==========
Args:Namespace(dataset='msmt17', batch_size=256, workers=4, height=256, width=128, num_instances=8, eps=0.7, eps_gap=0.02, k1=30, k2=6, arch='vit_small', pretrained_path='/home/ma-user/work/Projects/ReIDNet_Finetune/TransReID/log/bs_exp/ViT_Small_Market1501/64bs_lr0.0004_ep120_warm20_seed0/vit_small_120.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/market2msmt/vit_small_cion', pooling_type='gem', use_hard=True)
==========
==> Load unlabeled dataset
=> MSMT17 loaded
Dataset statistics:
  ----------------------------------------
  subset   | # ids | # images | # cameras
  ----------------------------------------
  train    |  1041 |    32621 |        15
  query    |  3060 |    11659 |        15
  gallery  |  3060 |    82161 |        15
  ----------------------------------------
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
Load 172 / 177 layers.
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.366)	Data 0.000 (0.073)	
Extract Features: [100/128]	Time 0.127 (0.271)	Data 0.000 (0.060)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.909541845321655
Clustering criterion: eps: 0.700
==> Statistics for epoch 0: 596 clusters
Epoch: [0][20/200]	Time 0.384 (1.367)	Data 0.001 (0.123)	Loss 5.670 (6.258)
Epoch: [0][40/200]	Time 0.381 (0.913)	Data 0.001 (0.100)	Loss 4.052 (5.703)
Epoch: [0][60/200]	Time 0.488 (0.762)	Data 0.001 (0.091)	Loss 4.189 (5.013)
Epoch: [0][80/200]	Time 0.378 (0.687)	Data 0.000 (0.088)	Loss 3.215 (4.503)
Epoch: [0][100/200]	Time 0.377 (0.640)	Data 0.000 (0.084)	Loss 2.484 (4.129)
Epoch: [0][120/200]	Time 0.385 (0.609)	Data 0.000 (0.082)	Loss 2.096 (3.875)
Epoch: [0][140/200]	Time 0.378 (0.587)	Data 0.000 (0.081)	Loss 1.793 (3.655)
Epoch: [0][160/200]	Time 0.385 (0.571)	Data 0.000 (0.080)	Loss 2.124 (3.480)
Epoch: [0][180/200]	Time 0.378 (0.558)	Data 0.000 (0.079)	Loss 2.167 (3.330)
Epoch: [0][200/200]	Time 0.377 (0.555)	Data 0.000 (0.086)	Loss 1.988 (3.206)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.206)	Data 0.000 (0.076)	
Extract Features: [100/128]	Time 0.130 (0.190)	Data 0.000 (0.060)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.43447828292847
==> Statistics for epoch 1: 882 clusters
Epoch: [1][20/200]	Time 0.380 (0.430)	Data 0.001 (0.047)	Loss 0.393 (0.573)
Epoch: [1][40/200]	Time 0.381 (0.444)	Data 0.001 (0.062)	Loss 1.476 (1.100)
Epoch: [1][60/200]	Time 0.382 (0.453)	Data 0.000 (0.069)	Loss 2.496 (1.458)
Epoch: [1][80/200]	Time 0.375 (0.436)	Data 0.000 (0.052)	Loss 2.298 (1.661)
Epoch: [1][100/200]	Time 0.380 (0.441)	Data 0.001 (0.058)	Loss 1.932 (1.757)
Epoch: [1][120/200]	Time 0.386 (0.444)	Data 0.001 (0.061)	Loss 1.988 (1.831)
Epoch: [1][140/200]	Time 0.389 (0.446)	Data 0.001 (0.063)	Loss 2.389 (1.867)
Epoch: [1][160/200]	Time 0.389 (0.439)	Data 0.000 (0.055)	Loss 2.240 (1.887)
Epoch: [1][180/200]	Time 0.382 (0.443)	Data 0.001 (0.058)	Loss 1.715 (1.905)
Epoch: [1][200/200]	Time 0.386 (0.445)	Data 0.000 (0.061)	Loss 1.755 (1.906)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.387 (0.211)	Data 0.263 (0.080)	
Extract Features: [100/128]	Time 0.126 (0.192)	Data 0.000 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.394899129867554
==> Statistics for epoch 2: 890 clusters
Epoch: [2][20/200]	Time 0.379 (0.426)	Data 0.001 (0.046)	Loss 0.473 (0.477)
Epoch: [2][40/200]	Time 0.383 (0.447)	Data 0.001 (0.063)	Loss 1.964 (0.946)
Epoch: [2][60/200]	Time 0.380 (0.453)	Data 0.001 (0.070)	Loss 2.112 (1.320)
Epoch: [2][80/200]	Time 0.377 (0.436)	Data 0.000 (0.052)	Loss 1.699 (1.482)
Epoch: [2][100/200]	Time 0.384 (0.440)	Data 0.001 (0.056)	Loss 2.097 (1.592)
Epoch: [2][120/200]	Time 0.377 (0.445)	Data 0.001 (0.061)	Loss 1.862 (1.664)
Epoch: [2][140/200]	Time 0.387 (0.447)	Data 0.001 (0.063)	Loss 1.613 (1.707)
Epoch: [2][160/200]	Time 0.379 (0.439)	Data 0.000 (0.056)	Loss 2.179 (1.733)
Epoch: [2][180/200]	Time 0.377 (0.442)	Data 0.001 (0.059)	Loss 1.723 (1.753)
Epoch: [2][200/200]	Time 0.378 (0.443)	Data 0.001 (0.060)	Loss 1.793 (1.756)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.163 (0.215)	Data 0.035 (0.084)	
Extract Features: [100/128]	Time 0.126 (0.194)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.48096251487732
==> Statistics for epoch 3: 935 clusters
Epoch: [3][20/200]	Time 0.383 (0.440)	Data 0.001 (0.057)	Loss 0.264 (0.447)
Epoch: [3][40/200]	Time 0.383 (0.457)	Data 0.001 (0.071)	Loss 1.231 (0.802)
Epoch: [3][60/200]	Time 0.380 (0.458)	Data 0.001 (0.073)	Loss 2.058 (1.160)
Epoch: [3][80/200]	Time 0.377 (0.439)	Data 0.001 (0.055)	Loss 1.717 (1.348)
Epoch: [3][100/200]	Time 0.383 (0.443)	Data 0.001 (0.059)	Loss 1.635 (1.469)
Epoch: [3][120/200]	Time 0.386 (0.447)	Data 0.001 (0.063)	Loss 1.695 (1.554)
Epoch: [3][140/200]	Time 0.382 (0.438)	Data 0.000 (0.054)	Loss 2.186 (1.600)
Epoch: [3][160/200]	Time 0.379 (0.442)	Data 0.000 (0.058)	Loss 2.242 (1.639)
Epoch: [3][180/200]	Time 0.384 (0.444)	Data 0.000 (0.060)	Loss 1.851 (1.664)
Epoch: [3][200/200]	Time 0.380 (0.438)	Data 0.000 (0.054)	Loss 2.329 (1.684)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.128 (0.209)	Data 0.000 (0.077)	
Extract Features: [100/128]	Time 0.130 (0.189)	Data 0.004 (0.059)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.190258502960205
==> Statistics for epoch 4: 967 clusters
Epoch: [4][20/200]	Time 0.379 (0.438)	Data 0.001 (0.051)	Loss 0.257 (0.462)
Epoch: [4][40/200]	Time 0.387 (0.451)	Data 0.001 (0.067)	Loss 2.202 (0.806)
Epoch: [4][60/200]	Time 0.380 (0.427)	Data 0.000 (0.045)	Loss 2.050 (1.173)
Epoch: [4][80/200]	Time 0.383 (0.438)	Data 0.001 (0.053)	Loss 1.751 (1.349)
Epoch: [4][100/200]	Time 0.381 (0.444)	Data 0.001 (0.060)	Loss 1.196 (1.458)
Epoch: [4][120/200]	Time 0.383 (0.434)	Data 0.000 (0.050)	Loss 2.056 (1.553)
Epoch: [4][140/200]	Time 0.383 (0.438)	Data 0.001 (0.054)	Loss 1.777 (1.604)
Epoch: [4][160/200]	Time 0.384 (0.442)	Data 0.001 (0.057)	Loss 2.133 (1.626)
Epoch: [4][180/200]	Time 0.380 (0.435)	Data 0.000 (0.051)	Loss 1.850 (1.660)
Epoch: [4][200/200]	Time 0.386 (0.438)	Data 0.001 (0.054)	Loss 1.822 (1.683)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.172 (0.208)	Data 0.047 (0.079)	
Extract Features: [100/128]	Time 0.130 (0.191)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.97732448577881
==> Statistics for epoch 5: 971 clusters
Epoch: [5][20/200]	Time 0.379 (0.436)	Data 0.001 (0.056)	Loss 0.448 (0.463)
Epoch: [5][40/200]	Time 0.382 (0.446)	Data 0.001 (0.065)	Loss 2.089 (0.771)
Epoch: [5][60/200]	Time 0.381 (0.426)	Data 0.000 (0.044)	Loss 1.879 (1.171)
Epoch: [5][80/200]	Time 0.386 (0.436)	Data 0.001 (0.053)	Loss 2.294 (1.353)
Epoch: [5][100/200]	Time 0.384 (0.442)	Data 0.001 (0.060)	Loss 1.966 (1.458)
Epoch: [5][120/200]	Time 0.388 (0.434)	Data 0.000 (0.050)	Loss 2.287 (1.546)
Epoch: [5][140/200]	Time 0.386 (0.439)	Data 0.000 (0.055)	Loss 2.199 (1.598)
Epoch: [5][160/200]	Time 0.383 (0.442)	Data 0.000 (0.058)	Loss 1.915 (1.641)
Epoch: [5][180/200]	Time 0.385 (0.435)	Data 0.000 (0.051)	Loss 1.487 (1.670)
