==========
Args:Namespace(dataset='market1501', batch_size=256, workers=4, height=256, width=128, num_instances=8, eps=0.6, eps_gap=0.02, k1=30, k2=6, arch='resnet_ibn50a', pretrained_path='/home/ma-user/work/Projects/ReIDNets_Checkpoints_TransReID/ResNet50_IBN.pth', features=0, dropout=0, momentum=0.2, drop_path_rate=0.3, hw_ratio=2, self_norm=True, conv_stem=False, lr=0.00035, weight_decay=0.0005, optimizer='SGD', epochs=50, iters=200, step_size=20, seed=1, print_freq=20, eval_step=10, temp=0.05, data_dir='/home/ma-user/work/Projects/ReIDNet_Finetune/CContrast/data', logs_dir='log/market/resnet50_ibn_cion', pooling_type='gem', use_hard=True)
==========
==> Load unlabeled dataset
=> Market1501 loaded
Dataset statistics:
  ----------------------------------------
  subset   | # ids | # images | # cameras
  ----------------------------------------
  train    |   751 |    12936 |         6
  query    |   750 |     3368 |         6
  gallery  |   751 |    15913 |         6
  ----------------------------------------
pooling_type: gem
checkpoint loaded!
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.118 (0.470)	Data 0.025 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 21.758543491363525
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 637 clusters
Epoch: [0][20/200]	Time 1.682 (0.693)	Data 1.387 (0.109)	Loss 4.378 (2.821)
Epoch: [0][40/200]	Time 0.258 (0.512)	Data 0.001 (0.088)	Loss 2.902 (3.427)
Epoch: [0][60/200]	Time 0.257 (0.451)	Data 0.001 (0.080)	Loss 3.087 (3.322)
Epoch: [0][80/200]	Time 0.259 (0.422)	Data 0.001 (0.078)	Loss 2.828 (3.254)
Epoch: [0][100/200]	Time 0.257 (0.405)	Data 0.001 (0.077)	Loss 2.714 (3.180)
Epoch: [0][120/200]	Time 0.255 (0.393)	Data 0.001 (0.075)	Loss 2.378 (3.097)
Epoch: [0][140/200]	Time 0.256 (0.385)	Data 0.001 (0.074)	Loss 2.612 (3.038)
Epoch: [0][160/200]	Time 0.266 (0.378)	Data 0.001 (0.074)	Loss 2.214 (2.982)
Epoch: [0][180/200]	Time 0.255 (0.373)	Data 0.000 (0.073)	Loss 2.618 (2.924)
Epoch: [0][200/200]	Time 0.259 (0.368)	Data 0.001 (0.072)	Loss 2.355 (2.880)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.139)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 21.084521055221558
==> Statistics for epoch 1: 616 clusters
Epoch: [1][20/200]	Time 1.678 (0.373)	Data 1.365 (0.108)	Loss 2.253 (0.662)
Epoch: [1][40/200]	Time 0.265 (0.358)	Data 0.001 (0.092)	Loss 2.508 (1.537)
Epoch: [1][60/200]	Time 0.281 (0.350)	Data 0.001 (0.084)	Loss 2.583 (1.836)
Epoch: [1][80/200]	Time 0.263 (0.347)	Data 0.001 (0.080)	Loss 2.727 (1.966)
Epoch: [1][100/200]	Time 0.276 (0.344)	Data 0.001 (0.078)	Loss 2.056 (2.040)
Epoch: [1][120/200]	Time 0.262 (0.343)	Data 0.001 (0.077)	Loss 2.002 (2.087)
Epoch: [1][140/200]	Time 0.259 (0.342)	Data 0.001 (0.076)	Loss 1.818 (2.109)
Epoch: [1][160/200]	Time 0.380 (0.341)	Data 0.001 (0.075)	Loss 2.585 (2.134)
Epoch: [1][180/200]	Time 0.257 (0.340)	Data 0.001 (0.074)	Loss 2.088 (2.141)
Epoch: [1][200/200]	Time 0.257 (0.340)	Data 0.001 (0.074)	Loss 2.244 (2.146)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.651599645614624
==> Statistics for epoch 2: 590 clusters
Epoch: [2][20/200]	Time 0.275 (0.374)	Data 0.001 (0.111)	Loss 2.040 (0.649)
Epoch: [2][40/200]	Time 0.257 (0.355)	Data 0.001 (0.090)	Loss 2.313 (1.371)
Epoch: [2][60/200]	Time 0.262 (0.350)	Data 0.001 (0.084)	Loss 1.945 (1.631)
Epoch: [2][80/200]	Time 0.260 (0.347)	Data 0.001 (0.082)	Loss 2.498 (1.762)
Epoch: [2][100/200]	Time 0.263 (0.345)	Data 0.001 (0.079)	Loss 1.861 (1.806)
Epoch: [2][120/200]	Time 0.257 (0.343)	Data 0.000 (0.077)	Loss 2.424 (1.866)
Epoch: [2][140/200]	Time 0.258 (0.342)	Data 0.000 (0.076)	Loss 2.105 (1.885)
Epoch: [2][160/200]	Time 0.259 (0.342)	Data 0.000 (0.075)	Loss 2.371 (1.918)
Epoch: [2][180/200]	Time 0.260 (0.341)	Data 0.000 (0.075)	Loss 1.981 (1.929)
Epoch: [2][200/200]	Time 0.263 (0.349)	Data 0.001 (0.082)	Loss 1.685 (1.939)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.10366725921631
==> Statistics for epoch 3: 598 clusters
Epoch: [3][20/200]	Time 0.266 (0.369)	Data 0.001 (0.103)	Loss 1.682 (0.546)
Epoch: [3][40/200]	Time 0.267 (0.354)	Data 0.001 (0.088)	Loss 1.970 (1.248)
Epoch: [3][60/200]	Time 0.261 (0.349)	Data 0.001 (0.083)	Loss 2.271 (1.511)
Epoch: [3][80/200]	Time 0.259 (0.346)	Data 0.001 (0.079)	Loss 2.224 (1.637)
Epoch: [3][100/200]	Time 0.259 (0.344)	Data 0.001 (0.077)	Loss 1.814 (1.701)
Epoch: [3][120/200]	Time 0.261 (0.343)	Data 0.000 (0.076)	Loss 1.954 (1.737)
Epoch: [3][140/200]	Time 0.260 (0.342)	Data 0.000 (0.075)	Loss 1.851 (1.762)
Epoch: [3][160/200]	Time 0.258 (0.342)	Data 0.000 (0.075)	Loss 1.706 (1.786)
Epoch: [3][180/200]	Time 0.259 (0.342)	Data 0.000 (0.075)	Loss 2.008 (1.808)
Epoch: [3][200/200]	Time 0.262 (0.348)	Data 0.001 (0.081)	Loss 1.729 (1.813)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.145 (0.135)	Data 0.052 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.15847897529602
==> Statistics for epoch 4: 595 clusters
Epoch: [4][20/200]	Time 0.264 (0.378)	Data 0.001 (0.111)	Loss 1.960 (0.496)
Epoch: [4][40/200]	Time 0.391 (0.361)	Data 0.001 (0.094)	Loss 1.897 (1.154)
Epoch: [4][60/200]	Time 0.257 (0.353)	Data 0.001 (0.089)	Loss 2.028 (1.384)
Epoch: [4][80/200]	Time 0.263 (0.350)	Data 0.001 (0.084)	Loss 1.602 (1.488)
Epoch: [4][100/200]	Time 0.261 (0.348)	Data 0.001 (0.082)	Loss 2.403 (1.568)
Epoch: [4][120/200]	Time 0.263 (0.347)	Data 0.000 (0.081)	Loss 1.545 (1.598)
Epoch: [4][140/200]	Time 0.259 (0.346)	Data 0.000 (0.080)	Loss 2.001 (1.642)
Epoch: [4][160/200]	Time 0.262 (0.345)	Data 0.000 (0.079)	Loss 1.918 (1.658)
Epoch: [4][180/200]	Time 0.258 (0.345)	Data 0.000 (0.079)	Loss 1.757 (1.675)
Epoch: [4][200/200]	Time 0.263 (0.353)	Data 0.001 (0.087)	Loss 1.941 (1.680)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.132)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.361576318740845
==> Statistics for epoch 5: 585 clusters
Epoch: [5][20/200]	Time 0.262 (0.380)	Data 0.001 (0.111)	Loss 1.672 (0.457)
Epoch: [5][40/200]	Time 0.303 (0.361)	Data 0.001 (0.094)	Loss 2.007 (1.070)
Epoch: [5][60/200]	Time 0.261 (0.354)	Data 0.001 (0.087)	Loss 1.868 (1.283)
Epoch: [5][80/200]	Time 0.263 (0.351)	Data 0.001 (0.083)	Loss 1.487 (1.392)
