==========
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='vit_tiny', pretrained_path='/home/ma-user/work/Projects/ReIDNets_Checkpoints_TransReID/ViT_Tiny_P16_2468_406080100_bs120_originaldino_noreid.pth', features=0, dropout=0, momentum=0.2, drop_path_rate=0.3, hw_ratio=2, self_norm=True, conv_stem=True, 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/vit_tiny_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
  ----------------------------------------
Using convolution stem
using drop_out rate is : 0.0
using attn_drop_out rate is : 0.0
using drop_path rate is : 0.3
Convert dino model......
Load 172 / 177 layers.
ViT Tiny Created!
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.070 (0.456)	Data 0.000 (0.047)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.681434154510498
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 536 clusters
Epoch: [0][20/200]	Time 0.180 (0.588)	Data 0.000 (0.105)	Loss 6.821 (5.486)
Epoch: [0][40/200]	Time 0.171 (0.421)	Data 0.000 (0.086)	Loss 4.701 (5.779)
Epoch: [0][60/200]	Time 0.172 (0.366)	Data 0.000 (0.081)	Loss 4.820 (5.438)
Epoch: [0][80/200]	Time 0.172 (0.336)	Data 0.000 (0.078)	Loss 4.267 (5.156)
Epoch: [0][100/200]	Time 0.170 (0.334)	Data 0.001 (0.090)	Loss 4.081 (4.906)
Epoch: [0][120/200]	Time 0.180 (0.320)	Data 0.001 (0.087)	Loss 4.094 (4.727)
Epoch: [0][140/200]	Time 0.176 (0.308)	Data 0.000 (0.083)	Loss 3.828 (4.584)
Epoch: [0][160/200]	Time 0.172 (0.302)	Data 0.000 (0.082)	Loss 3.878 (4.478)
Epoch: [0][180/200]	Time 0.175 (0.303)	Data 0.001 (0.088)	Loss 3.493 (4.361)
Epoch: [0][200/200]	Time 0.177 (0.298)	Data 0.001 (0.086)	Loss 3.303 (4.262)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.069 (0.157)	Data 0.000 (0.045)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.160412311553955
==> Statistics for epoch 1: 554 clusters
Epoch: [1][20/200]	Time 0.195 (0.278)	Data 0.001 (0.097)	Loss 3.576 (1.205)
Epoch: [1][40/200]	Time 0.189 (0.266)	Data 0.001 (0.082)	Loss 3.028 (2.227)
Epoch: [1][60/200]	Time 0.186 (0.262)	Data 0.001 (0.077)	Loss 3.619 (2.539)
Epoch: [1][80/200]	Time 0.173 (0.260)	Data 0.000 (0.075)	Loss 3.216 (2.697)
Epoch: [1][100/200]	Time 0.174 (0.258)	Data 0.000 (0.073)	Loss 2.800 (2.757)
Epoch: [1][120/200]	Time 1.616 (0.268)	Data 1.395 (0.083)	Loss 2.868 (2.787)
Epoch: [1][140/200]	Time 0.176 (0.266)	Data 0.001 (0.082)	Loss 3.563 (2.829)
Epoch: [1][160/200]	Time 0.174 (0.264)	Data 0.001 (0.080)	Loss 2.385 (2.850)
Epoch: [1][180/200]	Time 0.171 (0.264)	Data 0.000 (0.079)	Loss 2.944 (2.849)
Epoch: [1][200/200]	Time 0.176 (0.262)	Data 0.000 (0.078)	Loss 2.963 (2.857)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.068 (0.134)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.33532190322876
==> Statistics for epoch 2: 544 clusters
Epoch: [2][20/200]	Time 0.186 (0.297)	Data 0.001 (0.108)	Loss 2.919 (0.955)
Epoch: [2][40/200]	Time 0.193 (0.274)	Data 0.001 (0.087)	Loss 2.742 (1.826)
Epoch: [2][60/200]	Time 0.182 (0.265)	Data 0.001 (0.080)	Loss 2.756 (2.142)
Epoch: [2][80/200]	Time 0.173 (0.262)	Data 0.000 (0.076)	Loss 2.603 (2.296)
Epoch: [2][100/200]	Time 0.169 (0.259)	Data 0.000 (0.074)	Loss 2.851 (2.401)
Epoch: [2][120/200]	Time 1.615 (0.269)	Data 1.412 (0.085)	Loss 3.045 (2.458)
Epoch: [2][140/200]	Time 0.178 (0.268)	Data 0.000 (0.083)	Loss 2.445 (2.494)
Epoch: [2][160/200]	Time 0.179 (0.266)	Data 0.000 (0.081)	Loss 2.961 (2.519)
Epoch: [2][180/200]	Time 0.168 (0.264)	Data 0.000 (0.080)	Loss 2.362 (2.534)
Epoch: [2][200/200]	Time 0.177 (0.263)	Data 0.000 (0.079)	Loss 2.199 (2.543)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.139)	Data 0.000 (0.054)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.922269105911255
==> Statistics for epoch 3: 544 clusters
Epoch: [3][20/200]	Time 0.178 (0.287)	Data 0.001 (0.107)	Loss 2.358 (0.785)
Epoch: [3][40/200]	Time 0.174 (0.270)	Data 0.001 (0.088)	Loss 2.661 (1.639)
Epoch: [3][60/200]	Time 0.177 (0.265)	Data 0.000 (0.082)	Loss 2.282 (1.907)
Epoch: [3][80/200]	Time 0.171 (0.263)	Data 0.000 (0.080)	Loss 2.760 (2.069)
Epoch: [3][100/200]	Time 0.174 (0.262)	Data 0.000 (0.077)	Loss 2.943 (2.168)
Epoch: [3][120/200]	Time 1.656 (0.272)	Data 1.453 (0.087)	Loss 2.397 (2.232)
Epoch: [3][140/200]	Time 0.179 (0.269)	Data 0.000 (0.085)	Loss 2.546 (2.262)
Epoch: [3][160/200]	Time 0.167 (0.267)	Data 0.000 (0.083)	Loss 2.357 (2.275)
Epoch: [3][180/200]	Time 0.178 (0.265)	Data 0.000 (0.081)	Loss 2.663 (2.292)
Epoch: [3][200/200]	Time 0.176 (0.264)	Data 0.000 (0.081)	Loss 2.637 (2.308)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.106 (0.129)	Data 0.031 (0.044)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.903372764587402
==> Statistics for epoch 4: 538 clusters
Epoch: [4][20/200]	Time 0.169 (0.295)	Data 0.001 (0.111)	Loss 1.985 (0.852)
Epoch: [4][40/200]	Time 0.171 (0.272)	Data 0.000 (0.091)	Loss 2.325 (1.551)
Epoch: [4][60/200]	Time 0.172 (0.265)	Data 0.000 (0.083)	Loss 1.773 (1.788)
Epoch: [4][80/200]	Time 0.176 (0.263)	Data 0.000 (0.080)	Loss 2.335 (1.941)
Epoch: [4][100/200]	Time 0.174 (0.273)	Data 0.001 (0.091)	Loss 2.133 (2.040)
Epoch: [4][120/200]	Time 0.188 (0.270)	Data 0.001 (0.087)	Loss 2.483 (2.096)
Epoch: [4][140/200]	Time 0.169 (0.267)	Data 0.000 (0.085)	Loss 1.831 (2.121)
Epoch: [4][160/200]	Time 0.178 (0.265)	Data 0.000 (0.082)	Loss 2.369 (2.146)
Epoch: [4][180/200]	Time 0.173 (0.272)	Data 0.001 (0.089)	Loss 1.912 (2.160)
Epoch: [4][200/200]	Time 0.185 (0.271)	Data 0.001 (0.088)	Loss 2.476 (2.187)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.069 (0.133)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.07309055328369
==> Statistics for epoch 5: 521 clusters
Epoch: [5][20/200]	Time 0.171 (0.292)	Data 0.001 (0.101)	Loss 1.996 (0.744)
Epoch: [5][40/200]	Time 0.181 (0.271)	Data 0.001 (0.087)	Loss 1.922 (1.392)
Epoch: [5][60/200]	Time 0.169 (0.265)	Data 0.000 (0.081)	Loss 2.097 (1.684)
Epoch: [5][80/200]	Time 0.170 (0.262)	Data 0.000 (0.078)	Loss 2.653 (1.848)
Epoch: [5][100/200]	Time 0.173 (0.274)	Data 0.001 (0.089)	Loss 2.223 (1.922)
Epoch: [5][120/200]	Time 0.173 (0.268)	Data 0.001 (0.085)	Loss 2.281 (1.958)
Epoch: [5][140/200]	Time 0.175 (0.266)	Data 0.000 (0.082)	Loss 2.215 (1.999)
Epoch: [5][160/200]	Time 0.176 (0.264)	Data 0.000 (0.080)	Loss 1.862 (2.018)
