==========
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_tiny', pretrained_path='/home/ma-user/work/Projects/ReIDNet_Finetune/TransReID/log/bs_exp/ViT_Tiny_Market1501/64bs_lr0.0004_ep120_warm20_seed0/vit_tiny_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_tiny_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.
ViT Tiny Created!
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.502 (0.347)	Data 0.433 (0.122)	
Extract Features: [100/128]	Time 0.076 (0.259)	Data 0.000 (0.110)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.25541424751282
Clustering criterion: eps: 0.700
==> Statistics for epoch 0: 649 clusters
Epoch: [0][20/200]	Time 0.201 (0.740)	Data 0.000 (0.046)	Loss 7.080 (7.480)
Epoch: [0][40/200]	Time 0.202 (0.506)	Data 0.000 (0.056)	Loss 6.670 (7.191)
Epoch: [0][60/200]	Time 0.203 (0.430)	Data 0.000 (0.062)	Loss 4.260 (6.460)
Epoch: [0][80/200]	Time 0.202 (0.393)	Data 0.000 (0.066)	Loss 4.400 (5.871)
Epoch: [0][100/200]	Time 0.205 (0.372)	Data 0.000 (0.067)	Loss 3.557 (5.426)
Epoch: [0][120/200]	Time 0.204 (0.357)	Data 0.000 (0.068)	Loss 3.325 (5.072)
Epoch: [0][140/200]	Time 0.203 (0.346)	Data 0.000 (0.069)	Loss 3.335 (4.807)
Epoch: [0][160/200]	Time 0.202 (0.338)	Data 0.000 (0.069)	Loss 3.990 (4.607)
Epoch: [0][180/200]	Time 0.200 (0.332)	Data 0.000 (0.070)	Loss 2.528 (4.411)
Epoch: [0][200/200]	Time 0.201 (0.326)	Data 0.000 (0.070)	Loss 2.265 (4.240)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.208)	Data 0.000 (0.138)	
Extract Features: [100/128]	Time 0.069 (0.188)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.07098913192749
==> Statistics for epoch 1: 817 clusters
Epoch: [1][20/200]	Time 0.204 (0.260)	Data 0.001 (0.053)	Loss 0.610 (0.751)
Epoch: [1][40/200]	Time 0.201 (0.276)	Data 0.001 (0.070)	Loss 2.297 (1.420)
Epoch: [1][60/200]	Time 0.205 (0.280)	Data 0.001 (0.074)	Loss 2.182 (1.837)
Epoch: [1][80/200]	Time 0.241 (0.279)	Data 0.039 (0.074)	Loss 2.069 (2.006)
Epoch: [1][100/200]	Time 0.201 (0.264)	Data 0.000 (0.059)	Loss 2.296 (2.117)
Epoch: [1][120/200]	Time 0.210 (0.267)	Data 0.000 (0.061)	Loss 2.578 (2.154)
Epoch: [1][140/200]	Time 0.210 (0.269)	Data 0.000 (0.063)	Loss 2.483 (2.187)
Epoch: [1][160/200]	Time 0.205 (0.270)	Data 0.000 (0.064)	Loss 2.648 (2.204)
Epoch: [1][180/200]	Time 0.206 (0.271)	Data 0.000 (0.065)	Loss 2.291 (2.212)
Epoch: [1][200/200]	Time 0.204 (0.265)	Data 0.000 (0.058)	Loss 2.887 (2.218)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.577 (0.209)	Data 0.507 (0.136)	
Extract Features: [100/128]	Time 0.070 (0.189)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.128010511398315
==> Statistics for epoch 2: 884 clusters
Epoch: [2][20/200]	Time 0.219 (0.265)	Data 0.000 (0.051)	Loss 0.825 (0.613)
Epoch: [2][40/200]	Time 0.248 (0.270)	Data 0.048 (0.061)	Loss 2.645 (1.137)
Epoch: [2][60/200]	Time 0.205 (0.274)	Data 0.000 (0.067)	Loss 2.585 (1.547)
Epoch: [2][80/200]	Time 0.208 (0.257)	Data 0.000 (0.051)	Loss 2.105 (1.746)
Epoch: [2][100/200]	Time 0.206 (0.264)	Data 0.000 (0.056)	Loss 1.692 (1.873)
Epoch: [2][120/200]	Time 0.212 (0.266)	Data 0.000 (0.058)	Loss 2.047 (1.935)
Epoch: [2][140/200]	Time 0.245 (0.269)	Data 0.043 (0.061)	Loss 1.740 (1.971)
Epoch: [2][160/200]	Time 0.202 (0.263)	Data 0.000 (0.054)	Loss 2.295 (2.012)
Epoch: [2][180/200]	Time 0.205 (0.264)	Data 0.001 (0.056)	Loss 2.727 (2.031)
Epoch: [2][200/200]	Time 0.204 (0.266)	Data 0.000 (0.057)	Loss 1.915 (2.044)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.210)	Data 0.000 (0.140)	
Extract Features: [100/128]	Time 0.067 (0.189)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.510698080062866
==> Statistics for epoch 3: 902 clusters
Epoch: [3][20/200]	Time 0.206 (0.258)	Data 0.001 (0.052)	Loss 0.494 (0.556)
Epoch: [3][40/200]	Time 0.208 (0.270)	Data 0.001 (0.065)	Loss 1.735 (0.988)
Epoch: [3][60/200]	Time 0.209 (0.276)	Data 0.001 (0.068)	Loss 1.989 (1.349)
Epoch: [3][80/200]	Time 0.212 (0.260)	Data 0.000 (0.052)	Loss 2.346 (1.527)
Epoch: [3][100/200]	Time 0.205 (0.265)	Data 0.001 (0.057)	Loss 1.880 (1.647)
Epoch: [3][120/200]	Time 0.217 (0.269)	Data 0.001 (0.060)	Loss 2.226 (1.718)
Epoch: [3][140/200]	Time 0.201 (0.260)	Data 0.000 (0.051)	Loss 1.950 (1.773)
Epoch: [3][160/200]	Time 0.205 (0.262)	Data 0.000 (0.053)	Loss 2.065 (1.805)
Epoch: [3][180/200]	Time 0.204 (0.263)	Data 0.000 (0.055)	Loss 2.320 (1.827)
Epoch: [3][200/200]	Time 0.204 (0.266)	Data 0.001 (0.058)	Loss 2.601 (1.851)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.221 (0.202)	Data 0.008 (0.129)	
Extract Features: [100/128]	Time 0.296 (0.185)	Data 0.227 (0.115)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.72563648223877
==> Statistics for epoch 4: 872 clusters
Epoch: [4][20/200]	Time 0.205 (0.264)	Data 0.000 (0.051)	Loss 0.296 (0.474)
