078)	Loss 0.455 (0.445)
Epoch: [23][100/200]	Time 0.511 (0.604)	Data 0.000 (0.075)	Loss 0.611 (0.484)
Epoch: [23][120/200]	Time 0.507 (0.602)	Data 0.001 (0.075)	Loss 0.572 (0.492)
Epoch: [23][140/200]	Time 0.630 (0.603)	Data 0.001 (0.074)	Loss 0.721 (0.504)
Epoch: [23][160/200]	Time 0.505 (0.601)	Data 0.001 (0.073)	Loss 0.470 (0.514)
Epoch: [23][180/200]	Time 0.502 (0.600)	Data 0.001 (0.072)	Loss 0.387 (0.525)
Epoch: [23][200/200]	Time 0.505 (0.599)	Data 0.001 (0.071)	Loss 0.585 (0.528)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.184 (0.219)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.84671640396118
==> Statistics for epoch 24: 615 clusters
Epoch: [24][20/200]	Time 1.885 (0.630)	Data 1.347 (0.105)	Loss 0.586 (0.118)
Epoch: [24][40/200]	Time 0.501 (0.609)	Data 0.001 (0.086)	Loss 0.818 (0.343)
Epoch: [24][60/200]	Time 0.603 (0.604)	Data 0.001 (0.083)	Loss 0.408 (0.429)
Epoch: [24][80/200]	Time 0.526 (0.600)	Data 0.001 (0.079)	Loss 0.658 (0.466)
Epoch: [24][100/200]	Time 0.513 (0.599)	Data 0.001 (0.078)	Loss 0.456 (0.490)
Epoch: [24][120/200]	Time 0.506 (0.597)	Data 0.001 (0.076)	Loss 0.520 (0.499)
Epoch: [24][140/200]	Time 0.505 (0.596)	Data 0.001 (0.074)	Loss 0.565 (0.507)
Epoch: [24][160/200]	Time 0.611 (0.597)	Data 0.001 (0.074)	Loss 0.606 (0.516)
Epoch: [24][180/200]	Time 0.510 (0.597)	Data 0.001 (0.074)	Loss 0.808 (0.523)
Epoch: [24][200/200]	Time 0.511 (0.598)	Data 0.001 (0.074)	Loss 0.589 (0.526)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.219)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.945669412612915
==> Statistics for epoch 25: 613 clusters
Epoch: [25][20/200]	Time 2.010 (0.642)	Data 1.434 (0.112)	Loss 0.525 (0.104)
Epoch: [25][40/200]	Time 0.525 (0.623)	Data 0.001 (0.094)	Loss 0.539 (0.317)
Epoch: [25][60/200]	Time 0.611 (0.617)	Data 0.001 (0.088)	Loss 0.727 (0.409)
Epoch: [25][80/200]	Time 0.503 (0.610)	Data 0.001 (0.084)	Loss 0.415 (0.457)
Epoch: [25][100/200]	Time 0.508 (0.607)	Data 0.002 (0.081)	Loss 0.467 (0.471)
Epoch: [25][120/200]	Time 0.513 (0.604)	Data 0.001 (0.078)	Loss 0.694 (0.494)
Epoch: [25][140/200]	Time 0.503 (0.602)	Data 0.001 (0.076)	Loss 0.692 (0.511)
Epoch: [25][160/200]	Time 0.518 (0.604)	Data 0.002 (0.076)	Loss 0.644 (0.516)
Epoch: [25][180/200]	Time 0.505 (0.604)	Data 0.001 (0.076)	Loss 0.581 (0.525)
Epoch: [25][200/200]	Time 0.581 (0.604)	Data 0.001 (0.076)	Loss 0.537 (0.528)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.217)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.555010557174683
==> Statistics for epoch 26: 610 clusters
Epoch: [26][20/200]	Time 2.202 (0.652)	Data 1.530 (0.117)	Loss 0.524 (0.089)
Epoch: [26][40/200]	Time 0.511 (0.622)	Data 0.001 (0.093)	Loss 0.639 (0.294)
Epoch: [26][60/200]	Time 0.508 (0.610)	Data 0.000 (0.084)	Loss 0.628 (0.384)
Epoch: [26][80/200]	Time 0.505 (0.608)	Data 0.000 (0.079)	Loss 0.296 (0.427)
Epoch: [26][100/200]	Time 0.633 (0.606)	Data 0.000 (0.077)	Loss 0.316 (0.458)
Epoch: [26][120/200]	Time 0.512 (0.603)	Data 0.000 (0.075)	Loss 0.623 (0.469)
Epoch: [26][140/200]	Time 0.502 (0.604)	Data 0.000 (0.075)	Loss 0.527 (0.483)
Epoch: [26][160/200]	Time 0.501 (0.602)	Data 0.000 (0.074)	Loss 0.564 (0.486)
Epoch: [26][180/200]	Time 0.507 (0.601)	Data 0.000 (0.073)	Loss 0.333 (0.492)