Epoch: [5][180/200]	Time 0.295 (0.271)	Data 0.001 (0.086)	Loss 2.613 (2.042)
Epoch: [5][200/200]	Time 0.179 (0.269)	Data 0.001 (0.085)	Loss 2.132 (2.062)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.132)	Data 0.000 (0.048)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.874117136001587
==> Statistics for epoch 6: 536 clusters
Epoch: [6][20/200]	Time 0.176 (0.292)	Data 0.000 (0.100)	Loss 1.813 (0.702)
Epoch: [6][40/200]	Time 0.170 (0.270)	Data 0.001 (0.085)	Loss 2.015 (1.378)
Epoch: [6][60/200]	Time 0.169 (0.265)	Data 0.000 (0.081)	Loss 1.859 (1.632)
Epoch: [6][80/200]	Time 0.174 (0.261)	Data 0.000 (0.079)	Loss 2.198 (1.763)
Epoch: [6][100/200]	Time 0.171 (0.274)	Data 0.001 (0.090)	Loss 2.377 (1.843)
Epoch: [6][120/200]	Time 0.179 (0.270)	Data 0.001 (0.087)	Loss 2.109 (1.889)
Epoch: [6][140/200]	Time 0.183 (0.267)	Data 0.000 (0.084)	Loss 1.990 (1.920)
Epoch: [6][160/200]	Time 0.173 (0.265)	Data 0.000 (0.082)	Loss 2.278 (1.947)
Epoch: [6][180/200]	Time 0.186 (0.271)	Data 0.001 (0.088)	Loss 1.822 (1.962)
Epoch: [6][200/200]	Time 0.189 (0.269)	Data 0.001 (0.086)	Loss 2.105 (1.974)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.127 (0.142)	Data 0.049 (0.056)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.577009439468384
==> Statistics for epoch 7: 540 clusters
Epoch: [7][20/200]	Time 0.174 (0.294)	Data 0.001 (0.106)	Loss 1.965 (0.635)
Epoch: [7][40/200]	Time 0.168 (0.270)	Data 0.000 (0.089)	Loss 1.804 (1.242)
Epoch: [7][60/200]	Time 0.176 (0.263)	Data 0.000 (0.082)	Loss 2.371 (1.525)
Epoch: [7][80/200]	Time 0.174 (0.260)	Data 0.000 (0.077)	Loss 2.116 (1.633)
Epoch: [7][100/200]	Time 0.171 (0.272)	Data 0.001 (0.089)	Loss 2.574 (1.709)
Epoch: [7][120/200]	Time 0.173 (0.268)	Data 0.001 (0.085)	Loss 1.864 (1.774)
Epoch: [7][140/200]	Time 0.175 (0.265)	Data 0.000 (0.082)	Loss 2.043 (1.815)
Epoch: [7][160/200]	Time 0.166 (0.262)	Data 0.000 (0.080)	Loss 1.980 (1.840)
Epoch: [7][180/200]	Time 0.172 (0.268)	Data 0.001 (0.086)	Loss 1.981 (1.855)
Epoch: [7][200/200]	Time 0.175 (0.266)	Data 0.000 (0.084)	Loss 2.151 (1.873)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.129)	Data 0.000 (0.048)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.405498027801514
==> Statistics for epoch 8: 549 clusters
Epoch: [8][20/200]	Time 0.198 (0.290)	Data 0.001 (0.108)	Loss 1.821 (0.613)
Epoch: [8][40/200]	Time 0.168 (0.272)	Data 0.000 (0.090)	Loss 1.832 (1.235)
Epoch: [8][60/200]	Time 0.170 (0.264)	Data 0.000 (0.083)	Loss 1.999 (1.470)
Epoch: [8][80/200]	Time 0.171 (0.261)	Data 0.000 (0.079)	Loss 2.004 (1.586)
Epoch: [8][100/200]	Time 0.174 (0.258)	Data 0.000 (0.077)	Loss 1.516 (1.651)
Epoch: [8][120/200]	Time 1.515 (0.267)	Data 1.286 (0.085)	Loss 2.561 (1.717)
Epoch: [8][140/200]	Time 0.172 (0.264)	Data 0.000 (0.082)	Loss 1.752 (1.756)
Epoch: [8][160/200]	Time 0.183 (0.263)	Data 0.001 (0.081)	Loss 1.917 (1.779)
Epoch: [8][180/200]	Time 0.170 (0.262)	Data 0.000 (0.080)	Loss 2.006 (1.798)
Epoch: [8][200/200]	Time 0.166 (0.260)	Data 0.000 (0.079)	Loss 2.145 (1.808)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.118 (0.132)	Data 0.043 (0.048)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.811481952667236
==> Statistics for epoch 9: 550 clusters
Epoch: [9][20/200]	Time 0.184 (0.293)	Data 0.001 (0.104)	Loss 2.164 (0.607)
Epoch: [9][40/200]	Time 0.181 (0.271)	Data 0.001 (0.088)	Loss 1.494 (1.240)
Epoch: [9][60/200]	Time 0.181 (0.266)	Data 0.001 (0.081)	Loss 2.462 (1.451)
Epoch: [9][80/200]	Time 0.178 (0.265)	Data 0.000 (0.079)	Loss 1.743 (1.570)
Epoch: [9][100/200]	Time 0.283 (0.262)	Data 0.000 (0.077)	Loss 2.093 (1.627)
Epoch: [9][120/200]	Time 1.606 (0.271)	Data 1.394 (0.087)	Loss 1.736 (1.658)
Epoch: [9][140/200]	Time 0.179 (0.268)	Data 0.001 (0.084)	Loss 1.898 (1.687)
Epoch: [9][160/200]	Time 0.177 (0.266)	Data 0.000 (0.083)	Loss 2.094 (1.707)
Epoch: [9][180/200]	Time 0.183 (0.265)	Data 0.000 (0.082)	Loss 2.052 (1.723)
Epoch: [9][200/200]	Time 0.174 (0.263)	Data 0.000 (0.080)	Loss 1.601 (1.734)
Extract Features: [50/76]	Time 0.097 (0.130)	Data 0.015 (0.048)	
Mean AP: 84.3%

 * Finished epoch   9  model mAP: 84.3%  best: 84.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.231 (0.141)	Data 0.156 (0.056)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.36836862564087
==> Statistics for epoch 10: 552 clusters
Epoch: [10][20/200]	Time 0.175 (0.340)	Data 0.001 (0.108)	Loss 1.486 (0.507)
Epoch: [10][40/200]	Time 0.181 (0.296)	Data 0.001 (0.090)	Loss 1.775 (1.119)
Epoch: [10][60/200]	Time 0.178 (0.281)	Data 0.001 (0.082)	Loss 1.781 (1.313)
Epoch: [10][80/200]	Time 0.167 (0.273)	Data 0.000 (0.078)	Loss 2.102 (1.443)
Epoch: [10][100/200]	Time 0.172 (0.270)	Data 0.000 (0.077)	Loss 2.033 (1.525)
Epoch: [10][120/200]	Time 1.553 (0.277)	Data 1.323 (0.086)	Loss 1.287 (1.566)
Epoch: [10][140/200]	Time 0.181 (0.274)	Data 0.001 (0.084)	Loss 1.713 (1.598)
Epoch: [10][160/200]	Time 0.168 (0.271)	Data 0.001 (0.082)	Loss 1.538 (1.618)
Epoch: [10][180/200]	Time 0.177 (0.268)	Data 0.000 (0.080)	Loss 1.637 (1.633)
Epoch: [10][200/200]	Time 0.181 (0.266)	Data 0.000 (0.079)	Loss 1.933 (1.650)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.071 (0.131)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.46984601020813
==> Statistics for epoch 11: 542 clusters
Epoch: [11][20/200]	Time 0.177 (0.287)	Data 0.001 (0.105)	Loss 1.860 (0.548)
Epoch: [11][40/200]	Time 0.172 (0.269)	Data 0.001 (0.086)	Loss 1.760 (1.102)
Epoch: [11][60/200]	Time 0.173 (0.263)	Data 0.000 (0.079)	Loss 1.344 (1.327)
Epoch: [11][80/200]	Time 0.163 (0.259)	Data 0.000 (0.077)	Loss 1.586 (1.443)
Epoch: [11][100/200]	Time 0.173 (0.273)	Data 0.000 (0.089)	Loss 1.787 (1.507)
Epoch: [11][120/200]	Time 0.186 (0.269)	Data 0.001 (0.085)	Loss 1.741 (1.547)
Epoch: [11][140/200]	Time 0.173 (0.266)	Data 0.000 (0.083)	Loss 2.002 (1.577)
Epoch: [11][160/200]	Time 0.173 (0.265)	Data 0.000 (0.081)	Loss 1.878 (1.597)
Epoch: [11][180/200]	Time 0.173 (0.271)	Data 0.000 (0.088)	Loss 1.648 (1.613)
Epoch: [11][200/200]	Time 0.179 (0.269)	Data 0.001 (0.086)	Loss 1.894 (1.625)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.135)	Data 0.000 (0.051)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.73362636566162
==> Statistics for epoch 12: 553 clusters