Epoch: [5][200/200]	Time 0.380 (0.439)	Data 0.001 (0.054)	Loss 2.057 (1.692)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.124 (0.211)	Data 0.000 (0.082)	
Extract Features: [100/128]	Time 0.127 (0.191)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.4833505153656
==> Statistics for epoch 6: 929 clusters
Epoch: [6][20/200]	Time 0.379 (0.430)	Data 0.001 (0.051)	Loss 0.413 (0.418)
Epoch: [6][40/200]	Time 0.380 (0.442)	Data 0.001 (0.063)	Loss 2.209 (0.760)
Epoch: [6][60/200]	Time 0.377 (0.446)	Data 0.001 (0.068)	Loss 2.043 (1.130)
Epoch: [6][80/200]	Time 0.381 (0.430)	Data 0.000 (0.051)	Loss 1.689 (1.308)
Epoch: [6][100/200]	Time 0.378 (0.435)	Data 0.001 (0.056)	Loss 2.402 (1.408)
Epoch: [6][120/200]	Time 0.383 (0.439)	Data 0.000 (0.059)	Loss 1.611 (1.486)
Epoch: [6][140/200]	Time 0.374 (0.431)	Data 0.000 (0.051)	Loss 1.714 (1.536)
Epoch: [6][160/200]	Time 0.380 (0.435)	Data 0.000 (0.054)	Loss 1.768 (1.573)
Epoch: [6][180/200]	Time 0.382 (0.438)	Data 0.001 (0.057)	Loss 1.837 (1.585)
Epoch: [6][200/200]	Time 0.371 (0.432)	Data 0.000 (0.051)	Loss 2.347 (1.616)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.279 (0.204)	Data 0.152 (0.074)	
Extract Features: [100/128]	Time 0.128 (0.187)	Data 0.000 (0.058)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.52188801765442
==> Statistics for epoch 7: 930 clusters
Epoch: [7][20/200]	Time 0.371 (0.431)	Data 0.001 (0.050)	Loss 0.335 (0.407)
Epoch: [7][40/200]	Time 0.380 (0.442)	Data 0.001 (0.061)	Loss 1.808 (0.762)
Epoch: [7][60/200]	Time 0.387 (0.448)	Data 0.001 (0.065)	Loss 1.388 (1.093)
Epoch: [7][80/200]	Time 0.382 (0.431)	Data 0.000 (0.049)	Loss 1.677 (1.276)
Epoch: [7][100/200]	Time 0.387 (0.437)	Data 0.001 (0.054)	Loss 1.588 (1.359)
Epoch: [7][120/200]	Time 0.387 (0.442)	Data 0.001 (0.057)	Loss 1.846 (1.441)
Epoch: [7][140/200]	Time 0.376 (0.433)	Data 0.000 (0.049)	Loss 1.617 (1.487)
Epoch: [7][160/200]	Time 0.377 (0.436)	Data 0.001 (0.053)	Loss 1.742 (1.530)
Epoch: [7][180/200]	Time 0.379 (0.438)	Data 0.000 (0.055)	Loss 2.064 (1.560)
Epoch: [7][200/200]	Time 0.383 (0.433)	Data 0.000 (0.049)	Loss 1.621 (1.575)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.128 (0.211)	Data 0.001 (0.081)	
Extract Features: [100/128]	Time 0.240 (0.193)	Data 0.115 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.00420951843262
==> Statistics for epoch 8: 965 clusters
Epoch: [8][20/200]	Time 0.380 (0.439)	Data 0.000 (0.051)	Loss 0.498 (0.407)
Epoch: [8][40/200]	Time 0.385 (0.447)	Data 0.001 (0.063)	Loss 1.647 (0.758)
Epoch: [8][60/200]	Time 0.372 (0.424)	Data 0.000 (0.042)	Loss 2.129 (1.087)
Epoch: [8][80/200]	Time 0.382 (0.434)	Data 0.001 (0.051)	Loss 1.597 (1.262)
Epoch: [8][100/200]	Time 0.380 (0.438)	Data 0.001 (0.055)	Loss 1.805 (1.368)
Epoch: [8][120/200]	Time 0.381 (0.428)	Data 0.000 (0.046)	Loss 1.933 (1.448)
Epoch: [8][140/200]	Time 0.380 (0.433)	Data 0.001 (0.050)	Loss 1.408 (1.507)
Epoch: [8][160/200]	Time 0.383 (0.437)	Data 0.001 (0.054)	Loss 1.622 (1.538)
Epoch: [8][180/200]	Time 0.378 (0.431)	Data 0.000 (0.048)	Loss 1.906 (1.565)
Epoch: [8][200/200]	Time 0.384 (0.434)	Data 0.001 (0.050)	Loss 2.164 (1.582)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.126 (0.215)	Data 0.000 (0.089)	
Extract Features: [100/128]	Time 0.125 (0.195)	Data 0.000 (0.067)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.28940796852112
==> Statistics for epoch 9: 958 clusters
Epoch: [9][20/200]	Time 0.383 (0.444)	Data 0.001 (0.056)	Loss 0.400 (0.399)
Epoch: [9][40/200]	Time 0.380 (0.448)	Data 0.000 (0.064)	Loss 2.321 (0.773)
Epoch: [9][60/200]	Time 0.382 (0.452)	Data 0.001 (0.069)	Loss 2.001 (1.075)
Epoch: [9][80/200]	Time 0.386 (0.438)	Data 0.002 (0.052)	Loss 1.756 (1.263)
Epoch: [9][100/200]	Time 0.386 (0.443)	Data 0.001 (0.057)	Loss 2.000 (1.374)
Epoch: [9][120/200]	Time 0.385 (0.445)	Data 0.001 (0.060)	Loss 1.591 (1.441)
Epoch: [9][140/200]	Time 0.380 (0.436)	Data 0.000 (0.052)	Loss 1.543 (1.500)
Epoch: [9][160/200]	Time 0.388 (0.439)	Data 0.001 (0.054)	Loss 1.822 (1.539)
Epoch: [9][180/200]	Time 0.382 (0.441)	Data 0.001 (0.057)	Loss 1.674 (1.565)
Epoch: [9][200/200]	Time 0.386 (0.436)	Data 0.000 (0.051)	Loss 1.689 (1.585)
Extract Features: [50/367]	Time 0.135 (0.211)	Data 0.000 (0.080)	
Extract Features: [100/367]	Time 0.128 (0.194)	Data 0.000 (0.065)	
Extract Features: [150/367]	Time 0.127 (0.192)	Data 0.000 (0.063)	
Extract Features: [200/367]	Time 0.126 (0.188)	Data 0.000 (0.058)	
Extract Features: [250/367]	Time 0.132 (0.187)	Data 0.001 (0.056)	
Extract Features: [300/367]	Time 0.135 (0.185)	Data 0.009 (0.055)	
Extract Features: [350/367]	Time 0.146 (0.186)	Data 0.000 (0.053)	
Mean AP: 45.3%

 * Finished epoch   9  model mAP: 45.3%  best: 45.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.214)	Data 0.000 (0.087)	
Extract Features: [100/128]	Time 0.128 (0.192)	Data 0.000 (0.066)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.42216181755066
==> Statistics for epoch 10: 945 clusters
Epoch: [10][20/200]	Time 0.381 (0.442)	Data 0.001 (0.055)	Loss 0.546 (0.369)
Epoch: [10][40/200]	Time 0.381 (0.455)	Data 0.001 (0.068)	Loss 1.468 (0.682)
Epoch: [10][60/200]	Time 0.381 (0.454)	Data 0.001 (0.069)	Loss 1.186 (1.026)
Epoch: [10][80/200]	Time 0.381 (0.435)	Data 0.001 (0.052)	Loss 1.780 (1.208)
Epoch: [10][100/200]	Time 0.383 (0.441)	Data 0.001 (0.057)	Loss 2.164 (1.305)
Epoch: [10][120/200]	Time 0.374 (0.444)	Data 0.001 (0.060)	Loss 1.569 (1.382)
Epoch: [10][140/200]	Time 0.373 (0.435)	Data 0.000 (0.052)	Loss 1.765 (1.426)
Epoch: [10][160/200]	Time 0.379 (0.439)	Data 0.001 (0.056)	Loss 1.971 (1.460)
Epoch: [10][180/200]	Time 0.383 (0.442)	Data 0.001 (0.058)	Loss 1.784 (1.496)
Epoch: [10][200/200]	Time 0.383 (0.436)	Data 0.000 (0.053)	Loss 1.265 (1.514)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.210)	Data 0.000 (0.082)	
Extract Features: [100/128]	Time 0.127 (0.193)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.01003551483154
==> Statistics for epoch 11: 942 clusters
Epoch: [11][20/200]	Time 0.384 (0.435)	Data 0.001 (0.054)	Loss 0.214 (0.341)
Epoch: [11][40/200]	Time 0.381 (0.450)	Data 0.001 (0.068)	Loss 1.571 (0.715)
Epoch: [11][60/200]	Time 0.375 (0.454)	Data 0.001 (0.072)	Loss 1.523 (1.025)
Epoch: [11][80/200]	Time 0.387 (0.437)	Data 0.001 (0.054)	Loss 1.833 (1.198)
Epoch: [11][100/200]	Time 0.377 (0.442)	Data 0.001 (0.060)	Loss 1.493 (1.302)
Epoch: [11][120/200]	Time 0.383 (0.447)	Data 0.001 (0.063)	Loss 1.417 (1.366)
Epoch: [11][140/200]	Time 0.386 (0.438)	Data 0.000 (0.054)	Loss 1.591 (1.407)
Epoch: [11][160/200]	Time 0.383 (0.441)	Data 0.001 (0.058)	Loss 1.869 (1.451)
Epoch: [11][180/200]	Time 0.382 (0.444)	Data 0.001 (0.060)	Loss 1.416 (1.475)
Epoch: [11][200/200]	Time 0.380 (0.438)	Data 0.000 (0.054)	Loss 1.830 (1.493)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.209)	Data 0.001 (0.081)	
Extract Features: [100/128]	Time 0.128 (0.191)	Data 0.000 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.16776704788208
==> Statistics for epoch 12: 951 clusters