Epoch: [26][200/200]	Time 0.514 (0.600)	Data 0.001 (0.072)	Loss 0.422 (0.496)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.182 (0.218)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.494344234466553
==> Statistics for epoch 27: 615 clusters
Epoch: [27][20/200]	Time 2.307 (0.659)	Data 1.743 (0.126)	Loss 0.497 (0.098)
Epoch: [27][40/200]	Time 0.507 (0.625)	Data 0.001 (0.098)	Loss 0.368 (0.290)
Epoch: [27][60/200]	Time 0.643 (0.620)	Data 0.008 (0.090)	Loss 0.627 (0.376)
Epoch: [27][80/200]	Time 0.506 (0.613)	Data 0.001 (0.085)	Loss 0.360 (0.419)
Epoch: [27][100/200]	Time 0.507 (0.609)	Data 0.001 (0.081)	Loss 0.446 (0.446)
Epoch: [27][120/200]	Time 0.503 (0.607)	Data 0.001 (0.080)	Loss 0.537 (0.466)
Epoch: [27][140/200]	Time 0.504 (0.605)	Data 0.001 (0.078)	Loss 0.582 (0.485)
Epoch: [27][160/200]	Time 0.511 (0.603)	Data 0.002 (0.076)	Loss 0.621 (0.492)
Epoch: [27][180/200]	Time 0.508 (0.603)	Data 0.001 (0.076)	Loss 0.465 (0.500)
Epoch: [27][200/200]	Time 0.514 (0.604)	Data 0.001 (0.077)	Loss 0.899 (0.504)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.256 (0.220)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.200076818466187
==> Statistics for epoch 28: 611 clusters
Epoch: [28][20/200]	Time 2.150 (0.653)	Data 1.610 (0.129)	Loss 0.841 (0.118)
Epoch: [28][40/200]	Time 0.514 (0.632)	Data 0.002 (0.103)	Loss 0.591 (0.329)
Epoch: [28][60/200]	Time 0.506 (0.620)	Data 0.002 (0.092)	Loss 0.667 (0.396)
Epoch: [28][80/200]	Time 0.506 (0.617)	Data 0.001 (0.089)	Loss 0.525 (0.439)
Epoch: [28][100/200]	Time 0.515 (0.616)	Data 0.001 (0.088)	Loss 0.547 (0.462)
Epoch: [28][120/200]	Time 0.502 (0.613)	Data 0.001 (0.086)	Loss 0.530 (0.470)
Epoch: [28][140/200]	Time 0.508 (0.611)	Data 0.001 (0.085)	Loss 0.478 (0.482)
Epoch: [28][160/200]	Time 0.509 (0.612)	Data 0.001 (0.085)	Loss 0.667 (0.492)
Epoch: [28][180/200]	Time 0.505 (0.611)	Data 0.001 (0.084)	Loss 0.640 (0.497)
Epoch: [28][200/200]	Time 0.521 (0.610)	Data 0.001 (0.083)	Loss 0.541 (0.501)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.220)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.030344247817993
==> Statistics for epoch 29: 614 clusters
Epoch: [29][20/200]	Time 1.958 (0.633)	Data 1.369 (0.107)	Loss 0.521 (0.100)
Epoch: [29][40/200]	Time 0.511 (0.609)	Data 0.001 (0.086)	Loss 0.711 (0.315)
Epoch: [29][60/200]	Time 0.524 (0.601)	Data 0.001 (0.078)	Loss 0.562 (0.384)
Epoch: [29][80/200]	Time 0.505 (0.599)	Data 0.002 (0.076)	Loss 0.575 (0.430)
Epoch: [29][100/200]	Time 0.509 (0.597)	Data 0.001 (0.073)	Loss 0.681 (0.449)
Epoch: [29][120/200]	Time 0.505 (0.599)	Data 0.001 (0.073)	Loss 0.532 (0.474)
Epoch: [29][140/200]	Time 0.503 (0.598)	Data 0.001 (0.073)	Loss 0.364 (0.479)
Epoch: [29][160/200]	Time 0.604 (0.599)	Data 0.001 (0.072)	Loss 0.527 (0.487)
Epoch: [29][180/200]	Time 0.507 (0.597)	Data 0.001 (0.072)	Loss 0.503 (0.492)
Epoch: [29][200/200]	Time 0.511 (0.597)	Data 0.001 (0.072)	Loss 0.608 (0.500)
Extract Features: [50/76]	Time 0.181 (0.217)	Data 0.000 (0.022)	
Mean AP: 93.5%

 * Finished epoch  29  model mAP: 93.5%  best: 93.5% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.218)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.388513565063477
==> Statistics for epoch 30: 611 clusters