Epoch: [4][40/200]	Time 0.204 (0.270)	Data 0.001 (0.062)	Loss 2.255 (0.946)
Epoch: [4][60/200]	Time 0.204 (0.277)	Data 0.001 (0.070)	Loss 2.376 (1.318)
Epoch: [4][80/200]	Time 0.202 (0.260)	Data 0.000 (0.054)	Loss 1.943 (1.495)
Epoch: [4][100/200]	Time 0.205 (0.264)	Data 0.001 (0.058)	Loss 1.541 (1.601)
Epoch: [4][120/200]	Time 0.207 (0.267)	Data 0.001 (0.061)	Loss 2.449 (1.667)
Epoch: [4][140/200]	Time 0.216 (0.268)	Data 0.001 (0.062)	Loss 1.410 (1.710)
Epoch: [4][160/200]	Time 0.205 (0.261)	Data 0.000 (0.055)	Loss 2.294 (1.737)
Epoch: [4][180/200]	Time 0.206 (0.264)	Data 0.001 (0.057)	Loss 2.072 (1.768)
Epoch: [4][200/200]	Time 0.208 (0.267)	Data 0.001 (0.060)	Loss 1.613 (1.771)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.248 (0.205)	Data 0.179 (0.132)	
Extract Features: [100/128]	Time 0.069 (0.188)	Data 0.000 (0.116)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.04068064689636
==> Statistics for epoch 5: 931 clusters
Epoch: [5][20/200]	Time 0.207 (0.269)	Data 0.001 (0.062)	Loss 0.495 (0.500)
Epoch: [5][40/200]	Time 0.206 (0.278)	Data 0.001 (0.072)	Loss 1.954 (0.921)
Epoch: [5][60/200]	Time 0.204 (0.279)	Data 0.001 (0.072)	Loss 1.883 (1.328)
Epoch: [5][80/200]	Time 0.204 (0.263)	Data 0.000 (0.054)	Loss 1.696 (1.519)
Epoch: [5][100/200]	Time 0.290 (0.267)	Data 0.085 (0.059)	Loss 1.812 (1.642)
Epoch: [5][120/200]	Time 0.210 (0.269)	Data 0.002 (0.061)	Loss 2.192 (1.720)
Epoch: [5][140/200]	Time 0.208 (0.260)	Data 0.000 (0.053)	Loss 1.936 (1.763)
Epoch: [5][160/200]	Time 0.204 (0.264)	Data 0.001 (0.056)	Loss 2.459 (1.812)
Epoch: [5][180/200]	Time 0.203 (0.266)	Data 0.001 (0.059)	Loss 1.981 (1.833)
Epoch: [5][200/200]	Time 0.205 (0.262)	Data 0.000 (0.055)	Loss 1.921 (1.854)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.096 (0.202)	Data 0.026 (0.132)	
Extract Features: [100/128]	Time 0.070 (0.190)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.10828995704651
==> Statistics for epoch 6: 921 clusters
Epoch: [6][20/200]	Time 0.206 (0.264)	Data 0.001 (0.049)	Loss 0.387 (0.443)
Epoch: [6][40/200]	Time 0.206 (0.273)	Data 0.001 (0.063)	Loss 1.887 (0.874)
Epoch: [6][60/200]	Time 0.207 (0.276)	Data 0.000 (0.067)	Loss 1.721 (1.211)
Epoch: [6][80/200]	Time 0.210 (0.259)	Data 0.000 (0.052)	Loss 2.082 (1.393)
Epoch: [6][100/200]	Time 0.205 (0.264)	Data 0.001 (0.056)	Loss 1.697 (1.529)
Epoch: [6][120/200]	Time 0.208 (0.266)	Data 0.001 (0.059)	Loss 1.460 (1.592)
Epoch: [6][140/200]	Time 0.204 (0.259)	Data 0.000 (0.052)	Loss 1.766 (1.640)
Epoch: [6][160/200]	Time 0.204 (0.263)	Data 0.001 (0.056)	Loss 1.417 (1.672)
Epoch: [6][180/200]	Time 0.204 (0.266)	Data 0.001 (0.059)	Loss 2.021 (1.703)
Epoch: [6][200/200]	Time 0.210 (0.267)	Data 0.001 (0.060)	Loss 1.905 (1.719)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.368 (0.211)	Data 0.300 (0.138)	
Extract Features: [100/128]	Time 0.069 (0.192)	Data 0.000 (0.119)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.56925129890442
==> Statistics for epoch 7: 927 clusters
Epoch: [7][20/200]	Time 0.214 (0.260)	Data 0.001 (0.048)	Loss 0.369 (0.455)
Epoch: [7][40/200]	Time 0.211 (0.275)	Data 0.001 (0.065)	Loss 1.749 (0.851)
Epoch: [7][60/200]	Time 0.208 (0.279)	Data 0.001 (0.068)	Loss 2.316 (1.188)
Epoch: [7][80/200]	Time 0.204 (0.262)	Data 0.000 (0.053)	Loss 2.229 (1.387)
Epoch: [7][100/200]	Time 0.205 (0.264)	Data 0.001 (0.056)	Loss 1.993 (1.496)
Epoch: [7][120/200]	Time 0.205 (0.268)	Data 0.001 (0.059)	Loss 1.975 (1.571)
Epoch: [7][140/200]	Time 0.202 (0.259)	Data 0.000 (0.051)	Loss 1.893 (1.624)
Epoch: [7][160/200]	Time 0.206 (0.262)	Data 0.001 (0.054)	Loss 1.434 (1.652)
Epoch: [7][180/200]	Time 0.209 (0.264)	Data 0.001 (0.056)	Loss 1.551 (1.691)
Epoch: [7][200/200]	Time 0.205 (0.266)	Data 0.001 (0.058)	Loss 1.725 (1.714)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.213)	Data 0.000 (0.144)	
Extract Features: [100/128]	Time 0.069 (0.192)	Data 0.000 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.70527148246765
==> Statistics for epoch 8: 887 clusters
Epoch: [8][20/200]	Time 0.203 (0.269)	Data 0.001 (0.057)	Loss 0.529 (0.455)
Epoch: [8][40/200]	Time 0.219 (0.274)	Data 0.017 (0.065)	Loss 1.351 (0.888)
Epoch: [8][60/200]	Time 0.203 (0.281)	Data 0.001 (0.074)	Loss 1.812 (1.195)
Epoch: [8][80/200]	Time 0.202 (0.264)	Data 0.000 (0.057)	Loss 0.935 (1.344)
Epoch: [8][100/200]	Time 0.206 (0.268)	Data 0.001 (0.061)	Loss 2.080 (1.468)
Epoch: [8][120/200]	Time 0.205 (0.270)	Data 0.001 (0.063)	Loss 1.860 (1.533)
Epoch: [8][140/200]	Time 0.206 (0.274)	Data 0.000 (0.067)	Loss 2.023 (1.589)
Epoch: [8][160/200]	Time 0.202 (0.266)	Data 0.000 (0.059)	Loss 1.580 (1.626)
Epoch: [8][180/200]	Time 0.205 (0.267)	Data 0.001 (0.061)	Loss 2.165 (1.654)