Epoch: [12][20/200]	Time 0.371 (0.434)	Data 0.001 (0.053)	Loss 0.358 (0.350)
Epoch: [12][40/200]	Time 0.375 (0.448)	Data 0.000 (0.065)	Loss 1.578 (0.699)
Epoch: [12][60/200]	Time 0.381 (0.452)	Data 0.001 (0.070)	Loss 1.608 (1.007)
Epoch: [12][80/200]	Time 0.382 (0.434)	Data 0.001 (0.053)	Loss 1.060 (1.159)
Epoch: [12][100/200]	Time 0.384 (0.441)	Data 0.000 (0.059)	Loss 1.792 (1.275)
Epoch: [12][120/200]	Time 0.384 (0.444)	Data 0.000 (0.062)	Loss 1.810 (1.350)
Epoch: [12][140/200]	Time 0.380 (0.435)	Data 0.000 (0.053)	Loss 1.216 (1.399)
Epoch: [12][160/200]	Time 0.393 (0.438)	Data 0.001 (0.056)	Loss 1.679 (1.439)
Epoch: [12][180/200]	Time 0.379 (0.441)	Data 0.000 (0.059)	Loss 1.510 (1.469)
Epoch: [12][200/200]	Time 0.382 (0.435)	Data 0.000 (0.053)	Loss 1.829 (1.487)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.357 (0.209)	Data 0.233 (0.079)	
Extract Features: [100/128]	Time 0.346 (0.191)	Data 0.214 (0.061)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.41526198387146
==> Statistics for epoch 13: 957 clusters
Epoch: [13][20/200]	Time 0.366 (0.432)	Data 0.001 (0.053)	Loss 0.349 (0.360)
Epoch: [13][40/200]	Time 0.378 (0.445)	Data 0.001 (0.066)	Loss 1.607 (0.701)
Epoch: [13][60/200]	Time 0.378 (0.453)	Data 0.001 (0.070)	Loss 1.667 (1.004)
Epoch: [13][80/200]	Time 0.380 (0.435)	Data 0.001 (0.053)	Loss 1.632 (1.169)
Epoch: [13][100/200]	Time 0.383 (0.439)	Data 0.001 (0.057)	Loss 2.157 (1.273)
Epoch: [13][120/200]	Time 0.379 (0.443)	Data 0.001 (0.062)	Loss 1.219 (1.341)
Epoch: [13][140/200]	Time 0.386 (0.436)	Data 0.000 (0.053)	Loss 1.568 (1.387)
Epoch: [13][160/200]	Time 0.393 (0.439)	Data 0.001 (0.057)	Loss 2.135 (1.432)
Epoch: [13][180/200]	Time 0.388 (0.442)	Data 0.001 (0.059)	Loss 1.624 (1.453)
Epoch: [13][200/200]	Time 0.381 (0.436)	Data 0.000 (0.053)	Loss 1.991 (1.484)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.215)	Data 0.000 (0.087)	
Extract Features: [100/128]	Time 0.126 (0.194)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.92729663848877
==> Statistics for epoch 14: 974 clusters
Epoch: [14][20/200]	Time 0.378 (0.434)	Data 0.001 (0.054)	Loss 0.163 (0.316)
Epoch: [14][40/200]	Time 0.386 (0.455)	Data 0.001 (0.070)	Loss 1.274 (0.619)
Epoch: [14][60/200]	Time 0.384 (0.430)	Data 0.000 (0.047)	Loss 1.110 (0.940)
Epoch: [14][80/200]	Time 0.382 (0.440)	Data 0.001 (0.057)	Loss 1.761 (1.095)
Epoch: [14][100/200]	Time 0.378 (0.446)	Data 0.001 (0.062)	Loss 1.924 (1.206)
Epoch: [14][120/200]	Time 0.384 (0.435)	Data 0.000 (0.052)	Loss 1.272 (1.276)
Epoch: [14][140/200]	Time 0.383 (0.441)	Data 0.001 (0.057)	Loss 1.894 (1.329)
Epoch: [14][160/200]	Time 0.382 (0.445)	Data 0.001 (0.061)	Loss 1.939 (1.363)
Epoch: [14][180/200]	Time 0.384 (0.438)	Data 0.000 (0.054)	Loss 1.539 (1.394)
Epoch: [14][200/200]	Time 0.381 (0.441)	Data 0.001 (0.057)	Loss 1.606 (1.421)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.128 (0.212)	Data 0.000 (0.083)	
Extract Features: [100/128]	Time 0.126 (0.191)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.14959907531738
==> Statistics for epoch 15: 968 clusters
Epoch: [15][20/200]	Time 0.379 (0.438)	Data 0.001 (0.058)	Loss 0.315 (0.348)
Epoch: [15][40/200]	Time 0.378 (0.452)	Data 0.001 (0.067)	Loss 1.755 (0.659)
Epoch: [15][60/200]	Time 0.384 (0.428)	Data 0.000 (0.045)	Loss 2.108 (0.995)
Epoch: [15][80/200]	Time 0.377 (0.437)	Data 0.001 (0.053)	Loss 1.460 (1.147)
Epoch: [15][100/200]	Time 0.381 (0.443)	Data 0.001 (0.059)	Loss 2.495 (1.238)
Epoch: [15][120/200]	Time 0.379 (0.433)	Data 0.000 (0.050)	Loss 1.096 (1.301)
Epoch: [15][140/200]	Time 0.383 (0.437)	Data 0.001 (0.054)	Loss 1.779 (1.347)
Epoch: [15][160/200]	Time 0.386 (0.440)	Data 0.001 (0.057)	Loss 1.569 (1.381)
Epoch: [15][180/200]	Time 0.378 (0.434)	Data 0.000 (0.051)	Loss 2.286 (1.401)
Epoch: [15][200/200]	Time 0.378 (0.437)	Data 0.001 (0.054)	Loss 1.859 (1.418)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.125 (0.213)	Data 0.000 (0.085)	
Extract Features: [100/128]	Time 0.126 (0.192)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.201093912124634
==> Statistics for epoch 16: 983 clusters
Epoch: [16][20/200]	Time 0.380 (0.428)	Data 0.001 (0.049)	Loss 0.307 (0.332)
Epoch: [16][40/200]	Time 0.382 (0.443)	Data 0.001 (0.063)	Loss 1.397 (0.646)
Epoch: [16][60/200]	Time 0.378 (0.421)	Data 0.000 (0.042)	Loss 1.724 (0.933)
Epoch: [16][80/200]	Time 0.378 (0.432)	Data 0.001 (0.051)	Loss 1.427 (1.090)
Epoch: [16][100/200]	Time 0.385 (0.437)	Data 0.001 (0.057)	Loss 1.662 (1.185)
Epoch: [16][120/200]	Time 0.377 (0.428)	Data 0.000 (0.047)	Loss 1.353 (1.241)
Epoch: [16][140/200]	Time 0.381 (0.434)	Data 0.001 (0.052)	Loss 1.872 (1.285)
Epoch: [16][160/200]	Time 0.383 (0.437)	Data 0.001 (0.055)	Loss 1.920 (1.326)
Epoch: [16][180/200]	Time 0.381 (0.431)	Data 0.000 (0.049)	Loss 1.946 (1.354)
Epoch: [16][200/200]	Time 0.381 (0.435)	Data 0.001 (0.052)	Loss 1.351 (1.362)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.126 (0.215)	Data 0.000 (0.086)	
Extract Features: [100/128]	Time 0.124 (0.193)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.46621608734131
==> Statistics for epoch 17: 994 clusters
Epoch: [17][20/200]	Time 0.383 (0.435)	Data 0.001 (0.054)	Loss 0.315 (0.269)
Epoch: [17][40/200]	Time 0.375 (0.448)	Data 0.001 (0.067)	Loss 1.185 (0.530)
Epoch: [17][60/200]	Time 0.381 (0.425)	Data 0.000 (0.045)	Loss 1.446 (0.846)
Epoch: [17][80/200]	Time 0.381 (0.436)	Data 0.001 (0.056)	Loss 1.716 (0.998)
Epoch: [17][100/200]	Time 0.380 (0.441)	Data 0.001 (0.060)	Loss 1.704 (1.104)
Epoch: [17][120/200]	Time 0.382 (0.431)	Data 0.000 (0.050)	Loss 1.378 (1.174)
Epoch: [17][140/200]	Time 0.385 (0.435)	Data 0.001 (0.053)	Loss 1.481 (1.227)
Epoch: [17][160/200]	Time 0.390 (0.440)	Data 0.001 (0.057)	Loss 1.396 (1.260)
Epoch: [17][180/200]	Time 0.376 (0.434)	Data 0.000 (0.051)	Loss 1.586 (1.288)
Epoch: [17][200/200]	Time 0.383 (0.437)	Data 0.001 (0.054)	Loss 1.598 (1.310)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.126 (0.213)	Data 0.000 (0.083)	
Extract Features: [100/128]	Time 0.253 (0.194)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.112871408462524
==> Statistics for epoch 18: 1010 clusters
Epoch: [18][20/200]	Time 0.381 (0.433)	Data 0.001 (0.050)	Loss 0.305 (0.278)
Epoch: [18][40/200]	Time 0.385 (0.448)	Data 0.001 (0.066)	Loss 1.427 (0.526)
Epoch: [18][60/200]	Time 0.380 (0.426)	Data 0.000 (0.044)	Loss 1.890 (0.866)
Epoch: [18][80/200]	Time 0.383 (0.436)	Data 0.001 (0.052)	Loss 1.539 (1.013)
Epoch: [18][100/200]	Time 0.399 (0.443)	Data 0.002 (0.059)	Loss 1.621 (1.092)
Epoch: [18][120/200]	Time 0.378 (0.433)	Data 0.000 (0.049)	Loss 1.374 (1.154)
Epoch: [18][140/200]	Time 0.386 (0.439)	Data 0.001 (0.054)	Loss 1.648 (1.200)
Epoch: [18][160/200]	Time 0.382 (0.442)	Data 0.001 (0.058)	Loss 1.061 (1.241)
Epoch: [18][180/200]	Time 0.384 (0.435)	Data 0.000 (0.051)	Loss 1.452 (1.270)
Epoch: [18][200/200]	Time 0.382 (0.438)	Data 0.001 (0.054)	Loss 1.084 (1.298)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.125 (0.210)	Data 0.000 (0.080)	