Epoch: [30][20/200]	Time 2.027 (0.643)	Data 1.478 (0.113)	Loss 0.487 (0.101)
Epoch: [30][40/200]	Time 0.504 (0.612)	Data 0.001 (0.092)	Loss 0.501 (0.327)
Epoch: [30][60/200]	Time 0.519 (0.613)	Data 0.001 (0.088)	Loss 0.503 (0.406)
Epoch: [30][80/200]	Time 0.508 (0.609)	Data 0.002 (0.083)	Loss 0.523 (0.442)
Epoch: [30][100/200]	Time 0.510 (0.605)	Data 0.001 (0.079)	Loss 0.430 (0.470)
Epoch: [30][120/200]	Time 0.503 (0.603)	Data 0.001 (0.077)	Loss 0.470 (0.482)
Epoch: [30][140/200]	Time 0.505 (0.603)	Data 0.001 (0.076)	Loss 0.366 (0.486)
Epoch: [30][160/200]	Time 0.503 (0.602)	Data 0.001 (0.075)	Loss 0.801 (0.495)
Epoch: [30][180/200]	Time 0.510 (0.599)	Data 0.001 (0.074)	Loss 0.450 (0.499)
Epoch: [30][200/200]	Time 0.662 (0.600)	Data 0.001 (0.074)	Loss 0.650 (0.500)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.219)	Data 0.000 (0.024)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.85970377922058
==> Statistics for epoch 31: 610 clusters
Epoch: [31][20/200]	Time 2.081 (0.650)	Data 1.512 (0.113)	Loss 0.456 (0.095)
Epoch: [31][40/200]	Time 0.503 (0.623)	Data 0.001 (0.093)	Loss 0.851 (0.318)
Epoch: [31][60/200]	Time 0.518 (0.612)	Data 0.001 (0.085)	Loss 0.516 (0.406)
Epoch: [31][80/200]	Time 0.508 (0.608)	Data 0.001 (0.083)	Loss 0.557 (0.444)
Epoch: [31][100/200]	Time 0.509 (0.604)	Data 0.001 (0.079)	Loss 0.320 (0.459)
Epoch: [31][120/200]	Time 0.514 (0.601)	Data 0.001 (0.077)	Loss 0.595 (0.471)
Epoch: [31][140/200]	Time 0.510 (0.599)	Data 0.001 (0.075)	Loss 0.573 (0.483)
Epoch: [31][160/200]	Time 0.607 (0.598)	Data 0.002 (0.074)	Loss 0.633 (0.489)
Epoch: [31][180/200]	Time 0.508 (0.598)	Data 0.001 (0.073)	Loss 0.612 (0.497)
Epoch: [31][200/200]	Time 0.512 (0.599)	Data 0.001 (0.074)	Loss 0.616 (0.499)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.275 (0.223)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.088237285614014
==> Statistics for epoch 32: 613 clusters
Epoch: [32][20/200]	Time 2.290 (0.650)	Data 1.766 (0.124)	Loss 0.581 (0.097)
Epoch: [32][40/200]	Time 0.515 (0.624)	Data 0.001 (0.099)	Loss 0.636 (0.299)
Epoch: [32][60/200]	Time 0.506 (0.616)	Data 0.001 (0.090)	Loss 0.941 (0.379)
Epoch: [32][80/200]	Time 0.626 (0.612)	Data 0.002 (0.087)	Loss 0.519 (0.420)
Epoch: [32][100/200]	Time 0.511 (0.610)	Data 0.002 (0.085)	Loss 0.575 (0.448)
Epoch: [32][120/200]	Time 0.508 (0.609)	Data 0.001 (0.084)	Loss 0.515 (0.467)
Epoch: [32][140/200]	Time 0.506 (0.608)	Data 0.001 (0.083)	Loss 0.499 (0.476)
Epoch: [32][160/200]	Time 0.603 (0.608)	Data 0.001 (0.082)	Loss 0.421 (0.478)
Epoch: [32][180/200]	Time 0.503 (0.607)	Data 0.001 (0.081)	Loss 0.459 (0.480)
Epoch: [32][200/200]	Time 0.597 (0.607)	Data 0.001 (0.081)	Loss 0.511 (0.482)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.260 (0.221)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.88393235206604
==> Statistics for epoch 33: 613 clusters
Epoch: [33][20/200]	Time 2.030 (0.646)	Data 1.336 (0.110)	Loss 0.583 (0.103)
Epoch: [33][40/200]	Time 0.511 (0.619)	Data 0.002 (0.088)	Loss 0.487 (0.315)
Epoch: [33][60/200]	Time 0.507 (0.611)	Data 0.001 (0.081)	Loss 0.528 (0.387)
Epoch: [33][80/200]	Time 0.608 (0.610)	Data 0.002 (0.078)	Loss 0.504 (0.418)
Epoch: [33][100/200]	Time 0.514 (0.606)	Data 0.002 (0.076)	Loss 0.489 (0.445)