Epoch: [12][20/200]	Time 0.181 (0.290)	Data 0.001 (0.103)	Loss 1.358 (0.495)
Epoch: [12][40/200]	Time 0.171 (0.269)	Data 0.001 (0.086)	Loss 1.812 (1.130)
Epoch: [12][60/200]	Time 0.173 (0.265)	Data 0.001 (0.081)	Loss 1.381 (1.313)
Epoch: [12][80/200]	Time 0.172 (0.263)	Data 0.000 (0.079)	Loss 1.590 (1.391)
Epoch: [12][100/200]	Time 0.172 (0.260)	Data 0.000 (0.078)	Loss 1.847 (1.471)
Epoch: [12][120/200]	Time 1.551 (0.270)	Data 1.332 (0.087)	Loss 1.767 (1.512)
Epoch: [12][140/200]	Time 0.173 (0.267)	Data 0.001 (0.084)	Loss 1.472 (1.538)
Epoch: [12][160/200]	Time 0.186 (0.266)	Data 0.001 (0.082)	Loss 1.562 (1.565)
Epoch: [12][180/200]	Time 0.173 (0.265)	Data 0.000 (0.081)	Loss 1.631 (1.582)
Epoch: [12][200/200]	Time 0.165 (0.263)	Data 0.000 (0.080)	Loss 1.424 (1.588)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.133)	Data 0.000 (0.047)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.544164180755615
==> Statistics for epoch 13: 554 clusters
Epoch: [13][20/200]	Time 0.186 (0.293)	Data 0.000 (0.103)	Loss 1.466 (0.473)
Epoch: [13][40/200]	Time 0.178 (0.272)	Data 0.000 (0.084)	Loss 1.809 (1.026)
Epoch: [13][60/200]	Time 0.176 (0.263)	Data 0.000 (0.078)	Loss 1.882 (1.206)
Epoch: [13][80/200]	Time 0.185 (0.260)	Data 0.000 (0.075)	Loss 1.783 (1.316)
Epoch: [13][100/200]	Time 0.174 (0.257)	Data 0.000 (0.073)	Loss 1.906 (1.387)
Epoch: [13][120/200]	Time 1.627 (0.268)	Data 1.397 (0.084)	Loss 1.403 (1.432)
Epoch: [13][140/200]	Time 0.184 (0.266)	Data 0.000 (0.081)	Loss 1.686 (1.466)
Epoch: [13][160/200]	Time 0.177 (0.265)	Data 0.001 (0.080)	Loss 1.336 (1.479)
Epoch: [13][180/200]	Time 0.173 (0.264)	Data 0.000 (0.079)	Loss 2.102 (1.492)
Epoch: [13][200/200]	Time 0.176 (0.262)	Data 0.000 (0.078)	Loss 1.810 (1.502)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.072 (0.135)	Data 0.000 (0.049)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.380225658416748
==> Statistics for epoch 14: 556 clusters
Epoch: [14][20/200]	Time 0.177 (0.291)	Data 0.001 (0.105)	Loss 1.822 (0.459)
Epoch: [14][40/200]	Time 0.189 (0.267)	Data 0.001 (0.085)	Loss 1.833 (0.938)
Epoch: [14][60/200]	Time 0.187 (0.261)	Data 0.001 (0.078)	Loss 1.333 (1.120)
Epoch: [14][80/200]	Time 0.176 (0.260)	Data 0.000 (0.076)	Loss 1.780 (1.239)
Epoch: [14][100/200]	Time 0.171 (0.258)	Data 0.000 (0.075)	Loss 1.627 (1.319)
Epoch: [14][120/200]	Time 1.599 (0.270)	Data 1.384 (0.086)	Loss 1.381 (1.357)
Epoch: [14][140/200]	Time 0.176 (0.268)	Data 0.001 (0.084)	Loss 1.686 (1.388)
Epoch: [14][160/200]	Time 0.179 (0.266)	Data 0.001 (0.082)	Loss 1.838 (1.402)
Epoch: [14][180/200]	Time 0.177 (0.264)	Data 0.000 (0.080)	Loss 2.006 (1.422)
Epoch: [14][200/200]	Time 0.177 (0.263)	Data 0.000 (0.079)	Loss 1.419 (1.433)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.069 (0.131)	Data 0.000 (0.046)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.28991723060608
==> Statistics for epoch 15: 553 clusters
Epoch: [15][20/200]	Time 0.186 (0.298)	Data 0.001 (0.109)	Loss 1.181 (0.394)
Epoch: [15][40/200]	Time 0.179 (0.275)	Data 0.001 (0.090)	Loss 1.711 (0.958)
Epoch: [15][60/200]	Time 0.179 (0.266)	Data 0.001 (0.082)	Loss 1.441 (1.134)
Epoch: [15][80/200]	Time 0.177 (0.264)	Data 0.000 (0.079)	Loss 1.397 (1.255)
Epoch: [15][100/200]	Time 0.167 (0.260)	Data 0.000 (0.076)	Loss 1.856 (1.305)
Epoch: [15][120/200]	Time 1.601 (0.271)	Data 1.387 (0.086)	Loss 1.438 (1.336)
Epoch: [15][140/200]	Time 0.169 (0.268)	Data 0.001 (0.084)	Loss 1.391 (1.364)
Epoch: [15][160/200]	Time 0.180 (0.267)	Data 0.000 (0.083)	Loss 1.023 (1.376)
Epoch: [15][180/200]	Time 0.172 (0.265)	Data 0.000 (0.081)	Loss 1.460 (1.384)
Epoch: [15][200/200]	Time 0.172 (0.263)	Data 0.000 (0.080)	Loss 1.383 (1.404)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.132)	Data 0.000 (0.049)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.307375192642212
==> Statistics for epoch 16: 557 clusters
Epoch: [16][20/200]	Time 0.182 (0.289)	Data 0.001 (0.101)	Loss 1.213 (0.404)
Epoch: [16][40/200]	Time 0.179 (0.268)	Data 0.001 (0.084)	Loss 1.451 (0.943)
Epoch: [16][60/200]	Time 0.183 (0.263)	Data 0.001 (0.080)	Loss 1.090 (1.108)
Epoch: [16][80/200]	Time 0.172 (0.262)	Data 0.000 (0.079)	Loss 1.414 (1.202)
Epoch: [16][100/200]	Time 0.169 (0.260)	Data 0.000 (0.077)	Loss 1.809 (1.270)
Epoch: [16][120/200]	Time 1.640 (0.270)	Data 1.430 (0.087)	Loss 1.095 (1.308)
Epoch: [16][140/200]	Time 0.173 (0.268)	Data 0.001 (0.085)	Loss 1.437 (1.325)
Epoch: [16][160/200]	Time 0.172 (0.266)	Data 0.000 (0.083)	Loss 1.665 (1.341)
Epoch: [16][180/200]	Time 0.177 (0.264)	Data 0.000 (0.082)	Loss 0.987 (1.350)
Epoch: [16][200/200]	Time 0.171 (0.262)	Data 0.000 (0.080)	Loss 1.321 (1.360)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.135)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.411478996276855
==> Statistics for epoch 17: 557 clusters
Epoch: [17][20/200]	Time 0.350 (0.288)	Data 0.001 (0.098)	Loss 1.174 (0.394)
Epoch: [17][40/200]	Time 0.186 (0.268)	Data 0.001 (0.083)	Loss 1.312 (0.873)
Epoch: [17][60/200]	Time 0.187 (0.264)	Data 0.001 (0.077)	Loss 1.714 (1.035)
Epoch: [17][80/200]	Time 0.179 (0.260)	Data 0.000 (0.075)	Loss 1.471 (1.152)
Epoch: [17][100/200]	Time 0.173 (0.259)	Data 0.000 (0.074)	Loss 1.097 (1.194)
Epoch: [17][120/200]	Time 1.496 (0.268)	Data 1.256 (0.084)	Loss 1.467 (1.229)
Epoch: [17][140/200]	Time 0.180 (0.266)	Data 0.001 (0.081)	Loss 1.325 (1.268)
Epoch: [17][160/200]	Time 0.183 (0.263)	Data 0.001 (0.079)	Loss 1.719 (1.292)
Epoch: [17][180/200]	Time 0.179 (0.262)	Data 0.000 (0.078)	Loss 1.316 (1.298)
Epoch: [17][200/200]	Time 0.273 (0.261)	Data 0.000 (0.077)	Loss 1.432 (1.308)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.073 (0.141)	Data 0.000 (0.059)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.419623851776123
==> Statistics for epoch 18: 560 clusters
Epoch: [18][20/200]	Time 0.182 (0.290)	Data 0.001 (0.109)	Loss 1.615 (0.400)
Epoch: [18][40/200]	Time 0.175 (0.273)	Data 0.000 (0.088)	Loss 1.226 (0.860)
Epoch: [18][60/200]	Time 0.176 (0.265)	Data 0.001 (0.081)	Loss 1.254 (1.058)
Epoch: [18][80/200]	Time 0.173 (0.261)	Data 0.000 (0.077)	Loss 1.775 (1.183)
Epoch: [18][100/200]	Time 0.169 (0.260)	Data 0.000 (0.076)	Loss 1.208 (1.224)
Epoch: [18][120/200]	Time 1.593 (0.270)	Data 1.372 (0.087)	Loss 1.304 (1.257)