Epoch: [8][200/200]	Time 0.206 (0.270)	Data 0.001 (0.063)	Loss 2.892 (1.673)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.164 (0.203)	Data 0.094 (0.130)	
Extract Features: [100/128]	Time 0.068 (0.186)	Data 0.000 (0.115)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.5361430644989
==> Statistics for epoch 9: 932 clusters
Epoch: [9][20/200]	Time 0.212 (0.263)	Data 0.001 (0.055)	Loss 0.257 (0.449)
Epoch: [9][40/200]	Time 0.204 (0.275)	Data 0.001 (0.068)	Loss 2.008 (0.818)
Epoch: [9][60/200]	Time 0.208 (0.281)	Data 0.001 (0.074)	Loss 2.611 (1.206)
Epoch: [9][80/200]	Time 0.258 (0.265)	Data 0.056 (0.059)	Loss 1.894 (1.393)
Epoch: [9][100/200]	Time 0.313 (0.272)	Data 0.110 (0.065)	Loss 1.234 (1.479)
Epoch: [9][120/200]	Time 0.206 (0.275)	Data 0.001 (0.068)	Loss 1.708 (1.553)
Epoch: [9][140/200]	Time 0.209 (0.266)	Data 0.000 (0.059)	Loss 2.130 (1.593)
Epoch: [9][160/200]	Time 0.205 (0.269)	Data 0.000 (0.062)	Loss 1.816 (1.638)
Epoch: [9][180/200]	Time 0.205 (0.270)	Data 0.000 (0.063)	Loss 1.796 (1.667)
Epoch: [9][200/200]	Time 0.208 (0.265)	Data 0.000 (0.057)	Loss 1.648 (1.681)
Extract Features: [50/367]	Time 0.400 (0.213)	Data 0.184 (0.140)	
Extract Features: [100/367]	Time 0.068 (0.194)	Data 0.000 (0.122)	
Extract Features: [150/367]	Time 0.326 (0.191)	Data 0.256 (0.119)	
Extract Features: [200/367]	Time 0.070 (0.188)	Data 0.000 (0.116)	
Extract Features: [250/367]	Time 0.506 (0.187)	Data 0.435 (0.115)	
Extract Features: [300/367]	Time 0.070 (0.184)	Data 0.000 (0.113)	
Extract Features: [350/367]	Time 0.624 (0.185)	Data 0.550 (0.113)	
Mean AP: 29.5%

 * Finished epoch   9  model mAP: 29.5%  best: 29.5% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.237 (0.205)	Data 0.169 (0.135)	
Extract Features: [100/128]	Time 0.068 (0.190)	Data 0.000 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.912227392196655
==> Statistics for epoch 10: 894 clusters
Epoch: [10][20/200]	Time 0.204 (0.289)	Data 0.001 (0.056)	Loss 0.363 (0.448)
Epoch: [10][40/200]	Time 0.204 (0.293)	Data 0.001 (0.074)	Loss 1.556 (0.876)
Epoch: [10][60/200]	Time 0.204 (0.290)	Data 0.001 (0.076)	Loss 1.400 (1.198)
Epoch: [10][80/200]	Time 0.205 (0.269)	Data 0.000 (0.057)	Loss 1.506 (1.363)
Epoch: [10][100/200]	Time 0.207 (0.276)	Data 0.001 (0.065)	Loss 1.838 (1.452)
Epoch: [10][120/200]	Time 0.205 (0.279)	Data 0.001 (0.069)	Loss 2.363 (1.537)
Epoch: [10][140/200]	Time 0.207 (0.281)	Data 0.001 (0.072)	Loss 1.714 (1.577)
Epoch: [10][160/200]	Time 0.206 (0.273)	Data 0.000 (0.063)	Loss 1.509 (1.615)
Epoch: [10][180/200]	Time 0.255 (0.275)	Data 0.049 (0.066)	Loss 2.142 (1.639)
Epoch: [10][200/200]	Time 0.206 (0.278)	Data 0.001 (0.068)	Loss 1.622 (1.667)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.258 (0.207)	Data 0.189 (0.134)	
Extract Features: [100/128]	Time 0.070 (0.187)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.534663915634155
==> Statistics for epoch 11: 902 clusters
Epoch: [11][20/200]	Time 0.204 (0.258)	Data 0.001 (0.051)	Loss 0.425 (0.469)
Epoch: [11][40/200]	Time 0.203 (0.275)	Data 0.001 (0.065)	Loss 1.540 (0.853)
Epoch: [11][60/200]	Time 0.204 (0.279)	Data 0.000 (0.071)	Loss 1.617 (1.195)
Epoch: [11][80/200]	Time 0.210 (0.261)	Data 0.000 (0.054)	Loss 2.167 (1.367)
Epoch: [11][100/200]	Time 0.210 (0.267)	Data 0.004 (0.059)	Loss 2.383 (1.461)
Epoch: [11][120/200]	Time 0.204 (0.272)	Data 0.000 (0.063)	Loss 2.071 (1.544)
Epoch: [11][140/200]	Time 0.203 (0.262)	Data 0.000 (0.054)	Loss 2.011 (1.587)
Epoch: [11][160/200]	Time 0.207 (0.267)	Data 0.001 (0.059)	Loss 1.686 (1.623)
Epoch: [11][180/200]	Time 0.209 (0.269)	Data 0.001 (0.061)	Loss 1.675 (1.657)
Epoch: [11][200/200]	Time 0.207 (0.272)	Data 0.000 (0.063)	Loss 1.504 (1.675)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.251 (0.209)	Data 0.179 (0.140)	
Extract Features: [100/128]	Time 0.066 (0.188)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.061288356781006
==> Statistics for epoch 12: 900 clusters
Epoch: [12][20/200]	Time 0.207 (0.268)	Data 0.001 (0.061)	Loss 0.361 (0.424)
Epoch: [12][40/200]	Time 0.207 (0.280)	Data 0.001 (0.074)	Loss 1.571 (0.844)
Epoch: [12][60/200]	Time 0.207 (0.284)	Data 0.001 (0.078)	Loss 1.982 (1.202)
Epoch: [12][80/200]	Time 0.207 (0.267)	Data 0.000 (0.060)	Loss 2.101 (1.341)
Epoch: [12][100/200]	Time 0.207 (0.273)	Data 0.000 (0.066)	Loss 2.040 (1.423)
Epoch: [12][120/200]	Time 0.206 (0.276)	Data 0.001 (0.070)	Loss 2.094 (1.516)
Epoch: [12][140/200]	Time 0.205 (0.267)	Data 0.000 (0.060)	Loss 2.392 (1.561)
Epoch: [12][160/200]	Time 0.205 (0.271)	Data 0.001 (0.064)	Loss 2.010 (1.602)
Epoch: [12][180/200]	Time 0.209 (0.274)	Data 0.001 (0.067)	Loss 1.592 (1.636)
Epoch: [12][200/200]	Time 0.211 (0.277)	Data 0.001 (0.069)	Loss 2.024 (1.654)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.211)	Data 0.000 (0.142)	