Epoch: [33][120/200]	Time 0.501 (0.603)	Data 0.001 (0.073)	Loss 0.508 (0.464)
Epoch: [33][140/200]	Time 0.510 (0.601)	Data 0.001 (0.073)	Loss 0.381 (0.472)
Epoch: [33][160/200]	Time 0.586 (0.602)	Data 0.001 (0.073)	Loss 0.615 (0.486)
Epoch: [33][180/200]	Time 0.508 (0.602)	Data 0.001 (0.073)	Loss 0.634 (0.492)
Epoch: [33][200/200]	Time 0.525 (0.601)	Data 0.001 (0.073)	Loss 0.634 (0.501)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.215)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.04818034172058
==> Statistics for epoch 34: 613 clusters
Epoch: [34][20/200]	Time 1.896 (0.633)	Data 1.341 (0.109)	Loss 0.474 (0.097)
Epoch: [34][40/200]	Time 0.528 (0.608)	Data 0.001 (0.088)	Loss 0.434 (0.295)
Epoch: [34][60/200]	Time 0.514 (0.605)	Data 0.001 (0.085)	Loss 0.590 (0.373)
Epoch: [34][80/200]	Time 0.508 (0.602)	Data 0.002 (0.081)	Loss 0.596 (0.416)
Epoch: [34][100/200]	Time 0.634 (0.601)	Data 0.001 (0.079)	Loss 0.328 (0.444)
Epoch: [34][120/200]	Time 0.504 (0.602)	Data 0.001 (0.078)	Loss 0.616 (0.458)
Epoch: [34][140/200]	Time 0.507 (0.601)	Data 0.001 (0.077)	Loss 0.689 (0.463)
Epoch: [34][160/200]	Time 0.634 (0.601)	Data 0.002 (0.075)	Loss 0.682 (0.470)
Epoch: [34][180/200]	Time 0.503 (0.600)	Data 0.001 (0.075)	Loss 0.491 (0.471)
Epoch: [34][200/200]	Time 0.505 (0.600)	Data 0.001 (0.075)	Loss 0.615 (0.473)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.257 (0.219)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.013123750686646
==> Statistics for epoch 35: 613 clusters
Epoch: [35][20/200]	Time 1.912 (0.640)	Data 1.348 (0.105)	Loss 0.658 (0.100)
Epoch: [35][40/200]	Time 0.502 (0.614)	Data 0.001 (0.086)	Loss 0.869 (0.305)
Epoch: [35][60/200]	Time 0.518 (0.610)	Data 0.001 (0.081)	Loss 0.523 (0.385)
Epoch: [35][80/200]	Time 0.507 (0.606)	Data 0.002 (0.078)	Loss 0.542 (0.421)
Epoch: [35][100/200]	Time 0.503 (0.602)	Data 0.001 (0.075)	Loss 0.348 (0.442)
Epoch: [35][120/200]	Time 0.621 (0.601)	Data 0.001 (0.074)	Loss 0.598 (0.454)
Epoch: [35][140/200]	Time 0.628 (0.599)	Data 0.001 (0.072)	Loss 0.402 (0.465)
Epoch: [35][160/200]	Time 0.514 (0.599)	Data 0.001 (0.073)	Loss 0.595 (0.473)
Epoch: [35][180/200]	Time 0.505 (0.600)	Data 0.001 (0.072)	Loss 0.662 (0.482)
Epoch: [35][200/200]	Time 0.515 (0.598)	Data 0.001 (0.071)	Loss 0.577 (0.487)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.260 (0.219)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.94736337661743
==> Statistics for epoch 36: 614 clusters
Epoch: [36][20/200]	Time 2.037 (0.642)	Data 1.452 (0.115)	Loss 0.507 (0.097)
Epoch: [36][40/200]	Time 0.517 (0.621)	Data 0.000 (0.094)	Loss 0.367 (0.307)
Epoch: [36][60/200]	Time 0.509 (0.615)	Data 0.001 (0.089)	Loss 0.585 (0.389)
Epoch: [36][80/200]	Time 0.505 (0.612)	Data 0.001 (0.087)	Loss 0.488 (0.433)
Epoch: [36][100/200]	Time 0.506 (0.610)	Data 0.001 (0.086)	Loss 0.489 (0.454)
Epoch: [36][120/200]	Time 0.510 (0.608)	Data 0.001 (0.083)	Loss 0.451 (0.470)
Epoch: [36][140/200]	Time 0.601 (0.608)	Data 0.000 (0.082)	Loss 0.438 (0.472)
Epoch: [36][160/200]	Time 0.505 (0.608)	Data 0.000 (0.082)	Loss 0.346 (0.476)
Epoch: [36][180/200]	Time 0.594 (0.608)	Data 0.000 (0.081)	Loss 0.584 (0.484)