Epoch: [5][100/200]	Time 0.267 (0.349)	Data 0.001 (0.081)	Loss 1.576 (1.458)
Epoch: [5][120/200]	Time 0.265 (0.348)	Data 0.000 (0.079)	Loss 1.985 (1.504)
Epoch: [5][140/200]	Time 0.257 (0.346)	Data 0.000 (0.078)	Loss 1.906 (1.528)
Epoch: [5][160/200]	Time 0.258 (0.345)	Data 0.000 (0.076)	Loss 1.967 (1.543)
Epoch: [5][180/200]	Time 0.259 (0.345)	Data 0.000 (0.076)	Loss 1.489 (1.556)
Epoch: [5][200/200]	Time 0.277 (0.352)	Data 0.001 (0.083)	Loss 1.836 (1.572)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.136)	Data 0.000 (0.039)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.158761739730835
==> Statistics for epoch 6: 574 clusters
Epoch: [6][20/200]	Time 0.275 (0.383)	Data 0.001 (0.109)	Loss 1.530 (0.513)
Epoch: [6][40/200]	Time 0.261 (0.360)	Data 0.001 (0.091)	Loss 1.196 (1.030)
Epoch: [6][60/200]	Time 0.374 (0.351)	Data 0.001 (0.083)	Loss 1.650 (1.235)
Epoch: [6][80/200]	Time 0.259 (0.347)	Data 0.000 (0.080)	Loss 1.587 (1.327)
Epoch: [6][100/200]	Time 0.259 (0.347)	Data 0.000 (0.080)	Loss 1.956 (1.392)
Epoch: [6][120/200]	Time 1.751 (0.358)	Data 1.453 (0.091)	Loss 1.665 (1.424)
Epoch: [6][140/200]	Time 0.265 (0.355)	Data 0.001 (0.088)	Loss 1.241 (1.441)
Epoch: [6][160/200]	Time 0.259 (0.354)	Data 0.001 (0.087)	Loss 1.653 (1.463)
Epoch: [6][180/200]	Time 0.257 (0.353)	Data 0.001 (0.085)	Loss 1.592 (1.474)
Epoch: [6][200/200]	Time 0.258 (0.351)	Data 0.000 (0.084)	Loss 1.840 (1.488)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.138)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.789713382720947
==> Statistics for epoch 7: 587 clusters
Epoch: [7][20/200]	Time 0.260 (0.379)	Data 0.001 (0.109)	Loss 1.611 (0.392)
Epoch: [7][40/200]	Time 0.261 (0.358)	Data 0.001 (0.089)	Loss 1.613 (0.927)
Epoch: [7][60/200]	Time 0.258 (0.354)	Data 0.001 (0.085)	Loss 1.340 (1.114)
Epoch: [7][80/200]	Time 0.256 (0.351)	Data 0.001 (0.083)	Loss 2.060 (1.243)
Epoch: [7][100/200]	Time 0.257 (0.348)	Data 0.001 (0.081)	Loss 1.385 (1.303)
Epoch: [7][120/200]	Time 0.256 (0.346)	Data 0.000 (0.079)	Loss 1.659 (1.337)
Epoch: [7][140/200]	Time 0.259 (0.344)	Data 0.000 (0.077)	Loss 1.451 (1.363)
Epoch: [7][160/200]	Time 0.262 (0.343)	Data 0.000 (0.077)	Loss 1.301 (1.378)
Epoch: [7][180/200]	Time 0.257 (0.342)	Data 0.000 (0.076)	Loss 1.670 (1.401)
Epoch: [7][200/200]	Time 0.260 (0.350)	Data 0.001 (0.083)	Loss 1.591 (1.411)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.138)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.225066423416138
==> Statistics for epoch 8: 597 clusters
Epoch: [8][20/200]	Time 0.254 (0.373)	Data 0.001 (0.103)	Loss 1.041 (0.327)
Epoch: [8][40/200]	Time 0.266 (0.356)	Data 0.002 (0.090)	Loss 1.264 (0.902)
Epoch: [8][60/200]	Time 0.259 (0.351)	Data 0.001 (0.085)	Loss 1.337 (1.078)
Epoch: [8][80/200]	Time 0.260 (0.347)	Data 0.001 (0.081)	Loss 1.525 (1.172)
Epoch: [8][100/200]	Time 0.270 (0.343)	Data 0.001 (0.078)	Loss 1.571 (1.228)
Epoch: [8][120/200]	Time 0.261 (0.343)	Data 0.000 (0.078)	Loss 1.661 (1.282)
Epoch: [8][140/200]	Time 0.257 (0.343)	Data 0.000 (0.077)	Loss 1.201 (1.310)
Epoch: [8][160/200]	Time 0.259 (0.342)	Data 0.000 (0.076)	Loss 1.579 (1.333)
Epoch: [8][180/200]	Time 0.258 (0.341)	Data 0.000 (0.075)	Loss 1.160 (1.343)
Epoch: [8][200/200]	Time 0.263 (0.348)	Data 0.001 (0.081)	Loss 1.218 (1.344)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.137)	Data 0.000 (0.040)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.298842191696167
==> Statistics for epoch 9: 598 clusters
Epoch: [9][20/200]	Time 0.455 (0.387)	Data 0.001 (0.113)	Loss 1.221 (0.357)
Epoch: [9][40/200]	Time 0.261 (0.364)	Data 0.001 (0.095)	Loss 1.413 (0.854)
Epoch: [9][60/200]	Time 0.265 (0.357)	Data 0.001 (0.089)	Loss 1.273 (1.015)
Epoch: [9][80/200]	Time 0.261 (0.352)	Data 0.001 (0.085)	Loss 1.284 (1.115)
Epoch: [9][100/200]	Time 0.260 (0.350)	Data 0.001 (0.083)	Loss 1.479 (1.179)
Epoch: [9][120/200]	Time 0.257 (0.348)	Data 0.000 (0.082)	Loss 1.460 (1.219)
Epoch: [9][140/200]	Time 0.258 (0.348)	Data 0.000 (0.081)	Loss 1.275 (1.241)
Epoch: [9][160/200]	Time 0.256 (0.347)	Data 0.000 (0.080)	Loss 1.378 (1.267)
Epoch: [9][180/200]	Time 0.257 (0.346)	Data 0.000 (0.079)	Loss 1.303 (1.290)
Epoch: [9][200/200]	Time 0.261 (0.353)	Data 0.001 (0.086)	Loss 1.390 (1.302)
Extract Features: [50/76]	Time 0.094 (0.135)	Data 0.000 (0.037)	
Mean AP: 89.4%

 * Finished epoch   9  model mAP: 89.4%  best: 89.4% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.132)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.672764778137207
==> Statistics for epoch 10: 603 clusters
Epoch: [10][20/200]	Time 0.268 (0.399)	Data 0.001 (0.119)	Loss 1.482 (0.324)
Epoch: [10][40/200]	Time 0.255 (0.366)	Data 0.001 (0.092)	Loss 1.427 (0.768)
Epoch: [10][60/200]	Time 0.258 (0.358)	Data 0.001 (0.087)	Loss 1.313 (0.964)
Epoch: [10][80/200]	Time 0.369 (0.358)	Data 0.001 (0.086)	Loss 1.236 (1.070)
Epoch: [10][100/200]	Time 0.259 (0.353)	Data 0.001 (0.083)	Loss 1.580 (1.133)
Epoch: [10][120/200]	Time 0.260 (0.349)	Data 0.000 (0.080)	Loss 1.593 (1.166)
Epoch: [10][140/200]	Time 0.257 (0.348)	Data 0.000 (0.079)	Loss 1.292 (1.203)
Epoch: [10][160/200]	Time 0.256 (0.346)	Data 0.000 (0.077)	Loss 1.702 (1.219)
Epoch: [10][180/200]	Time 0.258 (0.345)	Data 0.000 (0.076)	Loss 1.133 (1.228)
Epoch: [10][200/200]	Time 0.259 (0.351)	Data 0.001 (0.083)	Loss 1.606 (1.238)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.08344864845276
==> Statistics for epoch 11: 602 clusters
Epoch: [11][20/200]	Time 0.259 (0.381)	Data 0.001 (0.109)	Loss 1.516 (0.342)
Epoch: [11][40/200]	Time 0.279 (0.358)	Data 0.001 (0.089)	Loss 1.271 (0.774)
Epoch: [11][60/200]	Time 0.259 (0.350)	Data 0.001 (0.083)	Loss 1.236 (0.959)
Epoch: [11][80/200]	Time 0.258 (0.346)	Data 0.001 (0.079)	Loss 1.314 (1.056)
Epoch: [11][100/200]	Time 0.259 (0.344)	Data 0.001 (0.077)	Loss 1.238 (1.104)
Epoch: [11][120/200]	Time 0.260 (0.343)	Data 0.000 (0.076)	Loss 1.568 (1.137)
Epoch: [11][140/200]	Time 0.259 (0.341)	Data 0.000 (0.075)	Loss 1.418 (1.170)
Epoch: [11][160/200]	Time 0.255 (0.341)	Data 0.000 (0.074)	Loss 1.039 (1.196)
Epoch: [11][180/200]	Time 0.258 (0.342)	Data 0.000 (0.075)	Loss 1.449 (1.210)