Epoch: [18][140/200]	Time 0.171 (0.268)	Data 0.000 (0.084)	Loss 1.303 (1.279)
Epoch: [18][160/200]	Time 0.181 (0.266)	Data 0.000 (0.082)	Loss 1.304 (1.294)
Epoch: [18][180/200]	Time 0.180 (0.265)	Data 0.000 (0.081)	Loss 1.563 (1.298)
Epoch: [18][200/200]	Time 0.173 (0.263)	Data 0.000 (0.080)	Loss 1.267 (1.309)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.137)	Data 0.000 (0.054)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.15043568611145
==> Statistics for epoch 19: 554 clusters
Epoch: [19][20/200]	Time 0.178 (0.289)	Data 0.001 (0.104)	Loss 1.475 (0.415)
Epoch: [19][40/200]	Time 0.186 (0.269)	Data 0.001 (0.085)	Loss 1.919 (0.932)
Epoch: [19][60/200]	Time 0.176 (0.262)	Data 0.001 (0.081)	Loss 1.130 (1.084)
Epoch: [19][80/200]	Time 0.178 (0.261)	Data 0.000 (0.078)	Loss 1.202 (1.143)
Epoch: [19][100/200]	Time 0.171 (0.257)	Data 0.000 (0.076)	Loss 1.411 (1.196)
Epoch: [19][120/200]	Time 1.610 (0.268)	Data 1.393 (0.086)	Loss 1.477 (1.226)
Epoch: [19][140/200]	Time 0.170 (0.265)	Data 0.001 (0.084)	Loss 1.507 (1.253)
Epoch: [19][160/200]	Time 0.179 (0.264)	Data 0.001 (0.082)	Loss 1.183 (1.266)
Epoch: [19][180/200]	Time 0.183 (0.262)	Data 0.001 (0.080)	Loss 1.215 (1.271)
Epoch: [19][200/200]	Time 0.182 (0.261)	Data 0.000 (0.080)	Loss 1.117 (1.272)
Extract Features: [50/76]	Time 0.224 (0.139)	Data 0.150 (0.054)	
Mean AP: 87.5%

 * Finished epoch  19  model mAP: 87.5%  best: 87.5% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.132)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.225502490997314
==> Statistics for epoch 20: 558 clusters
Epoch: [20][20/200]	Time 0.184 (0.280)	Data 0.000 (0.100)	Loss 1.277 (0.374)
Epoch: [20][40/200]	Time 0.175 (0.269)	Data 0.001 (0.086)	Loss 1.470 (0.843)
Epoch: [20][60/200]	Time 0.178 (0.265)	Data 0.000 (0.082)	Loss 1.246 (0.950)
Epoch: [20][80/200]	Time 0.172 (0.263)	Data 0.000 (0.079)	Loss 1.171 (1.033)
Epoch: [20][100/200]	Time 0.178 (0.260)	Data 0.000 (0.077)	Loss 1.097 (1.069)
Epoch: [20][120/200]	Time 1.577 (0.271)	Data 1.340 (0.088)	Loss 1.195 (1.106)
Epoch: [20][140/200]	Time 0.178 (0.268)	Data 0.001 (0.085)	Loss 1.392 (1.128)
Epoch: [20][160/200]	Time 0.185 (0.266)	Data 0.001 (0.083)	Loss 1.129 (1.147)
Epoch: [20][180/200]	Time 0.171 (0.265)	Data 0.000 (0.081)	Loss 1.286 (1.158)
Epoch: [20][200/200]	Time 0.171 (0.263)	Data 0.000 (0.080)	Loss 1.222 (1.168)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.070 (0.136)	Data 0.000 (0.051)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.273460388183594
==> Statistics for epoch 21: 562 clusters
Epoch: [21][20/200]	Time 0.184 (0.292)	Data 0.001 (0.098)	Loss 1.324 (0.390)
Epoch: [21][40/200]	Time 0.185 (0.273)	Data 0.001 (0.083)	Loss 1.122 (0.785)
Epoch: [21][60/200]	Time 0.174 (0.264)	Data 0.001 (0.079)	Loss 1.173 (0.955)
Epoch: [21][80/200]	Time 0.171 (0.262)	Data 0.000 (0.076)	Loss 1.039 (1.038)
Epoch: [21][100/200]	Time 0.312 (0.261)	Data 0.000 (0.075)	Loss 1.711 (1.090)
Epoch: [21][120/200]	Time 1.513 (0.270)	Data 1.281 (0.086)	Loss 1.165 (1.113)
Epoch: [21][140/200]	Time 0.179 (0.268)	Data 0.001 (0.083)	Loss 1.166 (1.132)
Epoch: [21][160/200]	Time 0.176 (0.266)	Data 0.001 (0.082)	Loss 1.293 (1.146)
Epoch: [21][180/200]	Time 0.180 (0.264)	Data 0.001 (0.080)	Loss 0.913 (1.155)
Epoch: [21][200/200]	Time 0.173 (0.264)	Data 0.000 (0.079)	Loss 1.464 (1.168)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.134)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.31347393989563
==> Statistics for epoch 22: 560 clusters
Epoch: [22][20/200]	Time 0.182 (0.285)	Data 0.001 (0.104)	Loss 1.179 (0.366)
Epoch: [22][40/200]	Time 0.174 (0.269)	Data 0.001 (0.086)	Loss 1.608 (0.830)
Epoch: [22][60/200]	Time 0.187 (0.264)	Data 0.001 (0.079)	Loss 1.528 (1.009)
Epoch: [22][80/200]	Time 0.174 (0.259)	Data 0.000 (0.075)	Loss 0.876 (1.056)
Epoch: [22][100/200]	Time 0.176 (0.259)	Data 0.000 (0.075)	Loss 1.166 (1.100)
Epoch: [22][120/200]	Time 1.578 (0.268)	Data 1.365 (0.085)	Loss 1.203 (1.116)
Epoch: [22][140/200]	Time 0.176 (0.266)	Data 0.001 (0.082)	Loss 1.351 (1.133)
Epoch: [22][160/200]	Time 0.173 (0.265)	Data 0.001 (0.081)	Loss 0.859 (1.147)
Epoch: [22][180/200]	Time 0.301 (0.264)	Data 0.001 (0.080)	Loss 1.179 (1.157)
Epoch: [22][200/200]	Time 0.166 (0.262)	Data 0.000 (0.079)	Loss 1.243 (1.165)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.135)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.54799509048462
==> Statistics for epoch 23: 558 clusters
Epoch: [23][20/200]	Time 0.187 (0.288)	Data 0.000 (0.100)	Loss 0.929 (0.309)
Epoch: [23][40/200]	Time 0.184 (0.269)	Data 0.001 (0.086)	Loss 1.363 (0.737)
Epoch: [23][60/200]	Time 0.196 (0.264)	Data 0.001 (0.080)	Loss 1.427 (0.916)
Epoch: [23][80/200]	Time 0.168 (0.262)	Data 0.000 (0.078)	Loss 1.280 (0.984)
Epoch: [23][100/200]	Time 0.173 (0.260)	Data 0.000 (0.075)	Loss 1.004 (1.039)
Epoch: [23][120/200]	Time 1.579 (0.270)	Data 1.355 (0.086)	Loss 1.410 (1.055)
Epoch: [23][140/200]	Time 0.177 (0.267)	Data 0.001 (0.083)	Loss 0.753 (1.076)
Epoch: [23][160/200]	Time 0.178 (0.265)	Data 0.001 (0.081)	Loss 0.964 (1.094)
Epoch: [23][180/200]	Time 0.178 (0.263)	Data 0.000 (0.079)	Loss 1.271 (1.102)
Epoch: [23][200/200]	Time 0.176 (0.261)	Data 0.000 (0.078)	Loss 1.232 (1.113)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.227 (0.136)	Data 0.149 (0.053)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.537787675857544
==> Statistics for epoch 24: 562 clusters
Epoch: [24][20/200]	Time 0.196 (0.296)	Data 0.001 (0.108)	Loss 1.282 (0.340)
Epoch: [24][40/200]	Time 0.186 (0.271)	Data 0.001 (0.087)	Loss 0.867 (0.742)
Epoch: [24][60/200]	Time 0.178 (0.265)	Data 0.001 (0.081)	Loss 1.437 (0.921)
Epoch: [24][80/200]	Time 0.299 (0.261)	Data 0.000 (0.077)	Loss 1.102 (1.008)
Epoch: [24][100/200]	Time 0.174 (0.258)	Data 0.000 (0.075)	Loss 0.966 (1.047)
Epoch: [24][120/200]	Time 1.569 (0.269)	Data 1.343 (0.085)	Loss 1.101 (1.080)
Epoch: [24][140/200]	Time 0.306 (0.268)	Data 0.001 (0.083)	Loss 1.462 (1.097)
Epoch: [24][160/200]	Time 0.180 (0.265)	Data 0.001 (0.081)	Loss 1.249 (1.107)
Epoch: [24][180/200]	Time 0.182 (0.264)	Data 0.001 (0.080)	Loss 1.412 (1.125)