Epoch: [36][200/200]	Time 0.506 (0.607)	Data 0.001 (0.081)	Loss 0.432 (0.485)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.223)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.84186553955078
==> Statistics for epoch 37: 611 clusters
Epoch: [37][20/200]	Time 2.020 (0.629)	Data 1.439 (0.113)	Loss 0.465 (0.100)
Epoch: [37][40/200]	Time 0.506 (0.617)	Data 0.001 (0.094)	Loss 0.408 (0.312)
Epoch: [37][60/200]	Time 0.614 (0.610)	Data 0.001 (0.086)	Loss 0.593 (0.388)
Epoch: [37][80/200]	Time 0.524 (0.608)	Data 0.002 (0.084)	Loss 0.492 (0.431)
Epoch: [37][100/200]	Time 0.504 (0.606)	Data 0.001 (0.082)	Loss 0.555 (0.451)
Epoch: [37][120/200]	Time 0.501 (0.604)	Data 0.001 (0.080)	Loss 0.793 (0.464)
Epoch: [37][140/200]	Time 0.508 (0.602)	Data 0.001 (0.079)	Loss 0.885 (0.480)
Epoch: [37][160/200]	Time 0.510 (0.601)	Data 0.001 (0.077)	Loss 0.540 (0.484)
Epoch: [37][180/200]	Time 0.509 (0.600)	Data 0.001 (0.076)	Loss 0.637 (0.488)
Epoch: [37][200/200]	Time 0.506 (0.601)	Data 0.001 (0.076)	Loss 0.355 (0.488)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.215)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.782891273498535
==> Statistics for epoch 38: 612 clusters
Epoch: [38][20/200]	Time 1.955 (0.633)	Data 1.378 (0.105)	Loss 0.533 (0.096)
Epoch: [38][40/200]	Time 0.517 (0.615)	Data 0.001 (0.087)	Loss 0.423 (0.298)
Epoch: [38][60/200]	Time 0.503 (0.609)	Data 0.001 (0.084)	Loss 0.547 (0.366)
Epoch: [38][80/200]	Time 0.635 (0.607)	Data 0.001 (0.081)	Loss 0.492 (0.386)
Epoch: [38][100/200]	Time 0.505 (0.603)	Data 0.001 (0.079)	Loss 0.521 (0.412)
Epoch: [38][120/200]	Time 0.500 (0.601)	Data 0.001 (0.077)	Loss 0.466 (0.426)
Epoch: [38][140/200]	Time 0.518 (0.601)	Data 0.002 (0.076)	Loss 0.631 (0.448)
Epoch: [38][160/200]	Time 0.523 (0.602)	Data 0.001 (0.076)	Loss 0.610 (0.460)
Epoch: [38][180/200]	Time 0.595 (0.602)	Data 0.001 (0.077)	Loss 0.404 (0.466)
Epoch: [38][200/200]	Time 0.503 (0.600)	Data 0.001 (0.075)	Loss 0.431 (0.476)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.178 (0.221)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.953566312789917
==> Statistics for epoch 39: 612 clusters
Epoch: [39][20/200]	Time 2.127 (0.649)	Data 1.544 (0.116)	Loss 0.359 (0.082)
Epoch: [39][40/200]	Time 0.505 (0.623)	Data 0.001 (0.094)	Loss 0.417 (0.287)
Epoch: [39][60/200]	Time 0.499 (0.613)	Data 0.001 (0.084)	Loss 0.653 (0.367)
Epoch: [39][80/200]	Time 0.511 (0.607)	Data 0.001 (0.079)	Loss 0.647 (0.397)
Epoch: [39][100/200]	Time 0.506 (0.601)	Data 0.000 (0.076)	Loss 0.394 (0.417)
Epoch: [39][120/200]	Time 0.513 (0.601)	Data 0.000 (0.075)	Loss 0.711 (0.436)
Epoch: [39][140/200]	Time 0.505 (0.601)	Data 0.000 (0.076)	Loss 0.358 (0.446)
Epoch: [39][160/200]	Time 0.514 (0.601)	Data 0.000 (0.076)	Loss 0.366 (0.454)
Epoch: [39][180/200]	Time 0.511 (0.601)	Data 0.001 (0.075)	Loss 0.461 (0.458)
Epoch: [39][200/200]	Time 0.592 (0.600)	Data 0.000 (0.074)	Loss 0.479 (0.462)
Extract Features: [50/76]	Time 0.184 (0.216)	Data 0.000 (0.023)	
Mean AP: 93.5%

 * Finished epoch  39  model mAP: 93.5%  best: 93.5% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.216)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.619168996810913
==> Statistics for epoch 40: 614 clusters
Epoch: [40][20/200]	Time 2.061 (0.648)	Data 1.482 (0.108)	Loss 0.311 (0.091)