Extract Features: [100/128]	Time 0.129 (0.193)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.1904091835022
==> Statistics for epoch 19: 990 clusters
Epoch: [19][20/200]	Time 0.379 (0.432)	Data 0.001 (0.051)	Loss 0.478 (0.299)
Epoch: [19][40/200]	Time 0.377 (0.445)	Data 0.000 (0.064)	Loss 1.284 (0.555)
Epoch: [19][60/200]	Time 0.379 (0.424)	Data 0.000 (0.043)	Loss 2.003 (0.864)
Epoch: [19][80/200]	Time 0.375 (0.433)	Data 0.000 (0.052)	Loss 1.471 (1.011)
Epoch: [19][100/200]	Time 0.378 (0.439)	Data 0.001 (0.058)	Loss 1.401 (1.125)
Epoch: [19][120/200]	Time 0.376 (0.430)	Data 0.000 (0.049)	Loss 1.467 (1.174)
Epoch: [19][140/200]	Time 0.378 (0.435)	Data 0.001 (0.053)	Loss 1.142 (1.219)
Epoch: [19][160/200]	Time 0.383 (0.438)	Data 0.001 (0.056)	Loss 1.592 (1.255)
Epoch: [19][180/200]	Time 0.383 (0.432)	Data 0.000 (0.050)	Loss 1.537 (1.281)
Epoch: [19][200/200]	Time 0.384 (0.435)	Data 0.000 (0.052)	Loss 1.036 (1.301)
Extract Features: [50/367]	Time 0.198 (0.210)	Data 0.076 (0.081)	
Extract Features: [100/367]	Time 0.127 (0.196)	Data 0.000 (0.069)	
Extract Features: [150/367]	Time 0.291 (0.192)	Data 0.172 (0.065)	
Extract Features: [200/367]	Time 0.126 (0.190)	Data 0.000 (0.062)	
Extract Features: [250/367]	Time 0.336 (0.189)	Data 0.214 (0.061)	
Extract Features: [300/367]	Time 0.229 (0.187)	Data 0.102 (0.059)	
Extract Features: [350/367]	Time 0.124 (0.186)	Data 0.000 (0.058)	
Mean AP: 55.2%

 * Finished epoch  19  model mAP: 55.2%  best: 55.2% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.123 (0.215)	Data 0.000 (0.088)	
Extract Features: [100/128]	Time 0.126 (0.194)	Data 0.001 (0.069)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.91979742050171
==> Statistics for epoch 20: 1016 clusters
Epoch: [20][20/200]	Time 0.378 (0.439)	Data 0.001 (0.058)	Loss 0.162 (0.281)
Epoch: [20][40/200]	Time 0.380 (0.450)	Data 0.001 (0.070)	Loss 2.022 (0.534)
Epoch: [20][60/200]	Time 0.384 (0.426)	Data 0.000 (0.047)	Loss 1.571 (0.814)
Epoch: [20][80/200]	Time 0.381 (0.436)	Data 0.001 (0.055)	Loss 1.860 (0.963)
Epoch: [20][100/200]	Time 0.384 (0.443)	Data 0.001 (0.061)	Loss 1.865 (1.062)
Epoch: [20][120/200]	Time 0.387 (0.433)	Data 0.000 (0.051)	Loss 1.622 (1.129)
Epoch: [20][140/200]	Time 0.385 (0.440)	Data 0.001 (0.057)	Loss 1.312 (1.165)
Epoch: [20][160/200]	Time 0.386 (0.444)	Data 0.001 (0.061)	Loss 1.491 (1.207)
Epoch: [20][180/200]	Time 0.381 (0.437)	Data 0.000 (0.054)	Loss 1.281 (1.232)
Epoch: [20][200/200]	Time 0.380 (0.440)	Data 0.001 (0.057)	Loss 1.532 (1.254)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.268 (0.217)	Data 0.000 (0.088)	
Extract Features: [100/128]	Time 0.126 (0.192)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.84030032157898
==> Statistics for epoch 21: 1036 clusters
Epoch: [21][20/200]	Time 0.383 (0.443)	Data 0.001 (0.056)	Loss 0.295 (0.286)
Epoch: [21][40/200]	Time 0.388 (0.453)	Data 0.001 (0.069)	Loss 1.387 (0.489)
Epoch: [21][60/200]	Time 0.380 (0.429)	Data 0.000 (0.046)	Loss 1.440 (0.787)
Epoch: [21][80/200]	Time 0.379 (0.439)	Data 0.001 (0.056)	Loss 1.366 (0.957)
Epoch: [21][100/200]	Time 0.387 (0.443)	Data 0.001 (0.061)	Loss 1.674 (1.048)
Epoch: [21][120/200]	Time 0.376 (0.434)	Data 0.001 (0.051)	Loss 1.068 (1.111)
Epoch: [21][140/200]	Time 0.382 (0.439)	Data 0.001 (0.055)	Loss 0.839 (1.154)
Epoch: [21][160/200]	Time 0.380 (0.432)	Data 0.000 (0.048)	Loss 1.685 (1.193)
Epoch: [21][180/200]	Time 0.395 (0.435)	Data 0.001 (0.052)	Loss 1.406 (1.219)
Epoch: [21][200/200]	Time 0.387 (0.440)	Data 0.001 (0.055)	Loss 1.330 (1.240)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.210)	Data 0.001 (0.081)	
Extract Features: [100/128]	Time 0.126 (0.191)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 57.66289758682251
==> Statistics for epoch 22: 1028 clusters
Epoch: [22][20/200]	Time 0.374 (0.435)	Data 0.001 (0.055)	Loss 0.482 (0.299)
Epoch: [22][40/200]	Time 0.378 (0.446)	Data 0.001 (0.066)	Loss 1.477 (0.496)
Epoch: [22][60/200]	Time 0.381 (0.427)	Data 0.000 (0.044)	Loss 1.588 (0.824)
Epoch: [22][80/200]	Time 0.383 (0.435)	Data 0.001 (0.053)	Loss 1.375 (0.975)
Epoch: [22][100/200]	Time 0.378 (0.441)	Data 0.001 (0.059)	Loss 1.414 (1.072)
Epoch: [22][120/200]	Time 0.378 (0.431)	Data 0.001 (0.049)	Loss 1.192 (1.138)
Epoch: [22][140/200]	Time 0.381 (0.437)	Data 0.001 (0.054)	Loss 1.272 (1.172)
Epoch: [22][160/200]	Time 0.378 (0.430)	Data 0.000 (0.048)	Loss 1.865 (1.199)
Epoch: [22][180/200]	Time 0.382 (0.434)	Data 0.001 (0.051)	Loss 1.867 (1.226)
Epoch: [22][200/200]	Time 0.392 (0.438)	Data 0.001 (0.055)	Loss 1.741 (1.255)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.125 (0.217)	Data 0.001 (0.091)	
Extract Features: [100/128]	Time 0.127 (0.196)	Data 0.000 (0.068)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.23072290420532
==> Statistics for epoch 23: 1034 clusters
Epoch: [23][20/200]	Time 0.372 (0.436)	Data 0.001 (0.053)	Loss 0.315 (0.282)
Epoch: [23][40/200]	Time 0.379 (0.447)	Data 0.001 (0.067)	Loss 1.121 (0.449)
Epoch: [23][60/200]	Time 0.375 (0.424)	Data 0.000 (0.045)	Loss 1.366 (0.755)
Epoch: [23][80/200]	Time 0.387 (0.433)	Data 0.001 (0.052)	Loss 1.835 (0.924)
Epoch: [23][100/200]	Time 0.386 (0.439)	Data 0.001 (0.058)	Loss 1.675 (1.022)
Epoch: [23][120/200]	Time 0.385 (0.429)	Data 0.001 (0.048)	Loss 1.637 (1.088)
Epoch: [23][140/200]	Time 0.380 (0.434)	Data 0.000 (0.053)	Loss 1.050 (1.126)
Epoch: [23][160/200]	Time 0.381 (0.427)	Data 0.000 (0.046)	Loss 1.268 (1.168)
Epoch: [23][180/200]	Time 0.379 (0.431)	Data 0.001 (0.050)	Loss 1.369 (1.189)
Epoch: [23][200/200]	Time 0.387 (0.434)	Data 0.001 (0.053)	Loss 1.643 (1.215)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.212)	Data 0.000 (0.083)	
Extract Features: [100/128]	Time 0.128 (0.194)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 63.31242275238037
==> Statistics for epoch 24: 1035 clusters
Epoch: [24][20/200]	Time 0.377 (0.437)	Data 0.001 (0.052)	Loss 0.341 (0.268)
Epoch: [24][40/200]	Time 0.395 (0.449)	Data 0.001 (0.066)	Loss 1.726 (0.493)
Epoch: [24][60/200]	Time 0.389 (0.427)	Data 0.000 (0.044)	Loss 1.478 (0.797)
Epoch: [24][80/200]	Time 0.380 (0.436)	Data 0.001 (0.053)	Loss 1.316 (0.942)
Epoch: [24][100/200]	Time 0.383 (0.442)	Data 0.001 (0.059)	Loss 1.573 (1.031)
Epoch: [24][120/200]	Time 0.377 (0.432)	Data 0.000 (0.049)	Loss 1.495 (1.094)
Epoch: [24][140/200]	Time 0.376 (0.437)	Data 0.001 (0.054)	Loss 1.223 (1.137)
Epoch: [24][160/200]	Time 0.378 (0.430)	Data 0.000 (0.047)	Loss 1.507 (1.170)
Epoch: [24][180/200]	Time 0.380 (0.434)	Data 0.000 (0.050)	Loss 1.545 (1.193)
Epoch: [24][200/200]	Time 0.388 (0.437)	Data 0.000 (0.053)	Loss 1.430 (1.222)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.133 (0.216)	Data 0.000 (0.088)	
Extract Features: [100/128]	Time 0.126 (0.195)	Data 0.000 (0.066)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.864582538604736
==> Statistics for epoch 25: 1034 clusters
Epoch: [25][20/200]	Time 0.376 (0.431)	Data 0.001 (0.052)	Loss 0.150 (0.274)