Extract Features: [100/128]	Time 0.067 (0.190)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.7850124835968
==> Statistics for epoch 13: 904 clusters
Epoch: [13][20/200]	Time 0.208 (0.270)	Data 0.001 (0.062)	Loss 0.306 (0.410)
Epoch: [13][40/200]	Time 0.216 (0.279)	Data 0.001 (0.073)	Loss 1.383 (0.763)
Epoch: [13][60/200]	Time 0.213 (0.290)	Data 0.001 (0.081)	Loss 1.291 (1.105)
Epoch: [13][80/200]	Time 0.206 (0.271)	Data 0.000 (0.063)	Loss 1.657 (1.275)
Epoch: [13][100/200]	Time 0.205 (0.275)	Data 0.001 (0.067)	Loss 1.882 (1.398)
Epoch: [13][120/200]	Time 0.217 (0.279)	Data 0.001 (0.071)	Loss 1.701 (1.471)
Epoch: [13][140/200]	Time 0.202 (0.269)	Data 0.000 (0.062)	Loss 1.864 (1.520)
Epoch: [13][160/200]	Time 0.208 (0.271)	Data 0.001 (0.063)	Loss 1.560 (1.554)
Epoch: [13][180/200]	Time 0.209 (0.273)	Data 0.001 (0.065)	Loss 1.554 (1.578)
Epoch: [13][200/200]	Time 0.210 (0.275)	Data 0.001 (0.067)	Loss 1.765 (1.603)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.535 (0.213)	Data 0.466 (0.140)	
Extract Features: [100/128]	Time 0.068 (0.191)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.06568455696106
==> Statistics for epoch 14: 910 clusters
Epoch: [14][20/200]	Time 0.214 (0.271)	Data 0.001 (0.054)	Loss 0.363 (0.374)
Epoch: [14][40/200]	Time 0.223 (0.282)	Data 0.001 (0.070)	Loss 2.095 (0.808)
Epoch: [14][60/200]	Time 0.205 (0.285)	Data 0.000 (0.074)	Loss 2.173 (1.139)
Epoch: [14][80/200]	Time 0.206 (0.268)	Data 0.000 (0.058)	Loss 1.673 (1.296)
Epoch: [14][100/200]	Time 0.207 (0.274)	Data 0.001 (0.064)	Loss 1.408 (1.397)
Epoch: [14][120/200]	Time 0.207 (0.276)	Data 0.000 (0.067)	Loss 1.928 (1.461)
Epoch: [14][140/200]	Time 0.203 (0.266)	Data 0.000 (0.058)	Loss 2.158 (1.505)
Epoch: [14][160/200]	Time 0.210 (0.270)	Data 0.000 (0.061)	Loss 2.006 (1.534)
Epoch: [14][180/200]	Time 0.205 (0.272)	Data 0.001 (0.063)	Loss 2.324 (1.572)
Epoch: [14][200/200]	Time 0.214 (0.274)	Data 0.001 (0.066)	Loss 1.625 (1.590)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.206)	Data 0.000 (0.134)	
Extract Features: [100/128]	Time 0.177 (0.188)	Data 0.109 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.12920784950256
==> Statistics for epoch 15: 916 clusters
Epoch: [15][20/200]	Time 0.209 (0.265)	Data 0.001 (0.059)	Loss 0.435 (0.391)
Epoch: [15][40/200]	Time 0.205 (0.278)	Data 0.001 (0.072)	Loss 1.813 (0.820)
Epoch: [15][60/200]	Time 0.212 (0.278)	Data 0.001 (0.072)	Loss 1.632 (1.160)
Epoch: [15][80/200]	Time 0.214 (0.262)	Data 0.000 (0.056)	Loss 2.234 (1.312)
Epoch: [15][100/200]	Time 0.210 (0.269)	Data 0.001 (0.062)	Loss 2.148 (1.410)
Epoch: [15][120/200]	Time 0.213 (0.274)	Data 0.001 (0.067)	Loss 1.981 (1.446)
Epoch: [15][140/200]	Time 0.201 (0.266)	Data 0.000 (0.059)	Loss 1.412 (1.507)
Epoch: [15][160/200]	Time 0.350 (0.270)	Data 0.001 (0.063)	Loss 1.785 (1.535)
Epoch: [15][180/200]	Time 0.226 (0.272)	Data 0.000 (0.064)	Loss 2.030 (1.568)
Epoch: [15][200/200]	Time 0.229 (0.273)	Data 0.001 (0.066)	Loss 2.509 (1.598)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.211)	Data 0.000 (0.141)	
Extract Features: [100/128]	Time 0.068 (0.193)	Data 0.000 (0.122)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.1605806350708
==> Statistics for epoch 16: 941 clusters
Epoch: [16][20/200]	Time 0.207 (0.261)	Data 0.001 (0.055)	Loss 0.365 (0.402)
Epoch: [16][40/200]	Time 0.205 (0.277)	Data 0.001 (0.071)	Loss 1.515 (0.757)
Epoch: [16][60/200]	Time 0.202 (0.284)	Data 0.001 (0.076)	Loss 1.766 (1.131)
Epoch: [16][80/200]	Time 0.206 (0.265)	Data 0.001 (0.057)	Loss 1.798 (1.313)
Epoch: [16][100/200]	Time 0.214 (0.271)	Data 0.001 (0.064)	Loss 2.189 (1.419)
Epoch: [16][120/200]	Time 0.204 (0.275)	Data 0.000 (0.067)	Loss 1.902 (1.492)
Epoch: [16][140/200]	Time 0.209 (0.266)	Data 0.000 (0.058)	Loss 1.874 (1.545)
Epoch: [16][160/200]	Time 0.204 (0.270)	Data 0.001 (0.062)	Loss 1.653 (1.592)
Epoch: [16][180/200]	Time 0.202 (0.272)	Data 0.000 (0.064)	Loss 2.232 (1.606)
Epoch: [16][200/200]	Time 0.289 (0.266)	Data 0.078 (0.059)	Loss 1.618 (1.632)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.208)	Data 0.000 (0.139)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.789894580841064
==> Statistics for epoch 17: 905 clusters
Epoch: [17][20/200]	Time 0.372 (0.279)	Data 0.001 (0.066)	Loss 0.461 (0.404)
Epoch: [17][40/200]	Time 0.205 (0.282)	Data 0.001 (0.072)	Loss 2.160 (0.773)
Epoch: [17][60/200]	Time 0.202 (0.283)	Data 0.001 (0.074)	Loss 1.273 (1.110)
Epoch: [17][80/200]	Time 0.210 (0.266)	Data 0.000 (0.057)	Loss 1.737 (1.272)
Epoch: [17][100/200]	Time 0.207 (0.273)	Data 0.001 (0.063)	Loss 1.862 (1.367)
Epoch: [17][120/200]	Time 0.208 (0.275)	Data 0.001 (0.065)	Loss 2.158 (1.436)