Epoch: [40][40/200]	Time 0.511 (0.622)	Data 0.001 (0.089)	Loss 0.514 (0.298)
Epoch: [40][60/200]	Time 0.598 (0.613)	Data 0.001 (0.083)	Loss 0.447 (0.366)
Epoch: [40][80/200]	Time 0.514 (0.610)	Data 0.001 (0.082)	Loss 0.667 (0.404)
Epoch: [40][100/200]	Time 0.503 (0.605)	Data 0.001 (0.079)	Loss 0.486 (0.424)
Epoch: [40][120/200]	Time 0.513 (0.603)	Data 0.001 (0.077)	Loss 0.575 (0.447)
Epoch: [40][140/200]	Time 0.508 (0.604)	Data 0.001 (0.077)	Loss 0.492 (0.455)
Epoch: [40][160/200]	Time 0.587 (0.603)	Data 0.001 (0.077)	Loss 0.489 (0.464)
Epoch: [40][180/200]	Time 0.519 (0.602)	Data 0.001 (0.075)	Loss 0.447 (0.469)
Epoch: [40][200/200]	Time 0.517 (0.601)	Data 0.001 (0.074)	Loss 0.598 (0.474)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.185 (0.220)	Data 0.000 (0.025)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.362627267837524
==> Statistics for epoch 41: 612 clusters
Epoch: [41][20/200]	Time 1.941 (0.636)	Data 1.393 (0.107)	Loss 0.349 (0.085)
Epoch: [41][40/200]	Time 0.507 (0.621)	Data 0.001 (0.092)	Loss 0.657 (0.279)
Epoch: [41][60/200]	Time 0.508 (0.613)	Data 0.001 (0.087)	Loss 0.506 (0.340)
Epoch: [41][80/200]	Time 0.510 (0.608)	Data 0.002 (0.083)	Loss 0.489 (0.377)
Epoch: [41][100/200]	Time 0.625 (0.605)	Data 0.002 (0.081)	Loss 0.661 (0.402)
Epoch: [41][120/200]	Time 0.510 (0.605)	Data 0.001 (0.080)	Loss 0.635 (0.425)
Epoch: [41][140/200]	Time 0.509 (0.607)	Data 0.001 (0.081)	Loss 0.431 (0.433)
Epoch: [41][160/200]	Time 0.600 (0.607)	Data 0.001 (0.080)	Loss 0.477 (0.440)
Epoch: [41][180/200]	Time 0.508 (0.605)	Data 0.001 (0.079)	Loss 0.404 (0.450)
Epoch: [41][200/200]	Time 0.513 (0.603)	Data 0.001 (0.077)	Loss 0.523 (0.460)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.183 (0.217)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.044310331344604
==> Statistics for epoch 42: 612 clusters
Epoch: [42][20/200]	Time 1.915 (0.630)	Data 1.328 (0.102)	Loss 0.901 (0.107)
Epoch: [42][40/200]	Time 0.526 (0.610)	Data 0.001 (0.086)	Loss 0.453 (0.278)
Epoch: [42][60/200]	Time 0.621 (0.603)	Data 0.001 (0.079)	Loss 0.336 (0.353)
Epoch: [42][80/200]	Time 0.507 (0.600)	Data 0.002 (0.075)	Loss 0.447 (0.403)
Epoch: [42][100/200]	Time 0.514 (0.599)	Data 0.001 (0.074)	Loss 0.369 (0.428)
Epoch: [42][120/200]	Time 0.518 (0.597)	Data 0.001 (0.073)	Loss 0.684 (0.448)
Epoch: [42][140/200]	Time 0.589 (0.597)	Data 0.001 (0.072)	Loss 0.487 (0.455)
Epoch: [42][160/200]	Time 0.515 (0.597)	Data 0.001 (0.071)	Loss 0.314 (0.464)
Epoch: [42][180/200]	Time 0.504 (0.597)	Data 0.001 (0.071)	Loss 0.505 (0.471)
Epoch: [42][200/200]	Time 0.507 (0.597)	Data 0.001 (0.071)	Loss 0.549 (0.472)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.180 (0.215)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.93120837211609
==> Statistics for epoch 43: 611 clusters
Epoch: [43][20/200]	Time 1.870 (0.622)	Data 1.313 (0.102)	Loss 0.419 (0.089)
Epoch: [43][40/200]	Time 0.508 (0.607)	Data 0.000 (0.085)	Loss 0.399 (0.295)
Epoch: [43][60/200]	Time 0.517 (0.607)	Data 0.005 (0.081)	Loss 0.802 (0.370)
Epoch: [43][80/200]	Time 0.505 (0.603)	Data 0.002 (0.077)	Loss 0.539 (0.411)
Epoch: [43][100/200]	Time 0.506 (0.602)	Data 0.001 (0.076)	Loss 0.840 (0.444)
Epoch: [43][120/200]	Time 0.623 (0.601)	Data 0.001 (0.074)	Loss 0.755 (0.452)