Epoch: [11][200/200]	Time 0.259 (0.348)	Data 0.001 (0.081)	Loss 1.173 (1.220)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.138)	Data 0.000 (0.038)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.84101128578186
==> Statistics for epoch 12: 609 clusters
Epoch: [12][20/200]	Time 1.746 (0.380)	Data 1.425 (0.108)	Loss 1.110 (0.261)
Epoch: [12][40/200]	Time 0.267 (0.360)	Data 0.002 (0.090)	Loss 0.958 (0.790)
Epoch: [12][60/200]	Time 0.259 (0.354)	Data 0.001 (0.085)	Loss 1.764 (0.970)
Epoch: [12][80/200]	Time 0.263 (0.351)	Data 0.001 (0.082)	Loss 1.252 (1.047)
Epoch: [12][100/200]	Time 0.263 (0.349)	Data 0.001 (0.081)	Loss 1.204 (1.108)
Epoch: [12][120/200]	Time 0.261 (0.347)	Data 0.001 (0.080)	Loss 1.276 (1.154)
Epoch: [12][140/200]	Time 0.261 (0.347)	Data 0.001 (0.079)	Loss 1.494 (1.170)
Epoch: [12][160/200]	Time 0.262 (0.347)	Data 0.001 (0.079)	Loss 1.454 (1.198)
Epoch: [12][180/200]	Time 0.265 (0.346)	Data 0.001 (0.078)	Loss 1.030 (1.206)
Epoch: [12][200/200]	Time 0.261 (0.345)	Data 0.001 (0.078)	Loss 0.917 (1.215)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.140)	Data 0.000 (0.042)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.574183464050293
==> Statistics for epoch 13: 596 clusters
Epoch: [13][20/200]	Time 0.266 (0.372)	Data 0.001 (0.109)	Loss 1.213 (0.297)
Epoch: [13][40/200]	Time 0.275 (0.356)	Data 0.001 (0.091)	Loss 1.370 (0.745)
Epoch: [13][60/200]	Time 0.257 (0.349)	Data 0.001 (0.085)	Loss 1.177 (0.905)
Epoch: [13][80/200]	Time 0.366 (0.346)	Data 0.001 (0.080)	Loss 1.087 (0.996)
Epoch: [13][100/200]	Time 0.261 (0.343)	Data 0.001 (0.078)	Loss 1.229 (1.044)
Epoch: [13][120/200]	Time 0.259 (0.341)	Data 0.000 (0.076)	Loss 1.272 (1.077)
Epoch: [13][140/200]	Time 0.259 (0.340)	Data 0.000 (0.075)	Loss 1.652 (1.101)
Epoch: [13][160/200]	Time 0.259 (0.339)	Data 0.000 (0.075)	Loss 1.415 (1.109)
Epoch: [13][180/200]	Time 0.259 (0.339)	Data 0.000 (0.074)	Loss 1.217 (1.122)
Epoch: [13][200/200]	Time 0.260 (0.347)	Data 0.001 (0.081)	Loss 1.138 (1.135)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.136)	Data 0.000 (0.036)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.461620330810547
==> Statistics for epoch 14: 602 clusters
Epoch: [14][20/200]	Time 0.259 (0.382)	Data 0.001 (0.114)	Loss 1.238 (0.287)
Epoch: [14][40/200]	Time 0.283 (0.361)	Data 0.001 (0.093)	Loss 1.257 (0.718)
Epoch: [14][60/200]	Time 0.260 (0.353)	Data 0.001 (0.086)	Loss 1.177 (0.893)
Epoch: [14][80/200]	Time 0.262 (0.350)	Data 0.001 (0.082)	Loss 1.096 (0.974)
Epoch: [14][100/200]	Time 0.258 (0.349)	Data 0.001 (0.081)	Loss 1.353 (1.043)
Epoch: [14][120/200]	Time 0.259 (0.346)	Data 0.000 (0.080)	Loss 1.067 (1.060)
Epoch: [14][140/200]	Time 0.256 (0.345)	Data 0.000 (0.078)	Loss 0.917 (1.087)
Epoch: [14][160/200]	Time 0.258 (0.344)	Data 0.000 (0.077)	Loss 1.363 (1.104)
Epoch: [14][180/200]	Time 0.259 (0.343)	Data 0.000 (0.076)	Loss 0.880 (1.119)
Epoch: [14][200/200]	Time 0.263 (0.350)	Data 0.001 (0.083)	Loss 1.012 (1.128)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.136)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.693242073059082
==> Statistics for epoch 15: 605 clusters
Epoch: [15][20/200]	Time 0.263 (0.380)	Data 0.001 (0.110)	Loss 1.015 (0.256)
Epoch: [15][40/200]	Time 0.260 (0.354)	Data 0.001 (0.088)	Loss 1.115 (0.692)
Epoch: [15][60/200]	Time 0.259 (0.349)	Data 0.001 (0.084)	Loss 1.339 (0.863)
Epoch: [15][80/200]	Time 0.262 (0.347)	Data 0.001 (0.081)	Loss 1.047 (0.956)
Epoch: [15][100/200]	Time 0.261 (0.346)	Data 0.001 (0.080)	Loss 0.973 (0.992)
Epoch: [15][120/200]	Time 0.260 (0.345)	Data 0.000 (0.079)	Loss 1.413 (1.029)
Epoch: [15][140/200]	Time 0.258 (0.344)	Data 0.000 (0.078)	Loss 1.254 (1.050)
Epoch: [15][160/200]	Time 0.262 (0.343)	Data 0.000 (0.078)	Loss 0.993 (1.076)
Epoch: [15][180/200]	Time 0.259 (0.343)	Data 0.000 (0.077)	Loss 1.106 (1.089)
Epoch: [15][200/200]	Time 0.271 (0.350)	Data 0.001 (0.084)	Loss 1.333 (1.101)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.134)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.679162740707397
==> Statistics for epoch 16: 611 clusters
Epoch: [16][20/200]	Time 1.716 (0.390)	Data 1.411 (0.120)	Loss 0.847 (0.200)
Epoch: [16][40/200]	Time 0.261 (0.365)	Data 0.001 (0.098)	Loss 1.134 (0.660)
Epoch: [16][60/200]	Time 0.266 (0.357)	Data 0.001 (0.090)	Loss 1.417 (0.840)
Epoch: [16][80/200]	Time 0.264 (0.355)	Data 0.001 (0.087)	Loss 1.447 (0.923)
Epoch: [16][100/200]	Time 0.262 (0.352)	Data 0.001 (0.085)	Loss 1.370 (0.961)
Epoch: [16][120/200]	Time 0.260 (0.352)	Data 0.001 (0.085)	Loss 1.245 (0.990)
Epoch: [16][140/200]	Time 0.260 (0.349)	Data 0.001 (0.082)	Loss 1.161 (1.008)
Epoch: [16][160/200]	Time 0.262 (0.348)	Data 0.001 (0.081)	Loss 0.751 (1.026)
Epoch: [16][180/200]	Time 0.262 (0.346)	Data 0.001 (0.080)	Loss 1.205 (1.055)
Epoch: [16][200/200]	Time 0.270 (0.346)	Data 0.001 (0.079)	Loss 1.234 (1.063)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.101 (0.132)	Data 0.007 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.668314933776855
==> Statistics for epoch 17: 607 clusters
Epoch: [17][20/200]	Time 0.260 (0.378)	Data 0.001 (0.103)	Loss 1.182 (0.272)
Epoch: [17][40/200]	Time 0.375 (0.358)	Data 0.001 (0.088)	Loss 1.345 (0.686)
Epoch: [17][60/200]	Time 0.258 (0.350)	Data 0.001 (0.083)	Loss 0.912 (0.851)
Epoch: [17][80/200]	Time 0.271 (0.349)	Data 0.001 (0.081)	Loss 1.247 (0.924)
Epoch: [17][100/200]	Time 0.257 (0.346)	Data 0.001 (0.078)	Loss 1.555 (0.975)
Epoch: [17][120/200]	Time 0.259 (0.343)	Data 0.000 (0.076)	Loss 1.166 (1.007)
Epoch: [17][140/200]	Time 0.259 (0.345)	Data 0.000 (0.078)	Loss 0.810 (1.029)
Epoch: [17][160/200]	Time 0.260 (0.344)	Data 0.000 (0.077)	Loss 1.059 (1.043)
Epoch: [17][180/200]	Time 0.260 (0.344)	Data 0.000 (0.076)	Loss 0.859 (1.046)
Epoch: [17][200/200]	Time 0.257 (0.351)	Data 0.001 (0.083)	Loss 1.010 (1.055)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.133)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.62001132965088
==> Statistics for epoch 18: 612 clusters
Epoch: [18][20/200]	Time 1.784 (0.382)	Data 1.487 (0.112)	Loss 1.174 (0.242)
Epoch: [18][40/200]	Time 0.264 (0.361)	Data 0.001 (0.091)	Loss 0.954 (0.688)
Epoch: [18][60/200]	Time 0.258 (0.356)	Data 0.001 (0.086)	Loss 1.022 (0.820)