Epoch: [24][200/200]	Time 0.167 (0.262)	Data 0.000 (0.078)	Loss 1.209 (1.138)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.138)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.543086528778076
==> Statistics for epoch 25: 560 clusters
Epoch: [25][20/200]	Time 0.178 (0.294)	Data 0.001 (0.102)	Loss 1.371 (0.351)
Epoch: [25][40/200]	Time 0.183 (0.276)	Data 0.001 (0.086)	Loss 1.058 (0.766)
Epoch: [25][60/200]	Time 0.185 (0.266)	Data 0.001 (0.079)	Loss 1.273 (0.912)
Epoch: [25][80/200]	Time 0.177 (0.263)	Data 0.000 (0.076)	Loss 1.227 (0.976)
Epoch: [25][100/200]	Time 0.170 (0.260)	Data 0.000 (0.075)	Loss 1.273 (1.011)
Epoch: [25][120/200]	Time 1.628 (0.271)	Data 1.416 (0.086)	Loss 1.255 (1.037)
Epoch: [25][140/200]	Time 0.184 (0.268)	Data 0.001 (0.084)	Loss 1.502 (1.065)
Epoch: [25][160/200]	Time 0.172 (0.266)	Data 0.001 (0.082)	Loss 1.391 (1.086)
Epoch: [25][180/200]	Time 0.187 (0.265)	Data 0.001 (0.080)	Loss 1.908 (1.108)
Epoch: [25][200/200]	Time 0.304 (0.264)	Data 0.000 (0.079)	Loss 1.094 (1.117)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.075 (0.134)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.75779366493225
==> Statistics for epoch 26: 558 clusters
Epoch: [26][20/200]	Time 0.175 (0.289)	Data 0.001 (0.110)	Loss 1.253 (0.335)
Epoch: [26][40/200]	Time 0.183 (0.272)	Data 0.001 (0.088)	Loss 0.945 (0.763)
Epoch: [26][60/200]	Time 0.173 (0.267)	Data 0.000 (0.083)	Loss 1.562 (0.911)
Epoch: [26][80/200]	Time 0.183 (0.263)	Data 0.000 (0.079)	Loss 0.921 (0.972)
Epoch: [26][100/200]	Time 0.171 (0.261)	Data 0.000 (0.078)	Loss 1.317 (1.012)
Epoch: [26][120/200]	Time 1.560 (0.271)	Data 1.340 (0.087)	Loss 0.919 (1.052)
Epoch: [26][140/200]	Time 0.177 (0.269)	Data 0.001 (0.084)	Loss 0.883 (1.083)
Epoch: [26][160/200]	Time 0.181 (0.266)	Data 0.001 (0.083)	Loss 1.351 (1.100)
Epoch: [26][180/200]	Time 0.178 (0.265)	Data 0.000 (0.081)	Loss 1.596 (1.109)
Epoch: [26][200/200]	Time 0.177 (0.264)	Data 0.000 (0.079)	Loss 1.056 (1.118)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.218 (0.138)	Data 0.025 (0.057)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.461848974227905
==> Statistics for epoch 27: 565 clusters
Epoch: [27][20/200]	Time 0.202 (0.294)	Data 0.001 (0.105)	Loss 1.313 (0.325)
Epoch: [27][40/200]	Time 0.191 (0.273)	Data 0.001 (0.085)	Loss 1.236 (0.707)
Epoch: [27][60/200]	Time 0.173 (0.266)	Data 0.001 (0.079)	Loss 1.181 (0.886)
Epoch: [27][80/200]	Time 0.170 (0.262)	Data 0.000 (0.077)	Loss 1.058 (0.971)
Epoch: [27][100/200]	Time 0.174 (0.262)	Data 0.000 (0.077)	Loss 1.377 (1.036)
Epoch: [27][120/200]	Time 1.595 (0.273)	Data 1.361 (0.087)	Loss 1.345 (1.071)
Epoch: [27][140/200]	Time 0.208 (0.271)	Data 0.002 (0.085)	Loss 0.842 (1.091)
Epoch: [27][160/200]	Time 0.181 (0.268)	Data 0.001 (0.083)	Loss 0.935 (1.097)
Epoch: [27][180/200]	Time 0.182 (0.266)	Data 0.001 (0.081)	Loss 1.069 (1.115)
Epoch: [27][200/200]	Time 0.180 (0.265)	Data 0.000 (0.080)	Loss 1.439 (1.120)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.141)	Data 0.000 (0.056)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.739994287490845
==> Statistics for epoch 28: 560 clusters
Epoch: [28][20/200]	Time 0.178 (0.288)	Data 0.000 (0.102)	Loss 1.635 (0.362)
Epoch: [28][40/200]	Time 0.181 (0.266)	Data 0.000 (0.085)	Loss 0.975 (0.748)
Epoch: [28][60/200]	Time 0.174 (0.260)	Data 0.000 (0.079)	Loss 1.385 (0.909)
Epoch: [28][80/200]	Time 0.316 (0.258)	Data 0.000 (0.077)	Loss 0.950 (0.996)
Epoch: [28][100/200]	Time 0.174 (0.255)	Data 0.000 (0.075)	Loss 1.323 (1.035)
Epoch: [28][120/200]	Time 1.552 (0.266)	Data 1.322 (0.084)	Loss 1.238 (1.057)
Epoch: [28][140/200]	Time 0.169 (0.264)	Data 0.000 (0.082)	Loss 1.440 (1.082)
Epoch: [28][160/200]	Time 0.178 (0.262)	Data 0.001 (0.081)	Loss 0.952 (1.090)
Epoch: [28][180/200]	Time 0.182 (0.261)	Data 0.000 (0.079)	Loss 0.986 (1.110)
Epoch: [28][200/200]	Time 0.167 (0.260)	Data 0.000 (0.078)	Loss 1.238 (1.123)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.192 (0.136)	Data 0.000 (0.053)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.06217885017395
==> Statistics for epoch 29: 561 clusters
Epoch: [29][20/200]	Time 0.182 (0.294)	Data 0.001 (0.107)	Loss 1.369 (0.320)
Epoch: [29][40/200]	Time 0.182 (0.271)	Data 0.001 (0.088)	Loss 0.740 (0.738)
Epoch: [29][60/200]	Time 0.178 (0.264)	Data 0.001 (0.080)	Loss 1.321 (0.883)
Epoch: [29][80/200]	Time 0.181 (0.261)	Data 0.000 (0.078)	Loss 1.220 (0.959)
Epoch: [29][100/200]	Time 0.170 (0.258)	Data 0.000 (0.075)	Loss 1.184 (0.999)
Epoch: [29][120/200]	Time 1.536 (0.267)	Data 1.287 (0.084)	Loss 1.250 (1.023)
Epoch: [29][140/200]	Time 0.174 (0.265)	Data 0.001 (0.082)	Loss 1.097 (1.052)
Epoch: [29][160/200]	Time 0.183 (0.263)	Data 0.001 (0.081)	Loss 0.947 (1.068)
Epoch: [29][180/200]	Time 0.180 (0.261)	Data 0.001 (0.079)	Loss 1.199 (1.088)
Epoch: [29][200/200]	Time 0.173 (0.261)	Data 0.000 (0.079)	Loss 1.124 (1.098)
Extract Features: [50/76]	Time 0.138 (0.138)	Data 0.067 (0.057)	
Mean AP: 88.2%

 * Finished epoch  29  model mAP: 88.2%  best: 88.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.073 (0.145)	Data 0.000 (0.061)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.962512493133545
==> Statistics for epoch 30: 561 clusters
Epoch: [30][20/200]	Time 0.198 (0.299)	Data 0.001 (0.109)	Loss 1.122 (0.317)
Epoch: [30][40/200]	Time 0.179 (0.273)	Data 0.001 (0.090)	Loss 1.110 (0.710)
Epoch: [30][60/200]	Time 0.177 (0.266)	Data 0.002 (0.083)	Loss 1.140 (0.850)
Epoch: [30][80/200]	Time 0.169 (0.260)	Data 0.000 (0.080)	Loss 1.560 (0.950)
Epoch: [30][100/200]	Time 0.175 (0.258)	Data 0.000 (0.077)	Loss 1.646 (0.999)
Epoch: [30][120/200]	Time 1.548 (0.269)	Data 1.328 (0.087)	Loss 0.944 (1.028)
Epoch: [30][140/200]	Time 0.171 (0.265)	Data 0.000 (0.084)	Loss 1.243 (1.064)
Epoch: [30][160/200]	Time 0.186 (0.264)	Data 0.000 (0.082)	Loss 1.273 (1.076)
Epoch: [30][180/200]	Time 0.179 (0.263)	Data 0.000 (0.081)	Loss 1.414 (1.088)
Epoch: [30][200/200]	Time 0.282 (0.261)	Data 0.000 (0.079)	Loss 1.083 (1.096)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.139 (0.136)	Data 0.068 (0.054)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.10895609855652
==> Statistics for epoch 31: 562 clusters
Epoch: [31][20/200]	Time 0.193 (0.284)	Data 0.001 (0.104)	Loss 0.857 (0.309)
Epoch: [31][40/200]	Time 0.194 (0.268)	Data 0.000 (0.084)	Loss 1.242 (0.740)