Epoch: [43][140/200]	Time 0.501 (0.599)	Data 0.001 (0.073)	Loss 0.734 (0.464)
Epoch: [43][160/200]	Time 0.502 (0.598)	Data 0.001 (0.072)	Loss 0.407 (0.471)
Epoch: [43][180/200]	Time 0.506 (0.597)	Data 0.001 (0.071)	Loss 0.475 (0.475)
Epoch: [43][200/200]	Time 0.509 (0.596)	Data 0.000 (0.071)	Loss 0.567 (0.481)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.184 (0.220)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.297102212905884
==> Statistics for epoch 44: 613 clusters
Epoch: [44][20/200]	Time 1.790 (0.635)	Data 1.225 (0.105)	Loss 0.498 (0.099)
Epoch: [44][40/200]	Time 0.510 (0.611)	Data 0.001 (0.086)	Loss 0.448 (0.302)
Epoch: [44][60/200]	Time 0.512 (0.608)	Data 0.001 (0.082)	Loss 0.689 (0.381)
Epoch: [44][80/200]	Time 0.510 (0.604)	Data 0.001 (0.079)	Loss 0.350 (0.414)
Epoch: [44][100/200]	Time 0.512 (0.601)	Data 0.001 (0.077)	Loss 0.540 (0.435)
Epoch: [44][120/200]	Time 0.520 (0.600)	Data 0.001 (0.074)	Loss 0.555 (0.446)
Epoch: [44][140/200]	Time 0.582 (0.599)	Data 0.001 (0.074)	Loss 0.406 (0.460)
Epoch: [44][160/200]	Time 0.504 (0.598)	Data 0.001 (0.072)	Loss 0.465 (0.468)
Epoch: [44][180/200]	Time 0.508 (0.597)	Data 0.001 (0.072)	Loss 0.496 (0.472)
Epoch: [44][200/200]	Time 0.513 (0.596)	Data 0.001 (0.071)	Loss 0.525 (0.473)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.183 (0.217)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.040666341781616
==> Statistics for epoch 45: 612 clusters
Epoch: [45][20/200]	Time 1.906 (0.646)	Data 1.346 (0.114)	Loss 0.635 (0.104)
Epoch: [45][40/200]	Time 0.504 (0.617)	Data 0.001 (0.091)	Loss 0.540 (0.314)
Epoch: [45][60/200]	Time 0.505 (0.608)	Data 0.001 (0.084)	Loss 0.630 (0.382)
Epoch: [45][80/200]	Time 0.518 (0.602)	Data 0.001 (0.080)	Loss 0.744 (0.416)
Epoch: [45][100/200]	Time 0.510 (0.602)	Data 0.001 (0.077)	Loss 0.462 (0.435)
Epoch: [45][120/200]	Time 0.600 (0.600)	Data 0.001 (0.075)	Loss 0.518 (0.453)
Epoch: [45][140/200]	Time 0.506 (0.599)	Data 0.001 (0.074)	Loss 0.533 (0.462)
Epoch: [45][160/200]	Time 0.506 (0.597)	Data 0.001 (0.073)	Loss 0.436 (0.466)
Epoch: [45][180/200]	Time 0.507 (0.596)	Data 0.001 (0.072)	Loss 0.628 (0.476)
Epoch: [45][200/200]	Time 0.619 (0.596)	Data 0.001 (0.072)	Loss 0.530 (0.486)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.217)	Data 0.000 (0.022)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.312309503555298
==> Statistics for epoch 46: 612 clusters
Epoch: [46][20/200]	Time 1.958 (0.646)	Data 1.398 (0.109)	Loss 0.643 (0.100)
Epoch: [46][40/200]	Time 0.604 (0.619)	Data 0.001 (0.086)	Loss 0.674 (0.277)
Epoch: [46][60/200]	Time 0.505 (0.608)	Data 0.001 (0.080)	Loss 0.515 (0.361)
Epoch: [46][80/200]	Time 0.507 (0.602)	Data 0.001 (0.076)	Loss 0.421 (0.403)
Epoch: [46][100/200]	Time 0.507 (0.599)	Data 0.001 (0.074)	Loss 0.426 (0.422)
Epoch: [46][120/200]	Time 0.512 (0.602)	Data 0.001 (0.076)	Loss 0.465 (0.435)
Epoch: [46][140/200]	Time 0.506 (0.601)	Data 0.001 (0.076)	Loss 0.416 (0.446)
Epoch: [46][160/200]	Time 0.514 (0.601)	Data 0.001 (0.075)	Loss 0.404 (0.456)
Epoch: [46][180/200]	Time 0.512 (0.600)	Data 0.001 (0.074)	Loss 0.567 (0.466)
Epoch: [46][200/200]	Time 0.520 (0.599)	Data 0.001 (0.074)	Loss 0.507 (0.470)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.253 (0.217)	Data 0.000 (0.021)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.851797819137573