Epoch: [18][80/200]	Time 0.263 (0.353)	Data 0.001 (0.084)	Loss 1.244 (0.899)
Epoch: [18][100/200]	Time 0.262 (0.350)	Data 0.001 (0.082)	Loss 0.778 (0.931)
Epoch: [18][120/200]	Time 0.266 (0.348)	Data 0.001 (0.080)	Loss 1.280 (0.957)
Epoch: [18][140/200]	Time 0.261 (0.347)	Data 0.001 (0.080)	Loss 0.928 (0.976)
Epoch: [18][160/200]	Time 0.261 (0.347)	Data 0.001 (0.079)	Loss 1.079 (0.991)
Epoch: [18][180/200]	Time 0.264 (0.347)	Data 0.001 (0.079)	Loss 1.199 (1.001)
Epoch: [18][200/200]	Time 0.265 (0.346)	Data 0.001 (0.078)	Loss 1.150 (1.011)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.134)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.925366401672363
==> Statistics for epoch 19: 610 clusters
Epoch: [19][20/200]	Time 1.785 (0.399)	Data 1.498 (0.128)	Loss 1.242 (0.228)
Epoch: [19][40/200]	Time 0.260 (0.368)	Data 0.001 (0.099)	Loss 1.086 (0.651)
Epoch: [19][60/200]	Time 0.258 (0.360)	Data 0.001 (0.091)	Loss 0.616 (0.781)
Epoch: [19][80/200]	Time 0.259 (0.353)	Data 0.001 (0.085)	Loss 1.177 (0.858)
Epoch: [19][100/200]	Time 0.265 (0.350)	Data 0.001 (0.082)	Loss 1.014 (0.896)
Epoch: [19][120/200]	Time 0.263 (0.349)	Data 0.001 (0.080)	Loss 0.942 (0.929)
Epoch: [19][140/200]	Time 0.260 (0.349)	Data 0.001 (0.080)	Loss 0.868 (0.964)
Epoch: [19][160/200]	Time 0.260 (0.349)	Data 0.002 (0.080)	Loss 0.805 (0.980)
Epoch: [19][180/200]	Time 0.376 (0.347)	Data 0.001 (0.078)	Loss 1.124 (0.995)
Epoch: [19][200/200]	Time 0.260 (0.347)	Data 0.001 (0.078)	Loss 1.124 (1.004)
Extract Features: [50/76]	Time 0.096 (0.136)	Data 0.000 (0.034)	
Mean AP: 90.9%

 * Finished epoch  19  model mAP: 90.9%  best: 90.9% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.139)	Data 0.000 (0.040)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.319290161132812
==> Statistics for epoch 20: 609 clusters
Epoch: [20][20/200]	Time 1.774 (0.383)	Data 1.474 (0.114)	Loss 1.130 (0.199)
Epoch: [20][40/200]	Time 0.260 (0.358)	Data 0.001 (0.092)	Loss 1.077 (0.606)
Epoch: [20][60/200]	Time 0.271 (0.349)	Data 0.001 (0.084)	Loss 1.078 (0.758)
Epoch: [20][80/200]	Time 0.262 (0.348)	Data 0.001 (0.082)	Loss 0.892 (0.842)
Epoch: [20][100/200]	Time 0.259 (0.345)	Data 0.001 (0.079)	Loss 1.301 (0.869)
Epoch: [20][120/200]	Time 0.261 (0.345)	Data 0.001 (0.078)	Loss 1.148 (0.902)
Epoch: [20][140/200]	Time 0.259 (0.345)	Data 0.001 (0.078)	Loss 1.045 (0.917)
Epoch: [20][160/200]	Time 0.260 (0.344)	Data 0.001 (0.078)	Loss 1.051 (0.935)
Epoch: [20][180/200]	Time 0.260 (0.344)	Data 0.001 (0.077)	Loss 0.897 (0.943)
Epoch: [20][200/200]	Time 0.260 (0.343)	Data 0.001 (0.077)	Loss 1.359 (0.946)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.136)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.699303150177002
==> Statistics for epoch 21: 610 clusters
Epoch: [21][20/200]	Time 1.632 (0.383)	Data 1.329 (0.112)	Loss 0.992 (0.182)
Epoch: [21][40/200]	Time 0.258 (0.358)	Data 0.001 (0.089)	Loss 1.135 (0.585)
Epoch: [21][60/200]	Time 0.262 (0.350)	Data 0.001 (0.082)	Loss 0.924 (0.731)
Epoch: [21][80/200]	Time 0.263 (0.347)	Data 0.001 (0.080)	Loss 1.072 (0.824)
Epoch: [21][100/200]	Time 0.258 (0.344)	Data 0.001 (0.077)	Loss 0.752 (0.866)
Epoch: [21][120/200]	Time 0.266 (0.342)	Data 0.001 (0.075)	Loss 1.175 (0.898)
Epoch: [21][140/200]	Time 0.259 (0.341)	Data 0.001 (0.075)	Loss 1.352 (0.916)
Epoch: [21][160/200]	Time 0.261 (0.340)	Data 0.001 (0.074)	Loss 1.389 (0.942)
Epoch: [21][180/200]	Time 0.265 (0.341)	Data 0.001 (0.074)	Loss 1.178 (0.954)
Epoch: [21][200/200]	Time 0.260 (0.341)	Data 0.001 (0.074)	Loss 1.071 (0.960)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.135)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.55011487007141
==> Statistics for epoch 22: 611 clusters
Epoch: [22][20/200]	Time 1.635 (0.374)	Data 1.328 (0.108)	Loss 0.926 (0.173)
Epoch: [22][40/200]	Time 0.258 (0.353)	Data 0.001 (0.088)	Loss 1.185 (0.568)
Epoch: [22][60/200]	Time 0.271 (0.346)	Data 0.001 (0.080)	Loss 0.830 (0.717)
Epoch: [22][80/200]	Time 0.261 (0.346)	Data 0.001 (0.080)	Loss 1.387 (0.795)
Epoch: [22][100/200]	Time 0.258 (0.343)	Data 0.001 (0.077)	Loss 1.223 (0.843)
Epoch: [22][120/200]	Time 0.260 (0.342)	Data 0.001 (0.077)	Loss 0.950 (0.863)
Epoch: [22][140/200]	Time 0.364 (0.341)	Data 0.001 (0.076)	Loss 1.385 (0.882)
Epoch: [22][160/200]	Time 0.262 (0.340)	Data 0.001 (0.075)	Loss 0.991 (0.899)
Epoch: [22][180/200]	Time 0.258 (0.340)	Data 0.001 (0.074)	Loss 0.960 (0.908)
Epoch: [22][200/200]	Time 0.260 (0.339)	Data 0.001 (0.074)	Loss 0.913 (0.918)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.892913103103638
==> Statistics for epoch 23: 610 clusters
Epoch: [23][20/200]	Time 1.787 (0.381)	Data 1.495 (0.110)	Loss 0.839 (0.177)
Epoch: [23][40/200]	Time 0.258 (0.363)	Data 0.001 (0.094)	Loss 1.075 (0.587)
Epoch: [23][60/200]	Time 0.263 (0.355)	Data 0.001 (0.086)	Loss 1.443 (0.736)
Epoch: [23][80/200]	Time 0.262 (0.351)	Data 0.001 (0.083)	Loss 0.919 (0.780)
Epoch: [23][100/200]	Time 0.411 (0.349)	Data 0.001 (0.080)	Loss 1.274 (0.827)
Epoch: [23][120/200]	Time 0.263 (0.347)	Data 0.001 (0.079)	Loss 0.863 (0.853)
Epoch: [23][140/200]	Time 0.263 (0.347)	Data 0.002 (0.078)	Loss 0.982 (0.887)
Epoch: [23][160/200]	Time 0.260 (0.345)	Data 0.001 (0.077)	Loss 1.189 (0.908)
Epoch: [23][180/200]	Time 0.259 (0.345)	Data 0.001 (0.077)	Loss 0.975 (0.924)
Epoch: [23][200/200]	Time 0.260 (0.346)	Data 0.001 (0.078)	Loss 0.831 (0.935)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.141)	Data 0.000 (0.041)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.6550452709198
==> Statistics for epoch 24: 607 clusters
Epoch: [24][20/200]	Time 0.280 (0.389)	Data 0.001 (0.114)	Loss 0.987 (0.244)
Epoch: [24][40/200]	Time 0.259 (0.365)	Data 0.001 (0.093)	Loss 0.761 (0.591)
Epoch: [24][60/200]	Time 0.259 (0.357)	Data 0.001 (0.087)	Loss 0.835 (0.693)
Epoch: [24][80/200]	Time 0.264 (0.355)	Data 0.001 (0.086)	Loss 1.108 (0.756)
Epoch: [24][100/200]	Time 0.266 (0.351)	Data 0.001 (0.083)	Loss 0.854 (0.804)
Epoch: [24][120/200]	Time 0.262 (0.350)	Data 0.000 (0.081)	Loss 0.886 (0.829)
Epoch: [24][140/200]	Time 0.257 (0.349)	Data 0.000 (0.080)	Loss 0.879 (0.862)
Epoch: [24][160/200]	Time 0.258 (0.348)	Data 0.000 (0.079)	Loss 0.997 (0.882)