Epoch: [17][140/200]	Time 0.205 (0.266)	Data 0.000 (0.057)	Loss 2.164 (1.489)
Epoch: [17][160/200]	Time 0.208 (0.271)	Data 0.000 (0.062)	Loss 2.003 (1.522)
Epoch: [17][180/200]	Time 0.204 (0.274)	Data 0.000 (0.065)	Loss 1.894 (1.568)
Epoch: [17][200/200]	Time 0.203 (0.275)	Data 0.001 (0.066)	Loss 1.953 (1.585)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.211)	Data 0.000 (0.137)	
Extract Features: [100/128]	Time 0.078 (0.192)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.57906246185303
==> Statistics for epoch 18: 930 clusters
Epoch: [18][20/200]	Time 0.203 (0.264)	Data 0.001 (0.060)	Loss 0.518 (0.395)
Epoch: [18][40/200]	Time 0.205 (0.277)	Data 0.001 (0.073)	Loss 1.310 (0.730)
Epoch: [18][60/200]	Time 0.203 (0.280)	Data 0.001 (0.075)	Loss 1.654 (1.089)
Epoch: [18][80/200]	Time 0.208 (0.261)	Data 0.000 (0.057)	Loss 1.485 (1.262)
Epoch: [18][100/200]	Time 0.206 (0.266)	Data 0.001 (0.062)	Loss 1.770 (1.360)
Epoch: [18][120/200]	Time 0.203 (0.270)	Data 0.001 (0.064)	Loss 1.260 (1.418)
Epoch: [18][140/200]	Time 0.207 (0.263)	Data 0.000 (0.057)	Loss 1.706 (1.481)
Epoch: [18][160/200]	Time 0.215 (0.267)	Data 0.001 (0.061)	Loss 1.468 (1.520)
Epoch: [18][180/200]	Time 0.206 (0.271)	Data 0.001 (0.064)	Loss 2.009 (1.552)
Epoch: [18][200/200]	Time 0.204 (0.265)	Data 0.000 (0.058)	Loss 1.745 (1.567)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.072 (0.207)	Data 0.000 (0.137)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.60673713684082
==> Statistics for epoch 19: 931 clusters
Epoch: [19][20/200]	Time 0.204 (0.266)	Data 0.001 (0.061)	Loss 0.233 (0.401)
Epoch: [19][40/200]	Time 0.203 (0.282)	Data 0.000 (0.073)	Loss 1.917 (0.723)
Epoch: [19][60/200]	Time 0.202 (0.285)	Data 0.001 (0.077)	Loss 2.067 (1.124)
Epoch: [19][80/200]	Time 0.210 (0.265)	Data 0.001 (0.058)	Loss 1.475 (1.284)
Epoch: [19][100/200]	Time 0.223 (0.271)	Data 0.019 (0.064)	Loss 1.765 (1.376)
Epoch: [19][120/200]	Time 0.211 (0.274)	Data 0.001 (0.067)	Loss 1.557 (1.438)
Epoch: [19][140/200]	Time 0.207 (0.265)	Data 0.000 (0.058)	Loss 1.979 (1.506)
Epoch: [19][160/200]	Time 0.202 (0.267)	Data 0.000 (0.061)	Loss 1.966 (1.544)
Epoch: [19][180/200]	Time 0.201 (0.269)	Data 0.000 (0.064)	Loss 1.895 (1.568)
Epoch: [19][200/200]	Time 0.199 (0.264)	Data 0.000 (0.059)	Loss 1.808 (1.594)
Extract Features: [50/367]	Time 0.069 (0.210)	Data 0.000 (0.138)	
Extract Features: [100/367]	Time 0.068 (0.197)	Data 0.000 (0.125)	
Extract Features: [150/367]	Time 0.068 (0.194)	Data 0.000 (0.123)	
Extract Features: [200/367]	Time 0.069 (0.192)	Data 0.000 (0.121)	
Extract Features: [250/367]	Time 0.068 (0.189)	Data 0.000 (0.118)	
Extract Features: [300/367]	Time 0.069 (0.189)	Data 0.000 (0.118)	
Extract Features: [350/367]	Time 0.070 (0.187)	Data 0.000 (0.116)	
Mean AP: 39.6%

 * Finished epoch  19  model mAP: 39.6%  best: 39.6% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.208)	Data 0.000 (0.139)	
Extract Features: [100/128]	Time 0.068 (0.187)	Data 0.000 (0.116)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.54834461212158
==> Statistics for epoch 20: 935 clusters
Epoch: [20][20/200]	Time 0.202 (0.258)	Data 0.001 (0.052)	Loss 0.200 (0.366)
Epoch: [20][40/200]	Time 0.204 (0.269)	Data 0.001 (0.064)	Loss 1.785 (0.742)
Epoch: [20][60/200]	Time 0.207 (0.279)	Data 0.001 (0.073)	Loss 2.291 (1.038)
Epoch: [20][80/200]	Time 0.204 (0.260)	Data 0.001 (0.055)	Loss 1.506 (1.197)
Epoch: [20][100/200]	Time 0.205 (0.268)	Data 0.001 (0.061)	Loss 1.735 (1.304)
Epoch: [20][120/200]	Time 0.202 (0.271)	Data 0.001 (0.064)	Loss 1.611 (1.373)
Epoch: [20][140/200]	Time 0.220 (0.263)	Data 0.001 (0.057)	Loss 1.742 (1.417)
Epoch: [20][160/200]	Time 0.203 (0.267)	Data 0.001 (0.060)	Loss 1.166 (1.443)
Epoch: [20][180/200]	Time 0.203 (0.270)	Data 0.001 (0.063)	Loss 1.697 (1.458)
Epoch: [20][200/200]	Time 0.200 (0.264)	Data 0.000 (0.058)	Loss 1.200 (1.476)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.212)	Data 0.000 (0.143)	
Extract Features: [100/128]	Time 0.068 (0.190)	Data 0.000 (0.119)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.33301663398743
==> Statistics for epoch 21: 919 clusters
Epoch: [21][20/200]	Time 0.208 (0.265)	Data 0.001 (0.059)	Loss 0.390 (0.376)
Epoch: [21][40/200]	Time 0.203 (0.280)	Data 0.000 (0.074)	Loss 1.676 (0.749)
Epoch: [21][60/200]	Time 0.210 (0.287)	Data 0.001 (0.082)	Loss 2.035 (1.051)
Epoch: [21][80/200]	Time 0.202 (0.269)	Data 0.000 (0.065)	Loss 1.460 (1.208)
Epoch: [21][100/200]	Time 0.203 (0.275)	Data 0.001 (0.070)	Loss 1.720 (1.297)
Epoch: [21][120/200]	Time 0.215 (0.279)	Data 0.001 (0.073)	Loss 1.584 (1.359)
Epoch: [21][140/200]	Time 0.206 (0.268)	Data 0.000 (0.063)	Loss 1.361 (1.403)
Epoch: [21][160/200]	Time 0.207 (0.271)	Data 0.000 (0.066)	Loss 1.458 (1.432)