Epoch: [31][60/200]	Time 0.184 (0.264)	Data 0.001 (0.078)	Loss 1.114 (0.889)
Epoch: [31][80/200]	Time 0.172 (0.259)	Data 0.000 (0.075)	Loss 1.050 (0.969)
Epoch: [31][100/200]	Time 0.176 (0.258)	Data 0.000 (0.073)	Loss 1.547 (1.036)
Epoch: [31][120/200]	Time 1.628 (0.268)	Data 1.426 (0.085)	Loss 1.103 (1.075)
Epoch: [31][140/200]	Time 0.183 (0.266)	Data 0.001 (0.082)	Loss 0.817 (1.095)
Epoch: [31][160/200]	Time 0.172 (0.264)	Data 0.000 (0.080)	Loss 0.952 (1.102)
Epoch: [31][180/200]	Time 0.190 (0.264)	Data 0.000 (0.079)	Loss 0.975 (1.119)
Epoch: [31][200/200]	Time 0.171 (0.262)	Data 0.000 (0.078)	Loss 1.312 (1.122)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.194 (0.134)	Data 0.118 (0.048)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.910634517669678
==> Statistics for epoch 32: 564 clusters
Epoch: [32][20/200]	Time 0.211 (0.295)	Data 0.001 (0.101)	Loss 0.856 (0.320)
Epoch: [32][40/200]	Time 0.176 (0.270)	Data 0.001 (0.084)	Loss 1.359 (0.762)
Epoch: [32][60/200]	Time 0.182 (0.265)	Data 0.001 (0.078)	Loss 1.407 (0.890)
Epoch: [32][80/200]	Time 0.179 (0.262)	Data 0.000 (0.075)	Loss 1.672 (0.958)
Epoch: [32][100/200]	Time 0.169 (0.259)	Data 0.000 (0.073)	Loss 1.095 (0.996)
Epoch: [32][120/200]	Time 1.514 (0.269)	Data 1.280 (0.083)	Loss 1.144 (1.022)
Epoch: [32][140/200]	Time 0.174 (0.266)	Data 0.001 (0.080)	Loss 1.326 (1.046)
Epoch: [32][160/200]	Time 0.187 (0.264)	Data 0.001 (0.079)	Loss 1.149 (1.059)
Epoch: [32][180/200]	Time 0.176 (0.264)	Data 0.000 (0.079)	Loss 1.217 (1.084)
Epoch: [32][200/200]	Time 0.172 (0.262)	Data 0.000 (0.078)	Loss 1.020 (1.091)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.143)	Data 0.000 (0.059)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.510985612869263
==> Statistics for epoch 33: 562 clusters
Epoch: [33][20/200]	Time 0.214 (0.290)	Data 0.000 (0.105)	Loss 1.244 (0.333)
Epoch: [33][40/200]	Time 0.183 (0.274)	Data 0.000 (0.087)	Loss 1.439 (0.731)
Epoch: [33][60/200]	Time 0.175 (0.269)	Data 0.001 (0.082)	Loss 1.453 (0.881)
Epoch: [33][80/200]	Time 0.171 (0.266)	Data 0.000 (0.079)	Loss 1.443 (0.956)
Epoch: [33][100/200]	Time 0.174 (0.262)	Data 0.000 (0.076)	Loss 1.228 (1.000)
Epoch: [33][120/200]	Time 1.536 (0.271)	Data 1.315 (0.085)	Loss 1.360 (1.049)
Epoch: [33][140/200]	Time 0.176 (0.268)	Data 0.000 (0.083)	Loss 1.134 (1.072)
Epoch: [33][160/200]	Time 0.189 (0.267)	Data 0.000 (0.081)	Loss 0.888 (1.089)
Epoch: [33][180/200]	Time 0.177 (0.265)	Data 0.000 (0.080)	Loss 1.017 (1.103)
Epoch: [33][200/200]	Time 0.166 (0.263)	Data 0.000 (0.079)	Loss 1.299 (1.108)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.107 (0.136)	Data 0.036 (0.051)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.37484359741211
==> Statistics for epoch 34: 563 clusters
Epoch: [34][20/200]	Time 0.199 (0.295)	Data 0.001 (0.107)	Loss 0.880 (0.310)
Epoch: [34][40/200]	Time 0.172 (0.274)	Data 0.001 (0.088)	Loss 1.436 (0.747)
Epoch: [34][60/200]	Time 0.186 (0.264)	Data 0.001 (0.080)	Loss 1.402 (0.884)
Epoch: [34][80/200]	Time 0.177 (0.262)	Data 0.000 (0.078)	Loss 1.526 (0.967)
Epoch: [34][100/200]	Time 0.173 (0.259)	Data 0.000 (0.077)	Loss 1.545 (1.018)
Epoch: [34][120/200]	Time 1.522 (0.269)	Data 1.302 (0.086)	Loss 1.196 (1.052)
Epoch: [34][140/200]	Time 0.179 (0.267)	Data 0.001 (0.083)	Loss 1.089 (1.079)
Epoch: [34][160/200]	Time 0.172 (0.264)	Data 0.001 (0.082)	Loss 0.854 (1.098)
Epoch: [34][180/200]	Time 0.176 (0.264)	Data 0.001 (0.080)	Loss 1.550 (1.105)
Epoch: [34][200/200]	Time 0.281 (0.262)	Data 0.000 (0.079)	Loss 1.168 (1.112)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.078 (0.138)	Data 0.000 (0.058)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.374414205551147
==> Statistics for epoch 35: 564 clusters
Epoch: [35][20/200]	Time 0.177 (0.283)	Data 0.000 (0.100)	Loss 1.383 (0.332)
Epoch: [35][40/200]	Time 0.177 (0.269)	Data 0.000 (0.085)	Loss 1.296 (0.719)
Epoch: [35][60/200]	Time 0.173 (0.264)	Data 0.000 (0.080)	Loss 0.853 (0.864)
Epoch: [35][80/200]	Time 0.176 (0.261)	Data 0.000 (0.077)	Loss 1.071 (0.938)
Epoch: [35][100/200]	Time 0.175 (0.258)	Data 0.000 (0.075)	Loss 1.337 (0.985)
Epoch: [35][120/200]	Time 1.615 (0.269)	Data 1.412 (0.086)	Loss 1.261 (1.027)
Epoch: [35][140/200]	Time 0.196 (0.266)	Data 0.000 (0.083)	Loss 1.436 (1.052)
Epoch: [35][160/200]	Time 0.181 (0.265)	Data 0.000 (0.081)	Loss 0.982 (1.067)
Epoch: [35][180/200]	Time 0.187 (0.264)	Data 0.000 (0.079)	Loss 1.030 (1.088)
Epoch: [35][200/200]	Time 0.168 (0.262)	Data 0.000 (0.078)	Loss 0.956 (1.093)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.075 (0.135)	Data 0.000 (0.045)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.484193563461304
==> Statistics for epoch 36: 565 clusters
Epoch: [36][20/200]	Time 0.194 (0.300)	Data 0.000 (0.111)	Loss 1.009 (0.317)
Epoch: [36][40/200]	Time 0.181 (0.276)	Data 0.001 (0.088)	Loss 1.166 (0.732)
Epoch: [36][60/200]	Time 0.329 (0.269)	Data 0.001 (0.081)	Loss 1.120 (0.885)
Epoch: [36][80/200]	Time 0.174 (0.264)	Data 0.000 (0.078)	Loss 1.062 (0.950)
Epoch: [36][100/200]	Time 0.171 (0.261)	Data 0.000 (0.075)	Loss 1.038 (0.996)
Epoch: [36][120/200]	Time 1.529 (0.272)	Data 1.289 (0.085)	Loss 1.561 (1.029)
Epoch: [36][140/200]	Time 0.178 (0.268)	Data 0.001 (0.082)	Loss 1.249 (1.052)
Epoch: [36][160/200]	Time 0.179 (0.267)	Data 0.001 (0.081)	Loss 0.991 (1.063)
Epoch: [36][180/200]	Time 0.178 (0.265)	Data 0.001 (0.080)	Loss 0.854 (1.079)
Epoch: [36][200/200]	Time 0.167 (0.263)	Data 0.000 (0.078)	Loss 1.116 (1.087)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.084 (0.133)	Data 0.009 (0.051)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.75965642929077
==> Statistics for epoch 37: 563 clusters
Epoch: [37][20/200]	Time 0.174 (0.301)	Data 0.000 (0.109)	Loss 1.088 (0.310)
Epoch: [37][40/200]	Time 0.181 (0.274)	Data 0.001 (0.090)	Loss 1.140 (0.728)
Epoch: [37][60/200]	Time 0.177 (0.267)	Data 0.001 (0.083)	Loss 0.856 (0.866)
Epoch: [37][80/200]	Time 0.169 (0.264)	Data 0.000 (0.080)	Loss 1.276 (0.953)
Epoch: [37][100/200]	Time 0.174 (0.261)	Data 0.000 (0.078)	Loss 1.265 (0.997)
Epoch: [37][120/200]	Time 1.504 (0.273)	Data 1.274 (0.088)	Loss 1.188 (1.017)
Epoch: [37][140/200]	Time 0.176 (0.270)	Data 0.001 (0.085)	Loss 1.097 (1.036)