Epoch: [25][40/200]	Time 0.398 (0.451)	Data 0.001 (0.068)	Loss 1.076 (0.483)
Epoch: [25][60/200]	Time 0.378 (0.428)	Data 0.000 (0.045)	Loss 1.651 (0.794)
Epoch: [25][80/200]	Time 0.380 (0.436)	Data 0.001 (0.054)	Loss 1.296 (0.947)
Epoch: [25][100/200]	Time 0.382 (0.443)	Data 0.001 (0.059)	Loss 1.776 (1.021)
Epoch: [25][120/200]	Time 0.383 (0.433)	Data 0.001 (0.050)	Loss 1.457 (1.092)
Epoch: [25][140/200]	Time 0.380 (0.438)	Data 0.001 (0.055)	Loss 1.628 (1.130)
Epoch: [25][160/200]	Time 0.381 (0.431)	Data 0.000 (0.048)	Loss 1.413 (1.161)
Epoch: [25][180/200]	Time 0.383 (0.436)	Data 0.001 (0.053)	Loss 1.390 (1.185)
Epoch: [25][200/200]	Time 0.377 (0.438)	Data 0.001 (0.055)	Loss 1.271 (1.209)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.163 (0.211)	Data 0.038 (0.082)	
Extract Features: [100/128]	Time 0.127 (0.193)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.14886116981506
==> Statistics for epoch 26: 1030 clusters
Epoch: [26][20/200]	Time 0.378 (0.439)	Data 0.001 (0.058)	Loss 0.191 (0.271)
Epoch: [26][40/200]	Time 0.383 (0.454)	Data 0.001 (0.072)	Loss 1.367 (0.511)
Epoch: [26][60/200]	Time 0.380 (0.430)	Data 0.000 (0.049)	Loss 1.505 (0.797)
Epoch: [26][80/200]	Time 0.381 (0.439)	Data 0.001 (0.057)	Loss 1.266 (0.943)
Epoch: [26][100/200]	Time 0.377 (0.445)	Data 0.001 (0.063)	Loss 1.272 (1.034)
Epoch: [26][120/200]	Time 0.377 (0.435)	Data 0.001 (0.052)	Loss 2.018 (1.098)
Epoch: [26][140/200]	Time 0.382 (0.440)	Data 0.001 (0.057)	Loss 1.388 (1.142)
Epoch: [26][160/200]	Time 0.384 (0.432)	Data 0.000 (0.050)	Loss 1.424 (1.182)
Epoch: [26][180/200]	Time 0.381 (0.437)	Data 0.001 (0.054)	Loss 1.268 (1.210)
Epoch: [26][200/200]	Time 0.389 (0.441)	Data 0.001 (0.058)	Loss 1.760 (1.234)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.300 (0.212)	Data 0.174 (0.082)	
Extract Features: [100/128]	Time 0.129 (0.192)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.26004767417908
==> Statistics for epoch 27: 1037 clusters
Epoch: [27][20/200]	Time 0.387 (0.435)	Data 0.001 (0.056)	Loss 0.333 (0.256)
Epoch: [27][40/200]	Time 0.379 (0.448)	Data 0.000 (0.068)	Loss 1.182 (0.470)
Epoch: [27][60/200]	Time 0.378 (0.426)	Data 0.000 (0.046)	Loss 1.329 (0.753)
Epoch: [27][80/200]	Time 0.374 (0.436)	Data 0.000 (0.055)	Loss 1.306 (0.922)
Epoch: [27][100/200]	Time 0.382 (0.441)	Data 0.001 (0.060)	Loss 1.461 (1.015)
Epoch: [27][120/200]	Time 0.386 (0.432)	Data 0.001 (0.050)	Loss 1.212 (1.078)
Epoch: [27][140/200]	Time 0.385 (0.437)	Data 0.001 (0.055)	Loss 1.148 (1.126)
Epoch: [27][160/200]	Time 0.383 (0.430)	Data 0.000 (0.048)	Loss 1.127 (1.161)
Epoch: [27][180/200]	Time 0.378 (0.435)	Data 0.000 (0.052)	Loss 1.150 (1.185)
Epoch: [27][200/200]	Time 0.390 (0.439)	Data 0.000 (0.055)	Loss 1.110 (1.207)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.126 (0.209)	Data 0.000 (0.080)	
Extract Features: [100/128]	Time 0.128 (0.190)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.2370023727417
==> Statistics for epoch 28: 1029 clusters
Epoch: [28][20/200]	Time 0.377 (0.433)	Data 0.001 (0.054)	Loss 0.268 (0.271)
Epoch: [28][40/200]	Time 0.387 (0.451)	Data 0.001 (0.070)	Loss 1.581 (0.487)
Epoch: [28][60/200]	Time 0.387 (0.428)	Data 0.000 (0.047)	Loss 1.255 (0.800)
Epoch: [28][80/200]	Time 0.375 (0.438)	Data 0.001 (0.058)	Loss 1.201 (0.925)
Epoch: [28][100/200]	Time 0.381 (0.445)	Data 0.001 (0.062)	Loss 1.415 (1.012)
Epoch: [28][120/200]	Time 0.391 (0.435)	Data 0.001 (0.052)	Loss 1.252 (1.078)
Epoch: [28][140/200]	Time 0.392 (0.440)	Data 0.001 (0.057)	Loss 1.309 (1.120)
Epoch: [28][160/200]	Time 0.382 (0.434)	Data 0.000 (0.050)	Loss 1.070 (1.159)
Epoch: [28][180/200]	Time 0.385 (0.438)	Data 0.001 (0.054)	Loss 1.475 (1.186)
Epoch: [28][200/200]	Time 0.385 (0.441)	Data 0.001 (0.057)	Loss 1.105 (1.203)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.208)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.127 (0.189)	Data 0.000 (0.060)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.301433801651
==> Statistics for epoch 29: 1032 clusters
Epoch: [29][20/200]	Time 0.379 (0.436)	Data 0.001 (0.055)	Loss 0.282 (0.256)
Epoch: [29][40/200]	Time 0.382 (0.456)	Data 0.001 (0.072)	Loss 1.175 (0.476)
Epoch: [29][60/200]	Time 0.382 (0.431)	Data 0.000 (0.048)	Loss 1.030 (0.751)
Epoch: [29][80/200]	Time 0.380 (0.441)	Data 0.001 (0.058)	Loss 1.020 (0.904)
Epoch: [29][100/200]	Time 0.499 (0.448)	Data 0.001 (0.063)	Loss 1.560 (0.991)
Epoch: [29][120/200]	Time 0.384 (0.437)	Data 0.001 (0.053)	Loss 1.645 (1.079)
Epoch: [29][140/200]	Time 0.392 (0.443)	Data 0.001 (0.058)	Loss 1.412 (1.102)
Epoch: [29][160/200]	Time 0.382 (0.435)	Data 0.000 (0.051)	Loss 1.529 (1.144)
Epoch: [29][180/200]	Time 0.383 (0.439)	Data 0.001 (0.055)	Loss 1.535 (1.174)
Epoch: [29][200/200]	Time 0.380 (0.442)	Data 0.001 (0.058)	Loss 1.666 (1.195)
Extract Features: [50/367]	Time 0.132 (0.208)	Data 0.000 (0.078)	
Extract Features: [100/367]	Time 0.126 (0.193)	Data 0.000 (0.065)	
Extract Features: [150/367]	Time 0.172 (0.190)	Data 0.049 (0.061)	
Extract Features: [200/367]	Time 0.133 (0.187)	Data 0.000 (0.058)	
Extract Features: [250/367]	Time 0.336 (0.186)	Data 0.209 (0.057)	
Extract Features: [300/367]	Time 0.126 (0.184)	Data 0.000 (0.055)	
Extract Features: [350/367]	Time 0.169 (0.184)	Data 0.041 (0.055)	
Mean AP: 57.3%

 * Finished epoch  29  model mAP: 57.3%  best: 57.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.388 (0.213)	Data 0.265 (0.088)	
Extract Features: [100/128]	Time 0.123 (0.194)	Data 0.000 (0.069)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.85130453109741
==> Statistics for epoch 30: 1025 clusters
Epoch: [30][20/200]	Time 0.380 (0.440)	Data 0.001 (0.054)	Loss 0.238 (0.270)
Epoch: [30][40/200]	Time 0.375 (0.454)	Data 0.001 (0.070)	Loss 1.319 (0.489)
Epoch: [30][60/200]	Time 0.381 (0.429)	Data 0.000 (0.047)	Loss 1.153 (0.770)
Epoch: [30][80/200]	Time 0.385 (0.439)	Data 0.001 (0.056)	Loss 1.607 (0.939)
Epoch: [30][100/200]	Time 0.379 (0.445)	Data 0.001 (0.061)	Loss 1.335 (1.047)
Epoch: [30][120/200]	Time 0.381 (0.434)	Data 0.001 (0.051)	Loss 1.516 (1.114)
Epoch: [30][140/200]	Time 0.387 (0.439)	Data 0.001 (0.055)	Loss 1.140 (1.164)
Epoch: [30][160/200]	Time 0.379 (0.432)	Data 0.000 (0.049)	Loss 1.432 (1.183)
Epoch: [30][180/200]	Time 0.386 (0.436)	Data 0.001 (0.053)	Loss 1.549 (1.209)
Epoch: [30][200/200]	Time 0.381 (0.439)	Data 0.001 (0.056)	Loss 2.130 (1.227)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.178 (0.209)	Data 0.052 (0.080)	
Extract Features: [100/128]	Time 0.125 (0.190)	Data 0.000 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.12007212638855
==> Statistics for epoch 31: 1031 clusters
Epoch: [31][20/200]	Time 0.375 (0.438)	Data 0.001 (0.051)	Loss 0.139 (0.252)
Epoch: [31][40/200]	Time 0.385 (0.448)	Data 0.000 (0.064)	Loss 1.701 (0.451)
Epoch: [31][60/200]	Time 0.374 (0.425)	Data 0.000 (0.043)	Loss 1.316 (0.731)
Epoch: [31][80/200]	Time 0.373 (0.437)	Data 0.001 (0.055)	Loss 1.325 (0.876)