Epoch: [21][180/200]	Time 0.208 (0.274)	Data 0.000 (0.068)	Loss 1.401 (1.452)
Epoch: [21][200/200]	Time 0.207 (0.275)	Data 0.001 (0.069)	Loss 1.717 (1.480)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.212)	Data 0.000 (0.139)	
Extract Features: [100/128]	Time 0.072 (0.194)	Data 0.000 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.883453607559204
==> Statistics for epoch 22: 925 clusters
Epoch: [22][20/200]	Time 0.209 (0.264)	Data 0.001 (0.056)	Loss 0.200 (0.369)
Epoch: [22][40/200]	Time 0.215 (0.279)	Data 0.001 (0.071)	Loss 1.590 (0.766)
Epoch: [22][60/200]	Time 0.204 (0.286)	Data 0.001 (0.073)	Loss 1.513 (1.078)
Epoch: [22][80/200]	Time 0.204 (0.266)	Data 0.000 (0.055)	Loss 1.811 (1.214)
Epoch: [22][100/200]	Time 0.206 (0.271)	Data 0.001 (0.061)	Loss 1.632 (1.301)
Epoch: [22][120/200]	Time 0.205 (0.275)	Data 0.001 (0.065)	Loss 1.611 (1.362)
Epoch: [22][140/200]	Time 0.201 (0.266)	Data 0.000 (0.056)	Loss 1.437 (1.402)
Epoch: [22][160/200]	Time 0.211 (0.270)	Data 0.001 (0.061)	Loss 1.888 (1.440)
Epoch: [22][180/200]	Time 0.210 (0.273)	Data 0.001 (0.064)	Loss 1.271 (1.461)
Epoch: [22][200/200]	Time 0.214 (0.274)	Data 0.001 (0.066)	Loss 1.459 (1.481)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.076 (0.206)	Data 0.006 (0.133)	
Extract Features: [100/128]	Time 0.148 (0.188)	Data 0.080 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.18586730957031
==> Statistics for epoch 23: 922 clusters
Epoch: [23][20/200]	Time 0.211 (0.273)	Data 0.001 (0.054)	Loss 0.502 (0.363)
Epoch: [23][40/200]	Time 0.225 (0.282)	Data 0.001 (0.066)	Loss 2.034 (0.713)
Epoch: [23][60/200]	Time 0.210 (0.283)	Data 0.002 (0.068)	Loss 1.980 (1.030)
Epoch: [23][80/200]	Time 0.208 (0.268)	Data 0.000 (0.053)	Loss 1.592 (1.208)
Epoch: [23][100/200]	Time 0.205 (0.273)	Data 0.001 (0.059)	Loss 1.599 (1.286)
Epoch: [23][120/200]	Time 0.210 (0.276)	Data 0.001 (0.064)	Loss 1.640 (1.343)
Epoch: [23][140/200]	Time 0.202 (0.266)	Data 0.000 (0.055)	Loss 1.839 (1.383)
Epoch: [23][160/200]	Time 0.204 (0.270)	Data 0.001 (0.060)	Loss 1.189 (1.420)
Epoch: [23][180/200]	Time 0.217 (0.274)	Data 0.001 (0.064)	Loss 1.697 (1.434)
Epoch: [23][200/200]	Time 0.210 (0.275)	Data 0.001 (0.065)	Loss 1.514 (1.461)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.072 (0.207)	Data 0.000 (0.134)	
Extract Features: [100/128]	Time 0.068 (0.189)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.91940093040466
==> Statistics for epoch 24: 919 clusters
Epoch: [24][20/200]	Time 0.201 (0.259)	Data 0.001 (0.053)	Loss 0.494 (0.341)
Epoch: [24][40/200]	Time 0.203 (0.276)	Data 0.001 (0.071)	Loss 1.483 (0.737)
Epoch: [24][60/200]	Time 0.203 (0.278)	Data 0.001 (0.073)	Loss 1.607 (1.022)
Epoch: [24][80/200]	Time 0.205 (0.262)	Data 0.000 (0.058)	Loss 1.738 (1.193)
Epoch: [24][100/200]	Time 0.206 (0.266)	Data 0.001 (0.062)	Loss 1.453 (1.275)
Epoch: [24][120/200]	Time 0.206 (0.270)	Data 0.001 (0.065)	Loss 1.762 (1.338)
Epoch: [24][140/200]	Time 0.204 (0.262)	Data 0.000 (0.057)	Loss 1.319 (1.374)
Epoch: [24][160/200]	Time 0.206 (0.266)	Data 0.001 (0.060)	Loss 1.936 (1.399)
Epoch: [24][180/200]	Time 0.213 (0.269)	Data 0.001 (0.063)	Loss 1.441 (1.423)
Epoch: [24][200/200]	Time 0.207 (0.271)	Data 0.001 (0.065)	Loss 1.585 (1.444)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.201)	Data 0.000 (0.129)	
Extract Features: [100/128]	Time 0.068 (0.185)	Data 0.000 (0.113)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.351757764816284
==> Statistics for epoch 25: 932 clusters
Epoch: [25][20/200]	Time 0.207 (0.269)	Data 0.001 (0.063)	Loss 0.287 (0.359)
Epoch: [25][40/200]	Time 0.202 (0.275)	Data 0.001 (0.070)	Loss 1.693 (0.684)
Epoch: [25][60/200]	Time 0.201 (0.277)	Data 0.000 (0.072)	Loss 1.570 (1.008)
Epoch: [25][80/200]	Time 0.203 (0.261)	Data 0.000 (0.055)	Loss 1.853 (1.176)
Epoch: [25][100/200]	Time 0.212 (0.266)	Data 0.000 (0.060)	Loss 0.998 (1.257)
Epoch: [25][120/200]	Time 0.205 (0.271)	Data 0.001 (0.065)	Loss 1.966 (1.318)
Epoch: [25][140/200]	Time 0.226 (0.263)	Data 0.000 (0.057)	Loss 1.576 (1.373)
Epoch: [25][160/200]	Time 0.208 (0.266)	Data 0.001 (0.059)	Loss 1.248 (1.402)
Epoch: [25][180/200]	Time 0.202 (0.268)	Data 0.001 (0.061)	Loss 1.446 (1.431)
Epoch: [25][200/200]	Time 0.200 (0.263)	Data 0.000 (0.056)	Loss 1.797 (1.451)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.200)	Data 0.000 (0.128)	
Extract Features: [100/128]	Time 0.070 (0.188)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.4662070274353
==> Statistics for epoch 26: 923 clusters
Epoch: [26][20/200]	Time 0.206 (0.269)	Data 0.001 (0.052)	Loss 0.287 (0.329)
Epoch: [26][40/200]	Time 0.205 (0.276)	Data 0.001 (0.064)	Loss 1.942 (0.697)
Epoch: [26][60/200]	Time 0.206 (0.279)	Data 0.001 (0.070)	Loss 1.510 (0.985)