Epoch: [24][180/200]	Time 0.258 (0.347)	Data 0.000 (0.079)	Loss 1.279 (0.900)
Epoch: [24][200/200]	Time 0.262 (0.354)	Data 0.001 (0.085)	Loss 0.837 (0.909)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.132)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.04465961456299
==> Statistics for epoch 25: 609 clusters
Epoch: [25][20/200]	Time 1.764 (0.385)	Data 1.475 (0.116)	Loss 1.045 (0.186)
Epoch: [25][40/200]	Time 0.264 (0.360)	Data 0.001 (0.093)	Loss 1.093 (0.580)
Epoch: [25][60/200]	Time 0.258 (0.351)	Data 0.001 (0.084)	Loss 0.765 (0.719)
Epoch: [25][80/200]	Time 0.259 (0.348)	Data 0.001 (0.081)	Loss 1.290 (0.806)
Epoch: [25][100/200]	Time 0.263 (0.346)	Data 0.001 (0.078)	Loss 0.877 (0.843)
Epoch: [25][120/200]	Time 0.260 (0.346)	Data 0.001 (0.078)	Loss 0.834 (0.868)
Epoch: [25][140/200]	Time 0.261 (0.343)	Data 0.001 (0.076)	Loss 1.028 (0.894)
Epoch: [25][160/200]	Time 0.258 (0.343)	Data 0.001 (0.075)	Loss 0.809 (0.904)
Epoch: [25][180/200]	Time 0.259 (0.342)	Data 0.001 (0.075)	Loss 0.928 (0.915)
Epoch: [25][200/200]	Time 0.263 (0.342)	Data 0.001 (0.074)	Loss 0.939 (0.924)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.129 (0.137)	Data 0.034 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.624321222305298
==> Statistics for epoch 26: 610 clusters
Epoch: [26][20/200]	Time 1.652 (0.379)	Data 1.340 (0.114)	Loss 0.769 (0.164)
Epoch: [26][40/200]	Time 0.265 (0.363)	Data 0.001 (0.096)	Loss 1.199 (0.566)
Epoch: [26][60/200]	Time 0.263 (0.355)	Data 0.001 (0.088)	Loss 0.926 (0.685)
Epoch: [26][80/200]	Time 0.269 (0.352)	Data 0.001 (0.085)	Loss 1.034 (0.757)
Epoch: [26][100/200]	Time 0.260 (0.349)	Data 0.001 (0.082)	Loss 0.740 (0.802)
Epoch: [26][120/200]	Time 0.262 (0.347)	Data 0.001 (0.080)	Loss 0.904 (0.828)
Epoch: [26][140/200]	Time 0.258 (0.345)	Data 0.001 (0.078)	Loss 0.927 (0.847)
Epoch: [26][160/200]	Time 0.263 (0.344)	Data 0.001 (0.076)	Loss 0.933 (0.866)
Epoch: [26][180/200]	Time 0.270 (0.343)	Data 0.001 (0.076)	Loss 1.174 (0.887)
Epoch: [26][200/200]	Time 0.265 (0.343)	Data 0.001 (0.076)	Loss 1.160 (0.896)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.133)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.817391872406006
==> Statistics for epoch 27: 613 clusters
Epoch: [27][20/200]	Time 1.704 (0.383)	Data 1.427 (0.116)	Loss 1.014 (0.184)
Epoch: [27][40/200]	Time 0.257 (0.358)	Data 0.001 (0.091)	Loss 0.926 (0.563)
Epoch: [27][60/200]	Time 0.260 (0.351)	Data 0.000 (0.085)	Loss 1.060 (0.700)
Epoch: [27][80/200]	Time 0.259 (0.345)	Data 0.001 (0.079)	Loss 1.289 (0.756)
Epoch: [27][100/200]	Time 0.261 (0.342)	Data 0.001 (0.077)	Loss 0.885 (0.802)
Epoch: [27][120/200]	Time 0.259 (0.342)	Data 0.001 (0.077)	Loss 0.799 (0.835)
Epoch: [27][140/200]	Time 0.365 (0.343)	Data 0.001 (0.077)	Loss 1.031 (0.857)
Epoch: [27][160/200]	Time 0.259 (0.341)	Data 0.001 (0.076)	Loss 0.955 (0.876)
Epoch: [27][180/200]	Time 0.261 (0.340)	Data 0.001 (0.075)	Loss 1.190 (0.889)
Epoch: [27][200/200]	Time 0.260 (0.339)	Data 0.000 (0.074)	Loss 0.696 (0.900)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.137)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.4893958568573
==> Statistics for epoch 28: 609 clusters
Epoch: [28][20/200]	Time 1.824 (0.385)	Data 1.435 (0.116)	Loss 0.869 (0.172)
Epoch: [28][40/200]	Time 0.271 (0.360)	Data 0.001 (0.094)	Loss 1.055 (0.578)
Epoch: [28][60/200]	Time 0.262 (0.354)	Data 0.001 (0.087)	Loss 0.688 (0.718)
Epoch: [28][80/200]	Time 0.261 (0.349)	Data 0.001 (0.082)	Loss 1.117 (0.796)
Epoch: [28][100/200]	Time 0.261 (0.346)	Data 0.001 (0.079)	Loss 1.296 (0.828)
Epoch: [28][120/200]	Time 0.261 (0.344)	Data 0.001 (0.077)	Loss 0.925 (0.858)
Epoch: [28][140/200]	Time 0.261 (0.344)	Data 0.001 (0.076)	Loss 1.020 (0.883)
Epoch: [28][160/200]	Time 0.280 (0.343)	Data 0.001 (0.075)	Loss 0.896 (0.895)
Epoch: [28][180/200]	Time 0.265 (0.342)	Data 0.001 (0.074)	Loss 1.111 (0.909)
Epoch: [28][200/200]	Time 0.259 (0.341)	Data 0.000 (0.073)	Loss 1.051 (0.917)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.133)	Data 0.000 (0.036)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.82885766029358
==> Statistics for epoch 29: 609 clusters
Epoch: [29][20/200]	Time 1.783 (0.374)	Data 1.494 (0.110)	Loss 0.858 (0.169)
Epoch: [29][40/200]	Time 0.297 (0.359)	Data 0.001 (0.093)	Loss 1.294 (0.567)
Epoch: [29][60/200]	Time 0.269 (0.350)	Data 0.001 (0.086)	Loss 0.849 (0.705)
Epoch: [29][80/200]	Time 0.261 (0.349)	Data 0.001 (0.084)	Loss 1.265 (0.781)
Epoch: [29][100/200]	Time 0.257 (0.347)	Data 0.001 (0.082)	Loss 1.024 (0.805)
Epoch: [29][120/200]	Time 0.272 (0.346)	Data 0.001 (0.081)	Loss 0.827 (0.832)
Epoch: [29][140/200]	Time 0.262 (0.344)	Data 0.001 (0.079)	Loss 0.783 (0.861)
Epoch: [29][160/200]	Time 0.260 (0.344)	Data 0.001 (0.078)	Loss 1.106 (0.875)
Epoch: [29][180/200]	Time 0.258 (0.343)	Data 0.001 (0.077)	Loss 0.972 (0.885)
Epoch: [29][200/200]	Time 0.261 (0.343)	Data 0.001 (0.077)	Loss 0.828 (0.896)
Extract Features: [50/76]	Time 0.096 (0.135)	Data 0.000 (0.037)	
Mean AP: 91.2%

 * Finished epoch  29  model mAP: 91.2%  best: 91.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.135)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.481682777404785
==> Statistics for epoch 30: 610 clusters
Epoch: [30][20/200]	Time 1.597 (0.375)	Data 1.286 (0.107)	Loss 1.002 (0.180)
Epoch: [30][40/200]	Time 0.258 (0.354)	Data 0.001 (0.087)	Loss 0.900 (0.539)
Epoch: [30][60/200]	Time 0.258 (0.347)	Data 0.000 (0.082)	Loss 0.879 (0.686)
Epoch: [30][80/200]	Time 0.261 (0.344)	Data 0.001 (0.078)	Loss 0.797 (0.761)
Epoch: [30][100/200]	Time 0.261 (0.341)	Data 0.001 (0.076)	Loss 0.846 (0.807)
Epoch: [30][120/200]	Time 0.258 (0.340)	Data 0.001 (0.074)	Loss 0.800 (0.843)
Epoch: [30][140/200]	Time 0.257 (0.339)	Data 0.001 (0.073)	Loss 0.917 (0.857)
Epoch: [30][160/200]	Time 0.261 (0.339)	Data 0.001 (0.073)	Loss 1.345 (0.873)
Epoch: [30][180/200]	Time 0.257 (0.338)	Data 0.001 (0.072)	Loss 1.144 (0.886)
Epoch: [30][200/200]	Time 0.257 (0.339)	Data 0.001 (0.073)	Loss 0.912 (0.901)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.145 (0.132)	Data 0.050 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.769330978393555