Epoch: [37][160/200]	Time 0.176 (0.268)	Data 0.001 (0.084)	Loss 0.763 (1.047)
Epoch: [37][180/200]	Time 0.184 (0.266)	Data 0.001 (0.082)	Loss 1.040 (1.060)
Epoch: [37][200/200]	Time 0.186 (0.265)	Data 0.000 (0.081)	Loss 1.254 (1.063)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.132)	Data 0.000 (0.050)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.130719900131226
==> Statistics for epoch 38: 565 clusters
Epoch: [38][20/200]	Time 0.195 (0.299)	Data 0.001 (0.109)	Loss 1.146 (0.292)
Epoch: [38][40/200]	Time 0.188 (0.277)	Data 0.001 (0.090)	Loss 0.921 (0.660)
Epoch: [38][60/200]	Time 0.176 (0.266)	Data 0.001 (0.081)	Loss 1.250 (0.795)
Epoch: [38][80/200]	Time 0.173 (0.263)	Data 0.000 (0.078)	Loss 1.070 (0.885)
Epoch: [38][100/200]	Time 0.180 (0.259)	Data 0.000 (0.076)	Loss 1.347 (0.952)
Epoch: [38][120/200]	Time 1.543 (0.269)	Data 1.302 (0.085)	Loss 1.459 (0.989)
Epoch: [38][140/200]	Time 0.176 (0.267)	Data 0.001 (0.082)	Loss 1.113 (1.009)
Epoch: [38][160/200]	Time 0.175 (0.265)	Data 0.001 (0.081)	Loss 1.429 (1.031)
Epoch: [38][180/200]	Time 0.176 (0.263)	Data 0.001 (0.080)	Loss 0.878 (1.048)
Epoch: [38][200/200]	Time 0.171 (0.262)	Data 0.000 (0.079)	Loss 0.858 (1.061)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.135)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.322461128234863
==> Statistics for epoch 39: 564 clusters
Epoch: [39][20/200]	Time 0.183 (0.292)	Data 0.001 (0.103)	Loss 1.037 (0.353)
Epoch: [39][40/200]	Time 0.179 (0.268)	Data 0.001 (0.084)	Loss 1.092 (0.721)
Epoch: [39][60/200]	Time 0.178 (0.263)	Data 0.001 (0.079)	Loss 1.006 (0.874)
Epoch: [39][80/200]	Time 0.181 (0.262)	Data 0.000 (0.077)	Loss 0.919 (0.922)
Epoch: [39][100/200]	Time 0.176 (0.259)	Data 0.000 (0.075)	Loss 1.098 (0.967)
Epoch: [39][120/200]	Time 1.566 (0.270)	Data 1.344 (0.086)	Loss 0.793 (0.985)
Epoch: [39][140/200]	Time 0.174 (0.267)	Data 0.001 (0.084)	Loss 1.132 (1.017)
Epoch: [39][160/200]	Time 0.184 (0.266)	Data 0.001 (0.082)	Loss 1.700 (1.042)
Epoch: [39][180/200]	Time 0.184 (0.265)	Data 0.001 (0.080)	Loss 1.211 (1.057)
Epoch: [39][200/200]	Time 0.177 (0.263)	Data 0.000 (0.079)	Loss 0.964 (1.058)
Extract Features: [50/76]	Time 0.082 (0.138)	Data 0.000 (0.056)	
Mean AP: 88.2%

 * Finished epoch  39  model mAP: 88.2%  best: 88.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.070 (0.140)	Data 0.000 (0.055)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.297971487045288
==> Statistics for epoch 40: 559 clusters
Epoch: [40][20/200]	Time 0.189 (0.298)	Data 0.001 (0.109)	Loss 0.875 (0.282)
Epoch: [40][40/200]	Time 0.174 (0.276)	Data 0.001 (0.089)	Loss 0.994 (0.672)
Epoch: [40][60/200]	Time 0.182 (0.267)	Data 0.001 (0.083)	Loss 1.083 (0.830)
Epoch: [40][80/200]	Time 0.174 (0.263)	Data 0.000 (0.078)	Loss 1.560 (0.922)
Epoch: [40][100/200]	Time 0.175 (0.261)	Data 0.000 (0.077)	Loss 1.549 (0.965)
Epoch: [40][120/200]	Time 1.610 (0.271)	Data 1.413 (0.088)	Loss 1.286 (1.000)
Epoch: [40][140/200]	Time 0.177 (0.267)	Data 0.001 (0.084)	Loss 1.226 (1.020)
Epoch: [40][160/200]	Time 0.173 (0.266)	Data 0.001 (0.082)	Loss 1.466 (1.038)
Epoch: [40][180/200]	Time 0.175 (0.264)	Data 0.001 (0.081)	Loss 1.013 (1.043)
Epoch: [40][200/200]	Time 0.171 (0.263)	Data 0.000 (0.079)	Loss 1.043 (1.047)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.070 (0.133)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.402073621749878
==> Statistics for epoch 41: 560 clusters
Epoch: [41][20/200]	Time 0.187 (0.289)	Data 0.001 (0.108)	Loss 1.045 (0.325)
Epoch: [41][40/200]	Time 0.190 (0.275)	Data 0.001 (0.089)	Loss 1.032 (0.730)
Epoch: [41][60/200]	Time 0.180 (0.269)	Data 0.001 (0.083)	Loss 1.285 (0.875)
Epoch: [41][80/200]	Time 0.295 (0.264)	Data 0.000 (0.078)	Loss 1.076 (0.932)
Epoch: [41][100/200]	Time 0.176 (0.260)	Data 0.000 (0.076)	Loss 1.065 (0.969)
Epoch: [41][120/200]	Time 1.543 (0.271)	Data 1.304 (0.086)	Loss 0.924 (0.992)
Epoch: [41][140/200]	Time 0.172 (0.267)	Data 0.001 (0.083)	Loss 1.250 (1.005)
Epoch: [41][160/200]	Time 0.182 (0.266)	Data 0.000 (0.082)	Loss 0.911 (1.021)
Epoch: [41][180/200]	Time 0.170 (0.264)	Data 0.000 (0.080)	Loss 0.958 (1.038)
Epoch: [41][200/200]	Time 0.173 (0.262)	Data 0.000 (0.079)	Loss 1.296 (1.040)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.077 (0.142)	Data 0.000 (0.059)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.14079713821411
==> Statistics for epoch 42: 561 clusters
Epoch: [42][20/200]	Time 0.185 (0.283)	Data 0.001 (0.101)	Loss 1.125 (0.301)
Epoch: [42][40/200]	Time 0.178 (0.272)	Data 0.001 (0.086)	Loss 1.122 (0.685)
Epoch: [42][60/200]	Time 0.182 (0.264)	Data 0.000 (0.079)	Loss 1.081 (0.817)
Epoch: [42][80/200]	Time 0.175 (0.260)	Data 0.000 (0.076)	Loss 1.474 (0.880)
Epoch: [42][100/200]	Time 0.173 (0.258)	Data 0.000 (0.075)	Loss 0.974 (0.929)
Epoch: [42][120/200]	Time 1.630 (0.270)	Data 1.425 (0.086)	Loss 1.361 (0.960)
Epoch: [42][140/200]	Time 0.182 (0.267)	Data 0.001 (0.084)	Loss 1.289 (0.991)
Epoch: [42][160/200]	Time 0.177 (0.265)	Data 0.001 (0.081)	Loss 1.243 (1.015)
Epoch: [42][180/200]	Time 0.189 (0.264)	Data 0.001 (0.080)	Loss 1.011 (1.026)
Epoch: [42][200/200]	Time 0.297 (0.263)	Data 0.000 (0.079)	Loss 1.031 (1.040)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.075 (0.137)	Data 0.000 (0.055)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.219644784927368
==> Statistics for epoch 43: 561 clusters
Epoch: [43][20/200]	Time 0.174 (0.295)	Data 0.001 (0.106)	Loss 0.783 (0.308)
Epoch: [43][40/200]	Time 0.177 (0.276)	Data 0.001 (0.088)	Loss 1.433 (0.675)
Epoch: [43][60/200]	Time 0.185 (0.267)	Data 0.001 (0.083)	Loss 1.270 (0.838)
Epoch: [43][80/200]	Time 0.168 (0.264)	Data 0.000 (0.080)	Loss 1.347 (0.931)
Epoch: [43][100/200]	Time 0.173 (0.261)	Data 0.000 (0.077)	Loss 1.307 (0.978)
Epoch: [43][120/200]	Time 1.610 (0.271)	Data 1.378 (0.086)	Loss 1.136 (1.010)
Epoch: [43][140/200]	Time 0.172 (0.268)	Data 0.001 (0.084)	Loss 0.968 (1.023)
Epoch: [43][160/200]	Time 0.172 (0.266)	Data 0.000 (0.082)	Loss 0.972 (1.045)
Epoch: [43][180/200]	Time 0.186 (0.264)	Data 0.000 (0.081)	Loss 1.342 (1.062)
Epoch: [43][200/200]	Time 0.173 (0.262)	Data 0.000 (0.079)	Loss 1.114 (1.071)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.070 (0.137)	Data 0.000 (0.055)	
Computing jaccard distance...