Epoch: [26][80/200]	Time 0.205 (0.264)	Data 0.000 (0.054)	Loss 1.079 (1.151)
Epoch: [26][100/200]	Time 0.216 (0.270)	Data 0.001 (0.061)	Loss 1.652 (1.240)
Epoch: [26][120/200]	Time 0.213 (0.272)	Data 0.000 (0.063)	Loss 1.233 (1.312)
Epoch: [26][140/200]	Time 0.205 (0.264)	Data 0.000 (0.056)	Loss 1.295 (1.355)
Epoch: [26][160/200]	Time 0.205 (0.267)	Data 0.001 (0.059)	Loss 1.666 (1.382)
Epoch: [26][180/200]	Time 0.205 (0.269)	Data 0.001 (0.061)	Loss 1.116 (1.418)
Epoch: [26][200/200]	Time 0.204 (0.271)	Data 0.001 (0.063)	Loss 1.720 (1.444)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.505 (0.210)	Data 0.436 (0.138)	
Extract Features: [100/128]	Time 0.071 (0.191)	Data 0.000 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.847867012023926
==> Statistics for epoch 27: 926 clusters
Epoch: [27][20/200]	Time 0.205 (0.270)	Data 0.001 (0.061)	Loss 0.331 (0.299)
Epoch: [27][40/200]	Time 0.203 (0.283)	Data 0.001 (0.076)	Loss 1.481 (0.665)
Epoch: [27][60/200]	Time 0.210 (0.282)	Data 0.001 (0.076)	Loss 2.354 (0.996)
Epoch: [27][80/200]	Time 0.213 (0.265)	Data 0.001 (0.059)	Loss 1.663 (1.141)
Epoch: [27][100/200]	Time 0.207 (0.271)	Data 0.001 (0.065)	Loss 1.623 (1.257)
Epoch: [27][120/200]	Time 0.203 (0.275)	Data 0.001 (0.069)	Loss 1.410 (1.308)
Epoch: [27][140/200]	Time 0.202 (0.266)	Data 0.000 (0.060)	Loss 1.495 (1.344)
Epoch: [27][160/200]	Time 0.218 (0.270)	Data 0.001 (0.063)	Loss 1.460 (1.376)
Epoch: [27][180/200]	Time 0.211 (0.272)	Data 0.001 (0.065)	Loss 2.076 (1.411)
Epoch: [27][200/200]	Time 0.206 (0.275)	Data 0.001 (0.067)	Loss 1.097 (1.429)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.208)	Data 0.000 (0.135)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.116)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.736783027648926
==> Statistics for epoch 28: 923 clusters
Epoch: [28][20/200]	Time 0.206 (0.266)	Data 0.001 (0.057)	Loss 0.290 (0.320)
Epoch: [28][40/200]	Time 0.205 (0.282)	Data 0.001 (0.075)	Loss 1.447 (0.688)
Epoch: [28][60/200]	Time 0.203 (0.284)	Data 0.001 (0.078)	Loss 2.186 (1.028)
Epoch: [28][80/200]	Time 0.217 (0.269)	Data 0.000 (0.061)	Loss 1.195 (1.156)
Epoch: [28][100/200]	Time 0.211 (0.274)	Data 0.001 (0.066)	Loss 1.338 (1.233)
Epoch: [28][120/200]	Time 0.207 (0.276)	Data 0.001 (0.068)	Loss 2.071 (1.284)
Epoch: [28][140/200]	Time 0.203 (0.266)	Data 0.000 (0.059)	Loss 1.475 (1.318)
Epoch: [28][160/200]	Time 0.211 (0.270)	Data 0.001 (0.062)	Loss 1.531 (1.362)
Epoch: [28][180/200]	Time 0.205 (0.271)	Data 0.001 (0.064)	Loss 1.423 (1.389)
Epoch: [28][200/200]	Time 0.202 (0.274)	Data 0.000 (0.066)	Loss 2.102 (1.421)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.071 (0.207)	Data 0.000 (0.134)	
Extract Features: [100/128]	Time 0.069 (0.188)	Data 0.000 (0.116)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.93126893043518
==> Statistics for epoch 29: 928 clusters
Epoch: [29][20/200]	Time 0.210 (0.270)	Data 0.001 (0.055)	Loss 0.367 (0.351)
Epoch: [29][40/200]	Time 0.202 (0.279)	Data 0.001 (0.069)	Loss 1.782 (0.688)
Epoch: [29][60/200]	Time 0.206 (0.283)	Data 0.001 (0.074)	Loss 1.570 (1.021)
Epoch: [29][80/200]	Time 0.204 (0.266)	Data 0.000 (0.056)	Loss 1.578 (1.185)
Epoch: [29][100/200]	Time 0.218 (0.272)	Data 0.015 (0.062)	Loss 1.836 (1.271)
Epoch: [29][120/200]	Time 0.208 (0.277)	Data 0.001 (0.069)	Loss 2.080 (1.323)
Epoch: [29][140/200]	Time 0.210 (0.268)	Data 0.000 (0.059)	Loss 2.089 (1.353)
Epoch: [29][160/200]	Time 0.203 (0.271)	Data 0.001 (0.063)	Loss 1.842 (1.376)
Epoch: [29][180/200]	Time 0.208 (0.272)	Data 0.000 (0.064)	Loss 1.345 (1.396)
Epoch: [29][200/200]	Time 0.207 (0.266)	Data 0.000 (0.058)	Loss 1.671 (1.409)
Extract Features: [50/367]	Time 0.130 (0.209)	Data 0.062 (0.136)	
Extract Features: [100/367]	Time 0.069 (0.193)	Data 0.000 (0.122)	
Extract Features: [150/367]	Time 0.349 (0.191)	Data 0.280 (0.120)	
Extract Features: [200/367]	Time 0.077 (0.187)	Data 0.000 (0.116)	
Extract Features: [250/367]	Time 0.517 (0.188)	Data 0.449 (0.117)	
Extract Features: [300/367]	Time 0.068 (0.186)	Data 0.000 (0.115)	
Extract Features: [350/367]	Time 0.611 (0.187)	Data 0.535 (0.115)	
Mean AP: 43.3%

 * Finished epoch  29  model mAP: 43.3%  best: 43.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.067 (0.208)	Data 0.000 (0.138)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.170111894607544
==> Statistics for epoch 30: 919 clusters
Epoch: [30][20/200]	Time 0.206 (0.261)	Data 0.001 (0.057)	Loss 0.246 (0.318)
Epoch: [30][40/200]	Time 0.203 (0.273)	Data 0.001 (0.070)	Loss 1.340 (0.688)
Epoch: [30][60/200]	Time 0.205 (0.278)	Data 0.000 (0.072)	Loss 1.508 (0.979)
Epoch: [30][80/200]	Time 0.203 (0.259)	Data 0.000 (0.054)	Loss 1.388 (1.133)
Epoch: [30][100/200]	Time 0.210 (0.265)	Data 0.001 (0.060)	Loss 1.674 (1.222)