==> Statistics for epoch 31: 613 clusters
Epoch: [31][20/200]	Time 1.725 (0.382)	Data 1.447 (0.115)	Loss 1.245 (0.191)
Epoch: [31][40/200]	Time 0.365 (0.357)	Data 0.001 (0.092)	Loss 0.777 (0.545)
Epoch: [31][60/200]	Time 0.291 (0.348)	Data 0.001 (0.083)	Loss 0.864 (0.695)
Epoch: [31][80/200]	Time 0.260 (0.345)	Data 0.001 (0.079)	Loss 1.011 (0.759)
Epoch: [31][100/200]	Time 0.262 (0.343)	Data 0.001 (0.077)	Loss 1.218 (0.806)
Epoch: [31][120/200]	Time 0.259 (0.342)	Data 0.001 (0.075)	Loss 1.287 (0.843)
Epoch: [31][140/200]	Time 0.262 (0.341)	Data 0.001 (0.074)	Loss 0.791 (0.860)
Epoch: [31][160/200]	Time 0.258 (0.341)	Data 0.001 (0.074)	Loss 0.790 (0.874)
Epoch: [31][180/200]	Time 0.256 (0.339)	Data 0.001 (0.073)	Loss 1.009 (0.886)
Epoch: [31][200/200]	Time 0.267 (0.338)	Data 0.001 (0.072)	Loss 1.117 (0.897)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.127 (0.134)	Data 0.032 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.986316442489624
==> Statistics for epoch 32: 609 clusters
Epoch: [32][20/200]	Time 1.656 (0.375)	Data 1.363 (0.111)	Loss 1.066 (0.198)
Epoch: [32][40/200]	Time 0.284 (0.356)	Data 0.001 (0.091)	Loss 0.840 (0.585)
Epoch: [32][60/200]	Time 0.265 (0.348)	Data 0.001 (0.082)	Loss 0.935 (0.709)
Epoch: [32][80/200]	Time 0.260 (0.345)	Data 0.001 (0.079)	Loss 0.792 (0.776)
Epoch: [32][100/200]	Time 0.262 (0.344)	Data 0.001 (0.078)	Loss 0.859 (0.816)
Epoch: [32][120/200]	Time 0.269 (0.342)	Data 0.001 (0.076)	Loss 1.055 (0.848)
Epoch: [32][140/200]	Time 0.258 (0.340)	Data 0.001 (0.075)	Loss 0.972 (0.865)
Epoch: [32][160/200]	Time 0.258 (0.339)	Data 0.001 (0.074)	Loss 0.908 (0.883)
Epoch: [32][180/200]	Time 0.267 (0.338)	Data 0.001 (0.073)	Loss 0.963 (0.895)
Epoch: [32][200/200]	Time 0.260 (0.337)	Data 0.001 (0.072)	Loss 0.819 (0.905)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.133)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.73368740081787
==> Statistics for epoch 33: 609 clusters
Epoch: [33][20/200]	Time 1.610 (0.375)	Data 1.280 (0.101)	Loss 1.002 (0.174)
Epoch: [33][40/200]	Time 0.265 (0.353)	Data 0.001 (0.085)	Loss 1.127 (0.562)
Epoch: [33][60/200]	Time 0.273 (0.350)	Data 0.001 (0.082)	Loss 1.327 (0.698)
Epoch: [33][80/200]	Time 0.264 (0.348)	Data 0.001 (0.079)	Loss 1.088 (0.774)
Epoch: [33][100/200]	Time 0.265 (0.345)	Data 0.001 (0.078)	Loss 1.033 (0.822)
Epoch: [33][120/200]	Time 0.261 (0.343)	Data 0.001 (0.076)	Loss 1.120 (0.849)
Epoch: [33][140/200]	Time 0.259 (0.343)	Data 0.001 (0.075)	Loss 0.995 (0.868)
Epoch: [33][160/200]	Time 0.372 (0.342)	Data 0.001 (0.075)	Loss 1.456 (0.885)
Epoch: [33][180/200]	Time 0.263 (0.342)	Data 0.001 (0.075)	Loss 1.092 (0.904)
Epoch: [33][200/200]	Time 0.261 (0.342)	Data 0.001 (0.075)	Loss 0.818 (0.918)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.135)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.22010850906372
==> Statistics for epoch 34: 612 clusters
Epoch: [34][20/200]	Time 1.593 (0.367)	Data 1.286 (0.101)	Loss 0.901 (0.172)
Epoch: [34][40/200]	Time 0.258 (0.350)	Data 0.001 (0.084)	Loss 1.031 (0.562)
Epoch: [34][60/200]	Time 0.262 (0.344)	Data 0.001 (0.078)	Loss 1.076 (0.699)
Epoch: [34][80/200]	Time 0.266 (0.344)	Data 0.001 (0.078)	Loss 0.771 (0.770)
Epoch: [34][100/200]	Time 0.258 (0.342)	Data 0.001 (0.076)	Loss 1.434 (0.827)
Epoch: [34][120/200]	Time 0.256 (0.340)	Data 0.000 (0.074)	Loss 0.976 (0.853)
Epoch: [34][140/200]	Time 0.259 (0.339)	Data 0.001 (0.074)	Loss 0.841 (0.867)
Epoch: [34][160/200]	Time 0.261 (0.338)	Data 0.000 (0.073)	Loss 0.863 (0.875)
Epoch: [34][180/200]	Time 0.357 (0.338)	Data 0.001 (0.073)	Loss 0.956 (0.884)
Epoch: [34][200/200]	Time 0.262 (0.337)	Data 0.001 (0.073)	Loss 0.924 (0.891)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.133)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.84932041168213
==> Statistics for epoch 35: 607 clusters
Epoch: [35][20/200]	Time 0.260 (0.369)	Data 0.001 (0.099)	Loss 1.180 (0.215)
Epoch: [35][40/200]	Time 0.257 (0.352)	Data 0.001 (0.083)	Loss 1.154 (0.558)
Epoch: [35][60/200]	Time 0.259 (0.345)	Data 0.001 (0.080)	Loss 0.829 (0.678)
Epoch: [35][80/200]	Time 0.259 (0.343)	Data 0.001 (0.078)	Loss 0.959 (0.752)
Epoch: [35][100/200]	Time 0.259 (0.340)	Data 0.001 (0.076)	Loss 0.939 (0.796)
Epoch: [35][120/200]	Time 0.258 (0.340)	Data 0.000 (0.075)	Loss 1.012 (0.824)
Epoch: [35][140/200]	Time 0.256 (0.340)	Data 0.000 (0.076)	Loss 1.088 (0.848)
Epoch: [35][160/200]	Time 0.257 (0.340)	Data 0.000 (0.075)	Loss 0.718 (0.869)
Epoch: [35][180/200]	Time 0.260 (0.340)	Data 0.000 (0.075)	Loss 0.938 (0.881)
Epoch: [35][200/200]	Time 0.271 (0.347)	Data 0.001 (0.081)	Loss 0.831 (0.888)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.144 (0.133)	Data 0.048 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.613646984100342
==> Statistics for epoch 36: 613 clusters
Epoch: [36][20/200]	Time 1.740 (0.370)	Data 1.447 (0.106)	Loss 0.788 (0.173)
Epoch: [36][40/200]	Time 0.262 (0.351)	Data 0.001 (0.085)	Loss 0.606 (0.571)
Epoch: [36][60/200]	Time 0.264 (0.347)	Data 0.001 (0.081)	Loss 1.079 (0.717)
Epoch: [36][80/200]	Time 0.263 (0.346)	Data 0.001 (0.079)	Loss 1.146 (0.793)
Epoch: [36][100/200]	Time 0.259 (0.342)	Data 0.001 (0.076)	Loss 1.214 (0.817)
Epoch: [36][120/200]	Time 0.266 (0.341)	Data 0.001 (0.075)	Loss 0.890 (0.837)
Epoch: [36][140/200]	Time 0.259 (0.341)	Data 0.001 (0.075)	Loss 0.713 (0.851)
Epoch: [36][160/200]	Time 0.261 (0.341)	Data 0.001 (0.074)	Loss 0.851 (0.868)
Epoch: [36][180/200]	Time 0.257 (0.341)	Data 0.001 (0.074)	Loss 0.860 (0.883)
Epoch: [36][200/200]	Time 0.258 (0.340)	Data 0.001 (0.073)	Loss 0.801 (0.894)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.140)	Data 0.000 (0.040)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.929158687591553
==> Statistics for epoch 37: 612 clusters
Epoch: [37][20/200]	Time 1.710 (0.383)	Data 1.394 (0.113)	Loss 0.930 (0.172)
Epoch: [37][40/200]	Time 0.261 (0.361)	Data 0.001 (0.092)	Loss 1.088 (0.557)
Epoch: [37][60/200]	Time 0.259 (0.353)	Data 0.001 (0.087)	Loss 0.644 (0.686)
Epoch: [37][80/200]	Time 0.259 (0.350)	Data 0.001 (0.084)	Loss 1.042 (0.761)
Epoch: [37][100/200]	Time 0.259 (0.347)	Data 0.001 (0.080)	Loss 1.036 (0.798)