Epoch: [31][100/200]	Time 0.384 (0.442)	Data 0.001 (0.060)	Loss 1.421 (0.996)
Epoch: [31][120/200]	Time 0.383 (0.433)	Data 0.002 (0.050)	Loss 1.474 (1.049)
Epoch: [31][140/200]	Time 0.376 (0.436)	Data 0.001 (0.054)	Loss 1.814 (1.097)
Epoch: [31][160/200]	Time 0.382 (0.430)	Data 0.000 (0.047)	Loss 1.882 (1.135)
Epoch: [31][180/200]	Time 0.388 (0.433)	Data 0.000 (0.050)	Loss 0.890 (1.159)
Epoch: [31][200/200]	Time 0.384 (0.437)	Data 0.000 (0.053)	Loss 1.482 (1.183)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.210)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.177 (0.190)	Data 0.050 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.73006868362427
==> Statistics for epoch 32: 1035 clusters
Epoch: [32][20/200]	Time 0.378 (0.428)	Data 0.001 (0.048)	Loss 0.355 (0.275)
Epoch: [32][40/200]	Time 0.379 (0.443)	Data 0.001 (0.063)	Loss 1.436 (0.484)
Epoch: [32][60/200]	Time 0.381 (0.425)	Data 0.000 (0.042)	Loss 1.174 (0.776)
Epoch: [32][80/200]	Time 0.386 (0.435)	Data 0.000 (0.052)	Loss 1.223 (0.943)
Epoch: [32][100/200]	Time 0.382 (0.440)	Data 0.001 (0.058)	Loss 1.536 (1.035)
Epoch: [32][120/200]	Time 0.394 (0.431)	Data 0.001 (0.048)	Loss 0.880 (1.102)
Epoch: [32][140/200]	Time 0.385 (0.437)	Data 0.001 (0.053)	Loss 1.400 (1.142)
Epoch: [32][160/200]	Time 0.379 (0.430)	Data 0.000 (0.047)	Loss 2.019 (1.173)
Epoch: [32][180/200]	Time 0.496 (0.436)	Data 0.001 (0.051)	Loss 1.550 (1.199)
Epoch: [32][200/200]	Time 0.378 (0.438)	Data 0.001 (0.054)	Loss 1.424 (1.221)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.128 (0.213)	Data 0.000 (0.086)	
Extract Features: [100/128]	Time 0.127 (0.195)	Data 0.000 (0.067)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.67203903198242
==> Statistics for epoch 33: 1027 clusters
Epoch: [33][20/200]	Time 0.385 (0.431)	Data 0.001 (0.050)	Loss 0.236 (0.257)
Epoch: [33][40/200]	Time 0.381 (0.450)	Data 0.001 (0.067)	Loss 1.190 (0.463)
Epoch: [33][60/200]	Time 0.379 (0.427)	Data 0.000 (0.045)	Loss 1.426 (0.770)
Epoch: [33][80/200]	Time 0.392 (0.437)	Data 0.000 (0.055)	Loss 1.630 (0.923)
Epoch: [33][100/200]	Time 0.376 (0.443)	Data 0.001 (0.060)	Loss 1.582 (1.006)
Epoch: [33][120/200]	Time 0.386 (0.433)	Data 0.001 (0.050)	Loss 1.214 (1.057)
Epoch: [33][140/200]	Time 0.390 (0.437)	Data 0.000 (0.054)	Loss 1.510 (1.107)
Epoch: [33][160/200]	Time 0.386 (0.431)	Data 0.000 (0.048)	Loss 1.602 (1.138)
Epoch: [33][180/200]	Time 0.388 (0.434)	Data 0.000 (0.051)	Loss 1.339 (1.158)
Epoch: [33][200/200]	Time 0.383 (0.437)	Data 0.001 (0.054)	Loss 1.623 (1.186)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.213)	Data 0.000 (0.083)	
Extract Features: [100/128]	Time 0.209 (0.192)	Data 0.088 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.288756370544434
==> Statistics for epoch 34: 1039 clusters
Epoch: [34][20/200]	Time 0.381 (0.435)	Data 0.001 (0.055)	Loss 0.341 (0.265)
Epoch: [34][40/200]	Time 0.385 (0.447)	Data 0.001 (0.066)	Loss 1.478 (0.444)
Epoch: [34][60/200]	Time 0.383 (0.425)	Data 0.000 (0.044)	Loss 1.600 (0.728)
Epoch: [34][80/200]	Time 0.377 (0.436)	Data 0.001 (0.054)	Loss 1.250 (0.901)
Epoch: [34][100/200]	Time 0.389 (0.444)	Data 0.001 (0.062)	Loss 1.154 (0.984)
Epoch: [34][120/200]	Time 0.389 (0.434)	Data 0.001 (0.052)	Loss 1.486 (1.059)
Epoch: [34][140/200]	Time 0.382 (0.438)	Data 0.001 (0.055)	Loss 1.407 (1.104)
Epoch: [34][160/200]	Time 0.386 (0.432)	Data 0.000 (0.048)	Loss 1.404 (1.134)
Epoch: [34][180/200]	Time 0.381 (0.435)	Data 0.001 (0.052)	Loss 1.182 (1.160)
Epoch: [34][200/200]	Time 0.379 (0.438)	Data 0.000 (0.055)	Loss 1.619 (1.189)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.125 (0.211)	Data 0.000 (0.084)	
Extract Features: [100/128]	Time 0.126 (0.195)	Data 0.000 (0.067)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.16002106666565
==> Statistics for epoch 35: 1024 clusters
Epoch: [35][20/200]	Time 0.380 (0.431)	Data 0.001 (0.050)	Loss 0.292 (0.250)
Epoch: [35][40/200]	Time 0.380 (0.447)	Data 0.001 (0.066)	Loss 1.017 (0.458)
Epoch: [35][60/200]	Time 0.381 (0.427)	Data 0.000 (0.044)	Loss 1.317 (0.752)
Epoch: [35][80/200]	Time 0.385 (0.438)	Data 0.001 (0.055)	Loss 1.130 (0.892)
Epoch: [35][100/200]	Time 0.380 (0.444)	Data 0.001 (0.061)	Loss 1.898 (0.993)
Epoch: [35][120/200]	Time 0.379 (0.434)	Data 0.001 (0.051)	Loss 1.388 (1.044)
Epoch: [35][140/200]	Time 0.387 (0.438)	Data 0.001 (0.056)	Loss 1.408 (1.093)
Epoch: [35][160/200]	Time 0.385 (0.432)	Data 0.000 (0.049)	Loss 1.122 (1.130)
Epoch: [35][180/200]	Time 0.386 (0.436)	Data 0.001 (0.053)	Loss 1.226 (1.152)
Epoch: [35][200/200]	Time 0.378 (0.440)	Data 0.001 (0.056)	Loss 1.414 (1.167)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.147 (0.210)	Data 0.018 (0.081)	
Extract Features: [100/128]	Time 0.126 (0.191)	Data 0.000 (0.061)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.89921736717224
==> Statistics for epoch 36: 1034 clusters
Epoch: [36][20/200]	Time 0.376 (0.436)	Data 0.001 (0.055)	Loss 0.341 (0.258)
Epoch: [36][40/200]	Time 0.381 (0.451)	Data 0.001 (0.067)	Loss 1.391 (0.463)
Epoch: [36][60/200]	Time 0.381 (0.428)	Data 0.000 (0.045)	Loss 1.082 (0.729)
Epoch: [36][80/200]	Time 0.383 (0.437)	Data 0.001 (0.054)	Loss 1.790 (0.895)
Epoch: [36][100/200]	Time 0.383 (0.442)	Data 0.001 (0.060)	Loss 1.826 (1.001)
Epoch: [36][120/200]	Time 0.381 (0.432)	Data 0.001 (0.050)	Loss 1.501 (1.073)
Epoch: [36][140/200]	Time 0.385 (0.436)	Data 0.000 (0.054)	Loss 1.130 (1.112)
Epoch: [36][160/200]	Time 0.380 (0.430)	Data 0.000 (0.047)	Loss 1.462 (1.145)
Epoch: [36][180/200]	Time 0.387 (0.433)	Data 0.001 (0.050)	Loss 1.150 (1.171)
Epoch: [36][200/200]	Time 0.382 (0.437)	Data 0.000 (0.054)	Loss 1.145 (1.195)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.123 (0.211)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.125 (0.194)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.53987407684326
==> Statistics for epoch 37: 1037 clusters
Epoch: [37][20/200]	Time 0.379 (0.434)	Data 0.001 (0.053)	Loss 0.266 (0.257)
Epoch: [37][40/200]	Time 0.381 (0.446)	Data 0.001 (0.063)	Loss 1.397 (0.501)
Epoch: [37][60/200]	Time 0.379 (0.424)	Data 0.000 (0.042)	Loss 1.493 (0.796)
Epoch: [37][80/200]	Time 0.389 (0.433)	Data 0.001 (0.051)	Loss 1.057 (0.927)
Epoch: [37][100/200]	Time 0.389 (0.440)	Data 0.001 (0.058)	Loss 1.377 (1.020)
Epoch: [37][120/200]	Time 0.381 (0.430)	Data 0.001 (0.049)	Loss 1.941 (1.097)
Epoch: [37][140/200]	Time 0.381 (0.435)	Data 0.001 (0.054)	Loss 1.669 (1.138)
Epoch: [37][160/200]	Time 0.378 (0.429)	Data 0.000 (0.047)	Loss 1.017 (1.166)
Epoch: [37][180/200]	Time 0.380 (0.433)	Data 0.001 (0.051)	Loss 1.073 (1.193)
Epoch: [37][200/200]	Time 0.382 (0.437)	Data 0.001 (0.054)	Loss 1.154 (1.216)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.227 (0.210)	Data 0.100 (0.081)	
Extract Features: [100/128]	Time 0.126 (0.191)	Data 0.000 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.93340039253235
==> Statistics for epoch 38: 1044 clusters
Epoch: [38][20/200]	Time 0.376 (0.439)	Data 0.001 (0.057)	Loss 0.319 (0.269)