Jaccard distance computing time cost: 17.965940237045288
==> Statistics for epoch 44: 562 clusters
Epoch: [44][20/200]	Time 0.190 (0.294)	Data 0.001 (0.108)	Loss 0.867 (0.298)
Epoch: [44][40/200]	Time 0.174 (0.272)	Data 0.001 (0.088)	Loss 1.334 (0.727)
Epoch: [44][60/200]	Time 0.185 (0.265)	Data 0.001 (0.081)	Loss 0.914 (0.847)
Epoch: [44][80/200]	Time 0.178 (0.262)	Data 0.000 (0.077)	Loss 1.223 (0.919)
Epoch: [44][100/200]	Time 0.175 (0.260)	Data 0.000 (0.077)	Loss 1.499 (0.965)
Epoch: [44][120/200]	Time 1.616 (0.270)	Data 1.411 (0.087)	Loss 1.366 (0.993)
Epoch: [44][140/200]	Time 0.171 (0.267)	Data 0.000 (0.084)	Loss 0.909 (1.013)
Epoch: [44][160/200]	Time 0.174 (0.265)	Data 0.000 (0.082)	Loss 1.001 (1.021)
Epoch: [44][180/200]	Time 0.345 (0.263)	Data 0.000 (0.080)	Loss 1.003 (1.043)
Epoch: [44][200/200]	Time 0.174 (0.261)	Data 0.000 (0.078)	Loss 0.774 (1.052)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.121 (0.136)	Data 0.050 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.34621810913086
==> Statistics for epoch 45: 564 clusters
Epoch: [45][20/200]	Time 0.189 (0.291)	Data 0.001 (0.099)	Loss 0.895 (0.301)
Epoch: [45][40/200]	Time 0.176 (0.270)	Data 0.001 (0.082)	Loss 0.745 (0.715)
Epoch: [45][60/200]	Time 0.192 (0.262)	Data 0.001 (0.077)	Loss 0.887 (0.842)
Epoch: [45][80/200]	Time 0.168 (0.260)	Data 0.000 (0.076)	Loss 1.137 (0.942)
Epoch: [45][100/200]	Time 0.169 (0.258)	Data 0.000 (0.074)	Loss 1.323 (0.990)
Epoch: [45][120/200]	Time 1.616 (0.268)	Data 1.398 (0.084)	Loss 1.309 (1.018)
Epoch: [45][140/200]	Time 0.177 (0.265)	Data 0.001 (0.082)	Loss 1.042 (1.030)
Epoch: [45][160/200]	Time 0.180 (0.263)	Data 0.001 (0.080)	Loss 0.992 (1.042)
Epoch: [45][180/200]	Time 0.180 (0.262)	Data 0.001 (0.078)	Loss 1.399 (1.057)
Epoch: [45][200/200]	Time 0.176 (0.261)	Data 0.000 (0.078)	Loss 1.004 (1.063)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.136)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.99934720993042
==> Statistics for epoch 46: 563 clusters
Epoch: [46][20/200]	Time 0.184 (0.300)	Data 0.001 (0.110)	Loss 1.041 (0.301)
Epoch: [46][40/200]	Time 0.186 (0.278)	Data 0.001 (0.092)	Loss 1.077 (0.681)
Epoch: [46][60/200]	Time 0.181 (0.267)	Data 0.001 (0.083)	Loss 1.341 (0.837)
Epoch: [46][80/200]	Time 0.170 (0.262)	Data 0.000 (0.078)	Loss 1.244 (0.900)
Epoch: [46][100/200]	Time 0.176 (0.261)	Data 0.000 (0.076)	Loss 1.362 (0.940)
Epoch: [46][120/200]	Time 1.583 (0.270)	Data 1.354 (0.086)	Loss 0.799 (0.962)
Epoch: [46][140/200]	Time 0.182 (0.267)	Data 0.002 (0.082)	Loss 1.103 (0.983)
Epoch: [46][160/200]	Time 0.182 (0.266)	Data 0.001 (0.081)	Loss 1.336 (1.005)
Epoch: [46][180/200]	Time 0.171 (0.264)	Data 0.001 (0.080)	Loss 0.816 (1.017)
Epoch: [46][200/200]	Time 0.174 (0.263)	Data 0.000 (0.078)	Loss 1.182 (1.030)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.132)	Data 0.000 (0.051)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.270344495773315
==> Statistics for epoch 47: 563 clusters
Epoch: [47][20/200]	Time 0.183 (0.280)	Data 0.001 (0.097)	Loss 0.871 (0.289)
Epoch: [47][40/200]	Time 0.184 (0.265)	Data 0.000 (0.081)	Loss 0.797 (0.705)
Epoch: [47][60/200]	Time 0.175 (0.262)	Data 0.001 (0.078)	Loss 1.247 (0.882)
Epoch: [47][80/200]	Time 0.165 (0.258)	Data 0.000 (0.076)	Loss 1.310 (0.961)
Epoch: [47][100/200]	Time 0.173 (0.258)	Data 0.000 (0.074)	Loss 1.076 (1.009)
Epoch: [47][120/200]	Time 1.564 (0.267)	Data 1.333 (0.084)	Loss 1.566 (1.041)
Epoch: [47][140/200]	Time 0.179 (0.266)	Data 0.001 (0.082)	Loss 1.127 (1.058)
Epoch: [47][160/200]	Time 0.171 (0.265)	Data 0.001 (0.080)	Loss 1.183 (1.071)
Epoch: [47][180/200]	Time 0.173 (0.264)	Data 0.000 (0.079)	Loss 0.829 (1.075)
Epoch: [47][200/200]	Time 0.176 (0.263)	Data 0.000 (0.078)	Loss 1.289 (1.078)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.076 (0.135)	Data 0.000 (0.052)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.081141233444214
==> Statistics for epoch 48: 562 clusters
Epoch: [48][20/200]	Time 0.187 (0.286)	Data 0.001 (0.103)	Loss 1.191 (0.315)
Epoch: [48][40/200]	Time 0.185 (0.271)	Data 0.001 (0.088)	Loss 1.104 (0.711)
Epoch: [48][60/200]	Time 0.173 (0.265)	Data 0.001 (0.082)	Loss 0.899 (0.842)
Epoch: [48][80/200]	Time 0.175 (0.263)	Data 0.000 (0.079)	Loss 1.307 (0.915)
Epoch: [48][100/200]	Time 0.176 (0.259)	Data 0.000 (0.076)	Loss 1.455 (0.969)
Epoch: [48][120/200]	Time 1.626 (0.270)	Data 1.413 (0.086)	Loss 0.830 (0.997)
Epoch: [48][140/200]	Time 0.186 (0.267)	Data 0.001 (0.083)	Loss 0.996 (1.018)
Epoch: [48][160/200]	Time 0.170 (0.265)	Data 0.001 (0.082)	Loss 1.420 (1.028)
Epoch: [48][180/200]	Time 0.183 (0.263)	Data 0.001 (0.080)	Loss 1.588 (1.047)
Epoch: [48][200/200]	Time 0.176 (0.263)	Data 0.000 (0.080)	Loss 1.009 (1.060)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.074 (0.139)	Data 0.000 (0.054)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.24273109436035
==> Statistics for epoch 49: 564 clusters
Epoch: [49][20/200]	Time 0.181 (0.282)	Data 0.001 (0.101)	Loss 1.113 (0.308)
Epoch: [49][40/200]	Time 0.177 (0.269)	Data 0.001 (0.085)	Loss 1.416 (0.704)
Epoch: [49][60/200]	Time 0.168 (0.264)	Data 0.000 (0.080)	Loss 0.798 (0.862)
Epoch: [49][80/200]	Time 0.179 (0.259)	Data 0.000 (0.077)	Loss 0.872 (0.936)
Epoch: [49][100/200]	Time 0.174 (0.259)	Data 0.000 (0.075)	Loss 1.238 (0.988)
Epoch: [49][120/200]	Time 1.635 (0.268)	Data 1.435 (0.085)	Loss 1.271 (1.018)
Epoch: [49][140/200]	Time 0.181 (0.266)	Data 0.000 (0.083)	Loss 1.039 (1.034)
Epoch: [49][160/200]	Time 0.181 (0.264)	Data 0.000 (0.081)	Loss 0.750 (1.041)
Epoch: [49][180/200]	Time 0.177 (0.263)	Data 0.000 (0.079)	Loss 0.866 (1.047)
Epoch: [49][200/200]	Time 0.173 (0.261)	Data 0.000 (0.078)	Loss 0.908 (1.048)
Extract Features: [50/76]	Time 0.082 (0.129)	Data 0.000 (0.044)	
Mean AP: 88.3%

 * Finished epoch  49  model mAP: 88.3%  best: 88.3% *

==> Test with the best model:
=> Loaded checkpoint 'log/market/vit_tiny_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.082 (0.137)	Data 0.003 (0.052)	
Mean AP: 88.3%
CMC Scores:
  top-1          94.5%
  top-5          97.9%
  top-10         98.7%
Total running time:  1:09:54.782042