Epoch: [37][120/200]	Time 0.262 (0.345)	Data 0.004 (0.079)	Loss 0.686 (0.822)
Epoch: [37][140/200]	Time 0.259 (0.343)	Data 0.001 (0.077)	Loss 1.307 (0.849)
Epoch: [37][160/200]	Time 0.257 (0.342)	Data 0.001 (0.075)	Loss 0.838 (0.860)
Epoch: [37][180/200]	Time 0.259 (0.342)	Data 0.001 (0.075)	Loss 0.874 (0.874)
Epoch: [37][200/200]	Time 0.262 (0.340)	Data 0.001 (0.074)	Loss 0.847 (0.885)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.133)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.846519947052002
==> Statistics for epoch 38: 612 clusters
Epoch: [38][20/200]	Time 1.746 (0.383)	Data 1.465 (0.114)	Loss 0.845 (0.156)
Epoch: [38][40/200]	Time 0.262 (0.361)	Data 0.001 (0.093)	Loss 0.931 (0.521)
Epoch: [38][60/200]	Time 0.258 (0.355)	Data 0.001 (0.087)	Loss 1.003 (0.663)
Epoch: [38][80/200]	Time 0.262 (0.351)	Data 0.001 (0.083)	Loss 0.768 (0.730)
Epoch: [38][100/200]	Time 0.258 (0.348)	Data 0.001 (0.081)	Loss 1.131 (0.783)
Epoch: [38][120/200]	Time 0.366 (0.348)	Data 0.001 (0.080)	Loss 0.930 (0.820)
Epoch: [38][140/200]	Time 0.261 (0.346)	Data 0.001 (0.078)	Loss 1.463 (0.849)
Epoch: [38][160/200]	Time 0.269 (0.344)	Data 0.001 (0.077)	Loss 1.142 (0.855)
Epoch: [38][180/200]	Time 0.260 (0.344)	Data 0.001 (0.077)	Loss 0.952 (0.861)
Epoch: [38][200/200]	Time 0.259 (0.344)	Data 0.001 (0.076)	Loss 0.942 (0.877)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.901920318603516
==> Statistics for epoch 39: 609 clusters
Epoch: [39][20/200]	Time 1.763 (0.379)	Data 1.478 (0.108)	Loss 0.908 (0.164)
Epoch: [39][40/200]	Time 0.269 (0.361)	Data 0.001 (0.092)	Loss 0.978 (0.540)
Epoch: [39][60/200]	Time 0.267 (0.354)	Data 0.001 (0.085)	Loss 0.919 (0.675)
Epoch: [39][80/200]	Time 0.265 (0.350)	Data 0.001 (0.081)	Loss 0.713 (0.727)
Epoch: [39][100/200]	Time 0.261 (0.348)	Data 0.001 (0.080)	Loss 1.116 (0.770)
Epoch: [39][120/200]	Time 0.259 (0.345)	Data 0.001 (0.077)	Loss 0.884 (0.803)
Epoch: [39][140/200]	Time 0.261 (0.344)	Data 0.001 (0.076)	Loss 1.006 (0.832)
Epoch: [39][160/200]	Time 0.265 (0.343)	Data 0.001 (0.075)	Loss 1.004 (0.848)
Epoch: [39][180/200]	Time 0.259 (0.343)	Data 0.001 (0.075)	Loss 1.211 (0.864)
Epoch: [39][200/200]	Time 0.259 (0.342)	Data 0.001 (0.075)	Loss 0.865 (0.873)
Extract Features: [50/76]	Time 0.094 (0.130)	Data 0.000 (0.032)	
Mean AP: 91.4%

 * Finished epoch  39  model mAP: 91.4%  best: 91.4% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.133 (0.135)	Data 0.039 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.844610452651978
==> Statistics for epoch 40: 612 clusters
Epoch: [40][20/200]	Time 1.720 (0.371)	Data 1.419 (0.109)	Loss 0.987 (0.180)
Epoch: [40][40/200]	Time 0.277 (0.354)	Data 0.001 (0.089)	Loss 0.927 (0.554)
Epoch: [40][60/200]	Time 0.262 (0.348)	Data 0.001 (0.083)	Loss 1.265 (0.689)
Epoch: [40][80/200]	Time 0.260 (0.345)	Data 0.001 (0.079)	Loss 1.117 (0.763)
Epoch: [40][100/200]	Time 0.264 (0.342)	Data 0.001 (0.076)	Loss 0.860 (0.811)
Epoch: [40][120/200]	Time 0.264 (0.343)	Data 0.001 (0.077)	Loss 1.002 (0.844)
Epoch: [40][140/200]	Time 0.262 (0.342)	Data 0.001 (0.076)	Loss 1.058 (0.866)
Epoch: [40][160/200]	Time 0.258 (0.341)	Data 0.001 (0.075)	Loss 1.207 (0.875)
Epoch: [40][180/200]	Time 0.261 (0.341)	Data 0.001 (0.075)	Loss 1.054 (0.887)
Epoch: [40][200/200]	Time 0.258 (0.341)	Data 0.001 (0.074)	Loss 0.574 (0.894)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.141)	Data 0.000 (0.041)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.97612977027893
==> Statistics for epoch 41: 613 clusters
Epoch: [41][20/200]	Time 1.714 (0.391)	Data 1.414 (0.121)	Loss 0.945 (0.167)
Epoch: [41][40/200]	Time 0.258 (0.360)	Data 0.001 (0.095)	Loss 0.823 (0.516)
Epoch: [41][60/200]	Time 0.262 (0.352)	Data 0.001 (0.085)	Loss 0.865 (0.647)
Epoch: [41][80/200]	Time 0.258 (0.346)	Data 0.001 (0.080)	Loss 0.845 (0.741)
Epoch: [41][100/200]	Time 0.258 (0.347)	Data 0.001 (0.080)	Loss 1.110 (0.776)
Epoch: [41][120/200]	Time 0.258 (0.345)	Data 0.001 (0.078)	Loss 0.718 (0.804)
Epoch: [41][140/200]	Time 0.269 (0.343)	Data 0.001 (0.077)	Loss 0.781 (0.831)
Epoch: [41][160/200]	Time 0.263 (0.342)	Data 0.001 (0.077)	Loss 0.901 (0.847)
Epoch: [41][180/200]	Time 0.262 (0.341)	Data 0.001 (0.076)	Loss 1.038 (0.862)
Epoch: [41][200/200]	Time 0.260 (0.341)	Data 0.001 (0.076)	Loss 0.797 (0.876)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.132)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.70159935951233
==> Statistics for epoch 42: 611 clusters
Epoch: [42][20/200]	Time 1.746 (0.381)	Data 1.429 (0.110)	Loss 0.870 (0.169)
Epoch: [42][40/200]	Time 0.265 (0.362)	Data 0.001 (0.091)	Loss 1.162 (0.541)
Epoch: [42][60/200]	Time 0.262 (0.352)	Data 0.001 (0.083)	Loss 0.844 (0.661)
Epoch: [42][80/200]	Time 0.263 (0.348)	Data 0.001 (0.080)	Loss 1.010 (0.744)
Epoch: [42][100/200]	Time 0.256 (0.346)	Data 0.001 (0.077)	Loss 1.093 (0.793)
Epoch: [42][120/200]	Time 0.258 (0.343)	Data 0.001 (0.075)	Loss 0.742 (0.822)
Epoch: [42][140/200]	Time 0.262 (0.342)	Data 0.001 (0.074)	Loss 1.178 (0.856)
Epoch: [42][160/200]	Time 0.261 (0.341)	Data 0.001 (0.073)	Loss 1.298 (0.869)
Epoch: [42][180/200]	Time 0.259 (0.341)	Data 0.001 (0.073)	Loss 1.168 (0.873)
Epoch: [42][200/200]	Time 0.260 (0.340)	Data 0.001 (0.072)	Loss 0.706 (0.888)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.131)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.538368940353394
==> Statistics for epoch 43: 611 clusters
Epoch: [43][20/200]	Time 1.562 (0.368)	Data 1.249 (0.097)	Loss 0.985 (0.175)
Epoch: [43][40/200]	Time 0.256 (0.348)	Data 0.001 (0.083)	Loss 0.968 (0.545)
Epoch: [43][60/200]	Time 0.259 (0.345)	Data 0.001 (0.079)	Loss 0.693 (0.684)
Epoch: [43][80/200]	Time 0.260 (0.343)	Data 0.001 (0.077)	Loss 1.234 (0.756)
Epoch: [43][100/200]	Time 0.259 (0.343)	Data 0.001 (0.077)	Loss 0.921 (0.801)
Epoch: [43][120/200]	Time 0.259 (0.342)	Data 0.001 (0.076)	Loss 0.965 (0.831)
Epoch: [43][140/200]	Time 0.263 (0.341)	Data 0.001 (0.075)	Loss 0.939 (0.846)
Epoch: [43][160/200]	Time 0.266 (0.341)	Data 0.001 (0.075)	Loss 1.291 (0.871)
Epoch: [43][180/200]	Time 0.265 (0.340)	Data 0.001 (0.074)	Loss 1.403 (0.890)