Epoch: [38][40/200]	Time 0.383 (0.451)	Data 0.001 (0.069)	Loss 0.787 (0.476)
Epoch: [38][60/200]	Time 0.378 (0.429)	Data 0.000 (0.046)	Loss 1.515 (0.747)
Epoch: [38][80/200]	Time 0.380 (0.436)	Data 0.000 (0.054)	Loss 0.907 (0.891)
Epoch: [38][100/200]	Time 0.384 (0.442)	Data 0.001 (0.060)	Loss 1.122 (0.990)
Epoch: [38][120/200]	Time 0.384 (0.433)	Data 0.001 (0.050)	Loss 1.783 (1.072)
Epoch: [38][140/200]	Time 0.380 (0.438)	Data 0.000 (0.055)	Loss 1.802 (1.126)
Epoch: [38][160/200]	Time 0.381 (0.431)	Data 0.000 (0.048)	Loss 1.079 (1.160)
Epoch: [38][180/200]	Time 0.385 (0.435)	Data 0.001 (0.052)	Loss 1.500 (1.191)
Epoch: [38][200/200]	Time 0.379 (0.437)	Data 0.001 (0.054)	Loss 1.313 (1.213)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.290 (0.211)	Data 0.167 (0.082)	
Extract Features: [100/128]	Time 0.126 (0.193)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.79074263572693
==> Statistics for epoch 39: 1036 clusters
Epoch: [39][20/200]	Time 0.380 (0.434)	Data 0.001 (0.050)	Loss 0.154 (0.235)
Epoch: [39][40/200]	Time 0.382 (0.451)	Data 0.001 (0.068)	Loss 1.356 (0.451)
Epoch: [39][60/200]	Time 0.378 (0.431)	Data 0.000 (0.045)	Loss 1.004 (0.721)
Epoch: [39][80/200]	Time 0.384 (0.440)	Data 0.000 (0.055)	Loss 1.237 (0.866)
Epoch: [39][100/200]	Time 0.389 (0.444)	Data 0.001 (0.060)	Loss 1.699 (0.954)
Epoch: [39][120/200]	Time 0.375 (0.433)	Data 0.001 (0.050)	Loss 1.320 (1.028)
Epoch: [39][140/200]	Time 0.379 (0.437)	Data 0.001 (0.055)	Loss 1.702 (1.076)
Epoch: [39][160/200]	Time 0.375 (0.430)	Data 0.000 (0.048)	Loss 1.428 (1.116)
Epoch: [39][180/200]	Time 0.386 (0.434)	Data 0.001 (0.051)	Loss 1.057 (1.144)
Epoch: [39][200/200]	Time 0.392 (0.437)	Data 0.002 (0.054)	Loss 1.248 (1.170)
Extract Features: [50/367]	Time 0.128 (0.216)	Data 0.000 (0.086)	
Extract Features: [100/367]	Time 0.130 (0.201)	Data 0.001 (0.071)	
Extract Features: [150/367]	Time 0.131 (0.196)	Data 0.000 (0.067)	
Extract Features: [200/367]	Time 0.126 (0.190)	Data 0.000 (0.062)	
Extract Features: [250/367]	Time 0.126 (0.188)	Data 0.000 (0.060)	
Extract Features: [300/367]	Time 0.125 (0.186)	Data 0.000 (0.058)	
Extract Features: [350/367]	Time 0.132 (0.185)	Data 0.000 (0.056)	
Mean AP: 57.8%

 * Finished epoch  39  model mAP: 57.8%  best: 57.8% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.124 (0.215)	Data 0.000 (0.088)	
Extract Features: [100/128]	Time 0.124 (0.193)	Data 0.000 (0.066)	
Computing jaccard distance...
Jaccard distance computing time cost: 57.7646164894104
==> Statistics for epoch 40: 1039 clusters
Epoch: [40][20/200]	Time 0.378 (0.435)	Data 0.001 (0.056)	Loss 0.182 (0.244)
Epoch: [40][40/200]	Time 0.380 (0.446)	Data 0.001 (0.067)	Loss 1.171 (0.465)
Epoch: [40][60/200]	Time 0.383 (0.428)	Data 0.000 (0.045)	Loss 1.095 (0.750)
Epoch: [40][80/200]	Time 0.382 (0.437)	Data 0.001 (0.055)	Loss 1.363 (0.897)
Epoch: [40][100/200]	Time 0.377 (0.442)	Data 0.001 (0.060)	Loss 1.736 (0.985)
Epoch: [40][120/200]	Time 0.384 (0.433)	Data 0.001 (0.050)	Loss 1.262 (1.040)
Epoch: [40][140/200]	Time 0.385 (0.437)	Data 0.001 (0.054)	Loss 0.888 (1.099)
Epoch: [40][160/200]	Time 0.383 (0.430)	Data 0.000 (0.047)	Loss 1.288 (1.140)
Epoch: [40][180/200]	Time 0.373 (0.433)	Data 0.000 (0.051)	Loss 1.312 (1.165)
Epoch: [40][200/200]	Time 0.381 (0.435)	Data 0.001 (0.053)	Loss 1.514 (1.187)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.130 (0.211)	Data 0.000 (0.080)	
Extract Features: [100/128]	Time 0.320 (0.192)	Data 0.194 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 57.68221998214722
==> Statistics for epoch 41: 1030 clusters
Epoch: [41][20/200]	Time 0.379 (0.440)	Data 0.001 (0.055)	Loss 0.147 (0.269)
Epoch: [41][40/200]	Time 0.381 (0.450)	Data 0.001 (0.068)	Loss 1.267 (0.459)
Epoch: [41][60/200]	Time 0.377 (0.427)	Data 0.000 (0.045)	Loss 1.675 (0.746)
Epoch: [41][80/200]	Time 0.373 (0.436)	Data 0.001 (0.055)	Loss 1.419 (0.897)
Epoch: [41][100/200]	Time 0.379 (0.441)	Data 0.001 (0.060)	Loss 1.321 (0.974)
Epoch: [41][120/200]	Time 0.379 (0.431)	Data 0.001 (0.050)	Loss 1.089 (1.044)
Epoch: [41][140/200]	Time 0.383 (0.436)	Data 0.001 (0.054)	Loss 1.026 (1.078)
Epoch: [41][160/200]	Time 0.379 (0.429)	Data 0.000 (0.047)	Loss 1.090 (1.114)
Epoch: [41][180/200]	Time 0.378 (0.432)	Data 0.001 (0.050)	Loss 1.266 (1.141)
Epoch: [41][200/200]	Time 0.389 (0.435)	Data 0.001 (0.053)	Loss 1.235 (1.157)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.128 (0.215)	Data 0.000 (0.089)	
Extract Features: [100/128]	Time 0.125 (0.193)	Data 0.000 (0.065)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.77461123466492
==> Statistics for epoch 42: 1032 clusters
Epoch: [42][20/200]	Time 0.382 (0.428)	Data 0.001 (0.048)	Loss 0.354 (0.251)
Epoch: [42][40/200]	Time 0.380 (0.443)	Data 0.001 (0.062)	Loss 1.503 (0.484)
Epoch: [42][60/200]	Time 0.374 (0.424)	Data 0.000 (0.042)	Loss 1.649 (0.759)
Epoch: [42][80/200]	Time 0.373 (0.433)	Data 0.001 (0.051)	Loss 1.162 (0.897)
Epoch: [42][100/200]	Time 0.381 (0.438)	Data 0.001 (0.056)	Loss 1.509 (0.989)
Epoch: [42][120/200]	Time 0.380 (0.429)	Data 0.001 (0.047)	Loss 0.888 (1.052)
Epoch: [42][140/200]	Time 0.383 (0.434)	Data 0.001 (0.051)	Loss 0.884 (1.097)
Epoch: [42][160/200]	Time 0.494 (0.428)	Data 0.000 (0.045)	Loss 1.190 (1.139)
Epoch: [42][180/200]	Time 0.384 (0.432)	Data 0.001 (0.049)	Loss 1.222 (1.156)
Epoch: [42][200/200]	Time 0.386 (0.436)	Data 0.001 (0.052)	Loss 1.548 (1.182)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.218)	Data 0.000 (0.089)	
Extract Features: [100/128]	Time 0.126 (0.198)	Data 0.000 (0.069)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.02304816246033
==> Statistics for epoch 43: 1030 clusters
Epoch: [43][20/200]	Time 0.379 (0.431)	Data 0.001 (0.049)	Loss 0.242 (0.243)
Epoch: [43][40/200]	Time 0.377 (0.445)	Data 0.001 (0.065)	Loss 1.115 (0.431)
Epoch: [43][60/200]	Time 0.379 (0.423)	Data 0.000 (0.044)	Loss 1.231 (0.733)
Epoch: [43][80/200]	Time 0.387 (0.433)	Data 0.001 (0.054)	Loss 1.412 (0.865)
Epoch: [43][100/200]	Time 0.379 (0.440)	Data 0.001 (0.059)	Loss 1.110 (0.967)
Epoch: [43][120/200]	Time 0.387 (0.431)	Data 0.001 (0.049)	Loss 1.524 (1.018)
Epoch: [43][140/200]	Time 0.383 (0.436)	Data 0.001 (0.054)	Loss 0.878 (1.051)
Epoch: [43][160/200]	Time 0.382 (0.429)	Data 0.000 (0.047)	Loss 1.586 (1.096)
Epoch: [43][180/200]	Time 0.384 (0.433)	Data 0.001 (0.050)	Loss 1.200 (1.122)
Epoch: [43][200/200]	Time 0.387 (0.437)	Data 0.001 (0.054)	Loss 1.700 (1.148)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.211)	Data 0.000 (0.080)	
Extract Features: [100/128]	Time 0.127 (0.194)	Data 0.000 (0.064)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.11856245994568
==> Statistics for epoch 44: 1027 clusters
Epoch: [44][20/200]	Time 0.377 (0.438)	Data 0.001 (0.057)	Loss 0.141 (0.249)
Epoch: [44][40/200]	Time 0.385 (0.447)	Data 0.001 (0.067)	Loss 1.300 (0.422)
Epoch: [44][60/200]	Time 0.384 (0.426)	Data 0.000 (0.045)	Loss 1.444 (0.713)
Epoch: [44][80/200]	Time 0.387 (0.439)	Data 0.001 (0.055)	Loss 1.347 (0.873)