==> Statistics for epoch 47: 612 clusters
Epoch: [47][20/200]	Time 1.965 (0.634)	Data 1.393 (0.107)	Loss 0.528 (0.087)
Epoch: [47][40/200]	Time 0.507 (0.619)	Data 0.000 (0.092)	Loss 0.489 (0.278)
Epoch: [47][60/200]	Time 0.510 (0.613)	Data 0.001 (0.087)	Loss 0.709 (0.372)
Epoch: [47][80/200]	Time 0.506 (0.606)	Data 0.001 (0.081)	Loss 0.512 (0.398)
Epoch: [47][100/200]	Time 0.636 (0.604)	Data 0.001 (0.079)	Loss 0.494 (0.419)
Epoch: [47][120/200]	Time 0.504 (0.602)	Data 0.001 (0.077)	Loss 0.704 (0.435)
Epoch: [47][140/200]	Time 0.501 (0.601)	Data 0.001 (0.076)	Loss 0.341 (0.452)
Epoch: [47][160/200]	Time 0.518 (0.600)	Data 0.001 (0.075)	Loss 0.721 (0.460)
Epoch: [47][180/200]	Time 0.533 (0.598)	Data 0.001 (0.074)	Loss 0.538 (0.466)
Epoch: [47][200/200]	Time 0.584 (0.597)	Data 0.000 (0.073)	Loss 0.404 (0.469)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.220)	Data 0.000 (0.023)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.024335384368896
==> Statistics for epoch 48: 612 clusters
Epoch: [48][20/200]	Time 1.882 (0.637)	Data 1.358 (0.109)	Loss 0.410 (0.087)
Epoch: [48][40/200]	Time 0.508 (0.621)	Data 0.001 (0.092)	Loss 0.556 (0.292)
Epoch: [48][60/200]	Time 0.502 (0.616)	Data 0.001 (0.088)	Loss 0.360 (0.364)
Epoch: [48][80/200]	Time 0.512 (0.611)	Data 0.002 (0.085)	Loss 0.608 (0.398)
Epoch: [48][100/200]	Time 0.509 (0.610)	Data 0.001 (0.083)	Loss 0.419 (0.416)
Epoch: [48][120/200]	Time 0.513 (0.609)	Data 0.001 (0.082)	Loss 0.594 (0.431)
Epoch: [48][140/200]	Time 0.515 (0.607)	Data 0.001 (0.079)	Loss 0.434 (0.443)
Epoch: [48][160/200]	Time 0.601 (0.606)	Data 0.001 (0.078)	Loss 0.518 (0.455)
Epoch: [48][180/200]	Time 0.505 (0.605)	Data 0.001 (0.078)	Loss 0.909 (0.460)
Epoch: [48][200/200]	Time 0.510 (0.605)	Data 0.001 (0.078)	Loss 0.614 (0.469)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.181 (0.220)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.57491135597229
==> Statistics for epoch 49: 613 clusters
Epoch: [49][20/200]	Time 1.961 (0.647)	Data 1.420 (0.118)	Loss 0.427 (0.087)
Epoch: [49][40/200]	Time 0.607 (0.617)	Data 0.001 (0.092)	Loss 0.578 (0.290)
Epoch: [49][60/200]	Time 0.503 (0.606)	Data 0.001 (0.081)	Loss 0.488 (0.357)
Epoch: [49][80/200]	Time 0.504 (0.601)	Data 0.001 (0.077)	Loss 0.558 (0.404)
Epoch: [49][100/200]	Time 0.608 (0.598)	Data 0.001 (0.075)	Loss 0.534 (0.426)
Epoch: [49][120/200]	Time 0.503 (0.595)	Data 0.001 (0.073)	Loss 0.580 (0.443)
Epoch: [49][140/200]	Time 0.498 (0.596)	Data 0.001 (0.073)	Loss 0.526 (0.457)
Epoch: [49][160/200]	Time 0.514 (0.595)	Data 0.001 (0.072)	Loss 0.597 (0.462)
Epoch: [49][180/200]	Time 0.504 (0.595)	Data 0.001 (0.071)	Loss 0.596 (0.466)
Epoch: [49][200/200]	Time 0.510 (0.594)	Data 0.001 (0.070)	Loss 0.404 (0.476)
Extract Features: [50/76]	Time 0.182 (0.217)	Data 0.000 (0.021)	
Mean AP: 93.6%

 * Finished epoch  49  model mAP: 93.6%  best: 93.6% *

==> Test with the best model:
=> Loaded checkpoint 'log/msmt2market/resnet152_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.181 (0.217)	Data 0.000 (0.023)	
Mean AP: 93.6%
CMC Scores:
  top-1          97.0%
  top-5          98.9%
  top-10         99.3%
Total running time:  2:11:26.926419