Epoch: [43][200/200]	Time 0.263 (0.339)	Data 0.001 (0.074)	Loss 0.815 (0.891)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.129)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.82078194618225
==> Statistics for epoch 44: 611 clusters
Epoch: [44][20/200]	Time 1.616 (0.365)	Data 1.301 (0.099)	Loss 0.935 (0.174)
Epoch: [44][40/200]	Time 0.268 (0.353)	Data 0.001 (0.086)	Loss 1.014 (0.559)
Epoch: [44][60/200]	Time 0.258 (0.348)	Data 0.001 (0.081)	Loss 0.864 (0.675)
Epoch: [44][80/200]	Time 0.258 (0.344)	Data 0.001 (0.078)	Loss 0.735 (0.752)
Epoch: [44][100/200]	Time 0.259 (0.342)	Data 0.001 (0.075)	Loss 1.380 (0.802)
Epoch: [44][120/200]	Time 0.265 (0.343)	Data 0.000 (0.076)	Loss 0.960 (0.830)
Epoch: [44][140/200]	Time 0.260 (0.342)	Data 0.000 (0.075)	Loss 0.958 (0.840)
Epoch: [44][160/200]	Time 0.263 (0.341)	Data 0.000 (0.074)	Loss 0.946 (0.852)
Epoch: [44][180/200]	Time 0.258 (0.341)	Data 0.001 (0.073)	Loss 0.794 (0.868)
Epoch: [44][200/200]	Time 0.263 (0.340)	Data 0.001 (0.073)	Loss 0.846 (0.876)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.093 (0.139)	Data 0.000 (0.042)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.239502429962158
==> Statistics for epoch 45: 615 clusters
Epoch: [45][20/200]	Time 1.745 (0.376)	Data 1.465 (0.108)	Loss 1.040 (0.167)
Epoch: [45][40/200]	Time 0.360 (0.353)	Data 0.001 (0.087)	Loss 0.811 (0.533)
Epoch: [45][60/200]	Time 0.269 (0.345)	Data 0.001 (0.080)	Loss 0.692 (0.677)
Epoch: [45][80/200]	Time 0.259 (0.345)	Data 0.001 (0.080)	Loss 1.115 (0.753)
Epoch: [45][100/200]	Time 0.260 (0.343)	Data 0.001 (0.079)	Loss 1.094 (0.798)
Epoch: [45][120/200]	Time 0.261 (0.343)	Data 0.001 (0.079)	Loss 0.723 (0.818)
Epoch: [45][140/200]	Time 0.261 (0.342)	Data 0.001 (0.077)	Loss 0.849 (0.839)
Epoch: [45][160/200]	Time 0.260 (0.342)	Data 0.001 (0.076)	Loss 1.304 (0.856)
Epoch: [45][180/200]	Time 0.262 (0.342)	Data 0.001 (0.076)	Loss 0.610 (0.869)
Epoch: [45][200/200]	Time 0.262 (0.341)	Data 0.001 (0.076)	Loss 1.050 (0.879)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.131)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.1605486869812
==> Statistics for epoch 46: 611 clusters
Epoch: [46][20/200]	Time 1.776 (0.393)	Data 1.479 (0.123)	Loss 1.014 (0.184)
Epoch: [46][40/200]	Time 0.259 (0.366)	Data 0.001 (0.098)	Loss 1.172 (0.557)
Epoch: [46][60/200]	Time 0.269 (0.357)	Data 0.001 (0.089)	Loss 0.729 (0.678)
Epoch: [46][80/200]	Time 0.258 (0.353)	Data 0.001 (0.085)	Loss 0.743 (0.725)
Epoch: [46][100/200]	Time 0.255 (0.348)	Data 0.001 (0.081)	Loss 1.013 (0.767)
Epoch: [46][120/200]	Time 0.258 (0.346)	Data 0.001 (0.080)	Loss 0.751 (0.795)
Epoch: [46][140/200]	Time 0.260 (0.344)	Data 0.004 (0.078)	Loss 0.710 (0.817)
Epoch: [46][160/200]	Time 0.260 (0.343)	Data 0.001 (0.077)	Loss 1.004 (0.834)
Epoch: [46][180/200]	Time 0.391 (0.343)	Data 0.001 (0.077)	Loss 0.989 (0.857)
Epoch: [46][200/200]	Time 0.263 (0.342)	Data 0.001 (0.076)	Loss 0.806 (0.870)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.185 (0.137)	Data 0.091 (0.036)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.548702478408813
==> Statistics for epoch 47: 615 clusters
Epoch: [47][20/200]	Time 1.759 (0.384)	Data 1.434 (0.111)	Loss 0.818 (0.160)
Epoch: [47][40/200]	Time 0.277 (0.358)	Data 0.001 (0.091)	Loss 1.014 (0.518)
Epoch: [47][60/200]	Time 0.279 (0.351)	Data 0.001 (0.083)	Loss 0.955 (0.668)
Epoch: [47][80/200]	Time 0.263 (0.348)	Data 0.001 (0.080)	Loss 1.207 (0.742)
Epoch: [47][100/200]	Time 0.261 (0.344)	Data 0.001 (0.077)	Loss 0.983 (0.786)
Epoch: [47][120/200]	Time 0.256 (0.343)	Data 0.001 (0.076)	Loss 1.216 (0.809)
Epoch: [47][140/200]	Time 0.262 (0.342)	Data 0.001 (0.075)	Loss 1.412 (0.839)
Epoch: [47][160/200]	Time 0.262 (0.341)	Data 0.001 (0.074)	Loss 1.095 (0.861)
Epoch: [47][180/200]	Time 0.263 (0.340)	Data 0.001 (0.073)	Loss 1.024 (0.874)
Epoch: [47][200/200]	Time 0.261 (0.339)	Data 0.001 (0.073)	Loss 1.224 (0.888)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.132)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.550020456314087
==> Statistics for epoch 48: 610 clusters
Epoch: [48][20/200]	Time 1.733 (0.385)	Data 1.448 (0.117)	Loss 1.106 (0.184)
Epoch: [48][40/200]	Time 0.289 (0.361)	Data 0.001 (0.094)	Loss 0.702 (0.551)
Epoch: [48][60/200]	Time 0.267 (0.350)	Data 0.001 (0.084)	Loss 1.139 (0.682)
Epoch: [48][80/200]	Time 0.259 (0.345)	Data 0.001 (0.079)	Loss 0.928 (0.749)
Epoch: [48][100/200]	Time 0.259 (0.343)	Data 0.001 (0.077)	Loss 1.194 (0.802)
Epoch: [48][120/200]	Time 0.258 (0.342)	Data 0.001 (0.076)	Loss 0.742 (0.820)
Epoch: [48][140/200]	Time 0.259 (0.343)	Data 0.001 (0.077)	Loss 0.964 (0.845)
Epoch: [48][160/200]	Time 0.257 (0.342)	Data 0.001 (0.076)	Loss 0.897 (0.849)
Epoch: [48][180/200]	Time 0.264 (0.341)	Data 0.001 (0.075)	Loss 1.203 (0.854)
Epoch: [48][200/200]	Time 0.260 (0.341)	Data 0.001 (0.075)	Loss 1.232 (0.862)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.134)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.127789735794067
==> Statistics for epoch 49: 615 clusters
Epoch: [49][20/200]	Time 1.684 (0.373)	Data 1.396 (0.110)	Loss 0.806 (0.160)
Epoch: [49][40/200]	Time 0.262 (0.354)	Data 0.001 (0.089)	Loss 0.733 (0.532)
Epoch: [49][60/200]	Time 0.261 (0.347)	Data 0.001 (0.082)	Loss 0.997 (0.682)
Epoch: [49][80/200]	Time 0.266 (0.345)	Data 0.001 (0.079)	Loss 0.861 (0.755)
Epoch: [49][100/200]	Time 0.258 (0.344)	Data 0.001 (0.078)	Loss 0.690 (0.807)
Epoch: [49][120/200]	Time 0.259 (0.342)	Data 0.001 (0.076)	Loss 0.894 (0.825)
Epoch: [49][140/200]	Time 0.261 (0.341)	Data 0.001 (0.076)	Loss 1.055 (0.848)
Epoch: [49][160/200]	Time 0.259 (0.341)	Data 0.001 (0.075)	Loss 1.128 (0.861)
Epoch: [49][180/200]	Time 0.262 (0.340)	Data 0.001 (0.075)	Loss 0.771 (0.871)
Epoch: [49][200/200]	Time 0.259 (0.340)	Data 0.001 (0.074)	Loss 1.028 (0.884)
Extract Features: [50/76]	Time 0.093 (0.132)	Data 0.000 (0.033)	
Mean AP: 91.4%

 * Finished epoch  49  model mAP: 91.4%  best: 91.4% *

==> Test with the best model:
=> Loaded checkpoint 'log/market/resnet50_ibn_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.093 (0.136)	Data 0.000 (0.036)	
Mean AP: 91.4%
CMC Scores:
  top-1          96.1%
  top-5          98.5%
  top-10         99.1%
Total running time:  1:24:46.687388