Epoch: [44][100/200]	Time 0.392 (0.445)	Data 0.002 (0.061)	Loss 1.831 (0.959)
Epoch: [44][120/200]	Time 0.386 (0.435)	Data 0.001 (0.051)	Loss 1.269 (1.033)
Epoch: [44][140/200]	Time 0.380 (0.440)	Data 0.001 (0.056)	Loss 1.178 (1.063)
Epoch: [44][160/200]	Time 0.381 (0.433)	Data 0.000 (0.049)	Loss 1.434 (1.093)
Epoch: [44][180/200]	Time 0.384 (0.436)	Data 0.001 (0.052)	Loss 1.575 (1.131)
Epoch: [44][200/200]	Time 0.379 (0.438)	Data 0.001 (0.054)	Loss 1.605 (1.157)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.125 (0.213)	Data 0.000 (0.087)	
Extract Features: [100/128]	Time 0.129 (0.194)	Data 0.000 (0.066)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.79157876968384
==> Statistics for epoch 45: 1029 clusters
Epoch: [45][20/200]	Time 0.385 (0.434)	Data 0.001 (0.054)	Loss 0.264 (0.228)
Epoch: [45][40/200]	Time 0.386 (0.451)	Data 0.001 (0.067)	Loss 1.470 (0.444)
Epoch: [45][60/200]	Time 0.378 (0.427)	Data 0.000 (0.045)	Loss 1.476 (0.731)
Epoch: [45][80/200]	Time 0.383 (0.435)	Data 0.000 (0.053)	Loss 1.182 (0.868)
Epoch: [45][100/200]	Time 0.395 (0.442)	Data 0.001 (0.058)	Loss 1.351 (0.954)
Epoch: [45][120/200]	Time 0.382 (0.432)	Data 0.000 (0.048)	Loss 1.356 (1.024)
Epoch: [45][140/200]	Time 0.379 (0.436)	Data 0.001 (0.052)	Loss 1.761 (1.075)
Epoch: [45][160/200]	Time 0.381 (0.429)	Data 0.000 (0.046)	Loss 1.311 (1.105)
Epoch: [45][180/200]	Time 0.382 (0.434)	Data 0.001 (0.050)	Loss 1.657 (1.132)
Epoch: [45][200/200]	Time 0.387 (0.436)	Data 0.000 (0.053)	Loss 1.117 (1.158)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.307 (0.208)	Data 0.181 (0.079)	
Extract Features: [100/128]	Time 0.126 (0.190)	Data 0.000 (0.061)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.02382946014404
==> Statistics for epoch 46: 1029 clusters
Epoch: [46][20/200]	Time 0.384 (0.434)	Data 0.001 (0.050)	Loss 0.172 (0.246)
Epoch: [46][40/200]	Time 0.384 (0.450)	Data 0.001 (0.067)	Loss 1.587 (0.486)
Epoch: [46][60/200]	Time 0.379 (0.428)	Data 0.000 (0.045)	Loss 1.341 (0.758)
Epoch: [46][80/200]	Time 0.384 (0.438)	Data 0.000 (0.056)	Loss 1.252 (0.905)
Epoch: [46][100/200]	Time 0.386 (0.444)	Data 0.001 (0.062)	Loss 1.396 (0.993)
Epoch: [46][120/200]	Time 0.385 (0.434)	Data 0.001 (0.051)	Loss 1.579 (1.054)
Epoch: [46][140/200]	Time 0.385 (0.440)	Data 0.001 (0.056)	Loss 1.068 (1.096)
Epoch: [46][160/200]	Time 0.386 (0.433)	Data 0.000 (0.049)	Loss 1.452 (1.130)
Epoch: [46][180/200]	Time 0.377 (0.436)	Data 0.000 (0.052)	Loss 0.943 (1.157)
Epoch: [46][200/200]	Time 0.385 (0.439)	Data 0.000 (0.055)	Loss 1.414 (1.178)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.127 (0.211)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.127 (0.191)	Data 0.000 (0.063)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.915663957595825
==> Statistics for epoch 47: 1034 clusters
Epoch: [47][20/200]	Time 0.381 (0.429)	Data 0.001 (0.048)	Loss 0.145 (0.255)
Epoch: [47][40/200]	Time 0.382 (0.450)	Data 0.001 (0.069)	Loss 1.635 (0.476)
Epoch: [47][60/200]	Time 0.373 (0.428)	Data 0.000 (0.046)	Loss 1.440 (0.762)
Epoch: [47][80/200]	Time 0.393 (0.437)	Data 0.001 (0.055)	Loss 1.329 (0.911)
Epoch: [47][100/200]	Time 0.389 (0.443)	Data 0.001 (0.060)	Loss 1.691 (1.000)
Epoch: [47][120/200]	Time 0.387 (0.433)	Data 0.001 (0.050)	Loss 1.178 (1.054)
Epoch: [47][140/200]	Time 0.381 (0.438)	Data 0.001 (0.055)	Loss 1.145 (1.097)
Epoch: [47][160/200]	Time 0.375 (0.431)	Data 0.000 (0.049)	Loss 1.396 (1.133)
Epoch: [47][180/200]	Time 0.381 (0.435)	Data 0.001 (0.052)	Loss 1.133 (1.150)
Epoch: [47][200/200]	Time 0.382 (0.438)	Data 0.001 (0.055)	Loss 1.943 (1.170)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.126 (0.210)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.134 (0.190)	Data 0.009 (0.061)	
Computing jaccard distance...
Jaccard distance computing time cost: 57.48074531555176
==> Statistics for epoch 48: 1033 clusters
Epoch: [48][20/200]	Time 0.380 (0.433)	Data 0.001 (0.052)	Loss 0.306 (0.244)
Epoch: [48][40/200]	Time 0.379 (0.446)	Data 0.001 (0.062)	Loss 1.284 (0.488)
Epoch: [48][60/200]	Time 0.389 (0.424)	Data 0.000 (0.042)	Loss 1.445 (0.746)
Epoch: [48][80/200]	Time 0.381 (0.434)	Data 0.001 (0.052)	Loss 1.194 (0.901)
Epoch: [48][100/200]	Time 0.382 (0.439)	Data 0.001 (0.057)	Loss 0.932 (1.004)
Epoch: [48][120/200]	Time 0.381 (0.430)	Data 0.001 (0.048)	Loss 1.281 (1.055)
Epoch: [48][140/200]	Time 0.375 (0.434)	Data 0.001 (0.052)	Loss 0.938 (1.107)
Epoch: [48][160/200]	Time 0.372 (0.427)	Data 0.000 (0.045)	Loss 1.062 (1.136)
Epoch: [48][180/200]	Time 0.386 (0.431)	Data 0.001 (0.049)	Loss 1.247 (1.159)
Epoch: [48][200/200]	Time 0.391 (0.434)	Data 0.000 (0.052)	Loss 1.783 (1.179)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.129 (0.210)	Data 0.000 (0.081)	
Extract Features: [100/128]	Time 0.126 (0.191)	Data 0.000 (0.062)	
Computing jaccard distance...
Jaccard distance computing time cost: 57.82575845718384
==> Statistics for epoch 49: 1029 clusters
Epoch: [49][20/200]	Time 0.387 (0.430)	Data 0.001 (0.051)	Loss 0.108 (0.243)
Epoch: [49][40/200]	Time 0.383 (0.445)	Data 0.001 (0.067)	Loss 1.800 (0.442)
Epoch: [49][60/200]	Time 0.377 (0.425)	Data 0.000 (0.045)	Loss 1.288 (0.727)
Epoch: [49][80/200]	Time 0.378 (0.434)	Data 0.001 (0.053)	Loss 0.941 (0.895)
Epoch: [49][100/200]	Time 0.383 (0.439)	Data 0.001 (0.058)	Loss 2.033 (0.987)
Epoch: [49][120/200]	Time 0.372 (0.430)	Data 0.001 (0.048)	Loss 1.766 (1.051)
Epoch: [49][140/200]	Time 0.378 (0.436)	Data 0.001 (0.053)	Loss 1.144 (1.088)
Epoch: [49][160/200]	Time 0.388 (0.429)	Data 0.000 (0.047)	Loss 1.351 (1.122)
Epoch: [49][180/200]	Time 0.390 (0.433)	Data 0.001 (0.051)	Loss 1.005 (1.144)
Epoch: [49][200/200]	Time 0.387 (0.435)	Data 0.001 (0.053)	Loss 1.657 (1.168)
Extract Features: [50/367]	Time 0.126 (0.215)	Data 0.000 (0.086)	
Extract Features: [100/367]	Time 0.250 (0.201)	Data 0.124 (0.071)	
Extract Features: [150/367]	Time 0.129 (0.196)	Data 0.000 (0.068)	
Extract Features: [200/367]	Time 0.124 (0.193)	Data 0.000 (0.065)	
Extract Features: [250/367]	Time 0.127 (0.191)	Data 0.000 (0.063)	
Extract Features: [300/367]	Time 0.127 (0.188)	Data 0.000 (0.061)	
Extract Features: [350/367]	Time 0.127 (0.187)	Data 0.000 (0.059)	
Mean AP: 58.0%

 * Finished epoch  49  model mAP: 58.0%  best: 58.0% *

==> Test with the best model:
=> Loaded checkpoint '.log/market2msmt/vit_small_cion/model_best.pth.tar'
Extract Features: [50/367]	Time 0.124 (0.215)	Data 0.000 (0.090)	
Extract Features: [100/367]	Time 0.125 (0.197)	Data 0.000 (0.071)	
Extract Features: [150/367]	Time 0.128 (0.193)	Data 0.000 (0.066)	
Extract Features: [200/367]	Time 0.126 (0.189)	Data 0.000 (0.063)	
Extract Features: [250/367]	Time 0.127 (0.188)	Data 0.000 (0.061)	
Extract Features: [300/367]	Time 0.126 (0.186)	Data 0.000 (0.059)	
Extract Features: [350/367]	Time 0.132 (0.185)	Data 0.000 (0.058)	
Mean AP: 58.0%
CMC Scores:
  top-1          79.9%
  top-5          88.2%
  top-10         90.5%
Total running time:  2:59:13.100725