Epoch: [30][120/200]	Time 0.202 (0.268)	Data 0.001 (0.063)	Loss 1.671 (1.282)
Epoch: [30][140/200]	Time 0.203 (0.260)	Data 0.000 (0.056)	Loss 1.501 (1.325)
Epoch: [30][160/200]	Time 0.206 (0.265)	Data 0.001 (0.060)	Loss 1.919 (1.362)
Epoch: [30][180/200]	Time 0.203 (0.268)	Data 0.001 (0.064)	Loss 1.336 (1.379)
Epoch: [30][200/200]	Time 0.206 (0.270)	Data 0.001 (0.066)	Loss 2.176 (1.403)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.667 (0.214)	Data 0.599 (0.142)	
Extract Features: [100/128]	Time 0.068 (0.193)	Data 0.000 (0.122)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.60343909263611
==> Statistics for epoch 31: 952 clusters
Epoch: [31][20/200]	Time 0.206 (0.270)	Data 0.001 (0.056)	Loss 0.378 (0.323)
Epoch: [31][40/200]	Time 0.203 (0.282)	Data 0.001 (0.072)	Loss 1.479 (0.648)
Epoch: [31][60/200]	Time 0.203 (0.284)	Data 0.001 (0.076)	Loss 2.122 (1.000)
Epoch: [31][80/200]	Time 0.205 (0.267)	Data 0.001 (0.057)	Loss 1.644 (1.159)
Epoch: [31][100/200]	Time 0.206 (0.272)	Data 0.001 (0.063)	Loss 1.525 (1.242)
Epoch: [31][120/200]	Time 0.208 (0.274)	Data 0.001 (0.066)	Loss 1.614 (1.308)
Epoch: [31][140/200]	Time 0.213 (0.267)	Data 0.000 (0.058)	Loss 1.199 (1.349)
Epoch: [31][160/200]	Time 0.206 (0.271)	Data 0.001 (0.062)	Loss 1.357 (1.379)
Epoch: [31][180/200]	Time 0.202 (0.274)	Data 0.001 (0.065)	Loss 2.077 (1.399)
Epoch: [31][200/200]	Time 0.206 (0.267)	Data 0.000 (0.059)	Loss 1.745 (1.422)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.236 (0.211)	Data 0.168 (0.138)	
Extract Features: [100/128]	Time 0.068 (0.190)	Data 0.000 (0.119)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.78485369682312
==> Statistics for epoch 32: 929 clusters
Epoch: [32][20/200]	Time 0.213 (0.263)	Data 0.001 (0.057)	Loss 0.391 (0.304)
Epoch: [32][40/200]	Time 0.202 (0.273)	Data 0.000 (0.068)	Loss 1.625 (0.646)
Epoch: [32][60/200]	Time 0.206 (0.280)	Data 0.001 (0.075)	Loss 1.422 (0.931)
Epoch: [32][80/200]	Time 0.204 (0.263)	Data 0.001 (0.058)	Loss 1.236 (1.075)
Epoch: [32][100/200]	Time 0.255 (0.270)	Data 0.054 (0.064)	Loss 1.681 (1.177)
Epoch: [32][120/200]	Time 0.208 (0.275)	Data 0.000 (0.069)	Loss 1.850 (1.250)
Epoch: [32][140/200]	Time 0.211 (0.266)	Data 0.000 (0.061)	Loss 1.792 (1.299)
Epoch: [32][160/200]	Time 0.205 (0.269)	Data 0.000 (0.063)	Loss 1.566 (1.328)
Epoch: [32][180/200]	Time 0.203 (0.272)	Data 0.001 (0.066)	Loss 1.620 (1.352)
Epoch: [32][200/200]	Time 0.206 (0.266)	Data 0.000 (0.060)	Loss 1.381 (1.373)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.209)	Data 0.000 (0.136)	
Extract Features: [100/128]	Time 0.106 (0.189)	Data 0.037 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.288378953933716
==> Statistics for epoch 33: 935 clusters
Epoch: [33][20/200]	Time 0.223 (0.270)	Data 0.001 (0.064)	Loss 0.317 (0.334)
Epoch: [33][40/200]	Time 0.202 (0.279)	Data 0.001 (0.074)	Loss 2.081 (0.642)
Epoch: [33][60/200]	Time 0.202 (0.283)	Data 0.001 (0.076)	Loss 1.199 (0.968)
Epoch: [33][80/200]	Time 0.205 (0.265)	Data 0.000 (0.059)	Loss 1.195 (1.136)
Epoch: [33][100/200]	Time 0.210 (0.271)	Data 0.001 (0.065)	Loss 1.070 (1.220)
Epoch: [33][120/200]	Time 0.208 (0.273)	Data 0.001 (0.067)	Loss 1.578 (1.268)
Epoch: [33][140/200]	Time 0.213 (0.264)	Data 0.001 (0.058)	Loss 1.494 (1.306)
Epoch: [33][160/200]	Time 0.201 (0.267)	Data 0.000 (0.061)	Loss 1.791 (1.333)
Epoch: [33][180/200]	Time 0.203 (0.269)	Data 0.001 (0.064)	Loss 1.610 (1.350)
Epoch: [33][200/200]	Time 0.204 (0.264)	Data 0.000 (0.058)	Loss 1.395 (1.378)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.211)	Data 0.000 (0.139)	
Extract Features: [100/128]	Time 0.068 (0.191)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.37291622161865
==> Statistics for epoch 34: 944 clusters
Epoch: [34][20/200]	Time 0.210 (0.266)	Data 0.001 (0.061)	Loss 0.193 (0.306)
Epoch: [34][40/200]	Time 0.205 (0.277)	Data 0.001 (0.073)	Loss 1.605 (0.663)
Epoch: [34][60/200]	Time 0.203 (0.279)	Data 0.001 (0.074)	Loss 1.917 (0.965)
Epoch: [34][80/200]	Time 0.205 (0.260)	Data 0.001 (0.056)	Loss 1.606 (1.118)
Epoch: [34][100/200]	Time 0.211 (0.270)	Data 0.001 (0.065)	Loss 1.687 (1.217)
Epoch: [34][120/200]	Time 0.201 (0.272)	Data 0.001 (0.066)	Loss 1.421 (1.269)
Epoch: [34][140/200]	Time 0.216 (0.263)	Data 0.000 (0.058)	Loss 1.713 (1.331)
Epoch: [34][160/200]	Time 0.203 (0.265)	Data 0.001 (0.060)	Loss 1.397 (1.367)
Epoch: [34][180/200]	Time 0.205 (0.269)	Data 0.001 (0.063)	Loss 1.543 (1.383)
Epoch: [34][200/200]	Time 0.204 (0.263)	Data 0.000 (0.057)	Loss 1.244 (1.398)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.080 (0.202)	Data 0.011 (0.132)	
Extract Features: [100/128]	Time 0.068 (0.193)	Data 0.000 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.651721239089966
==> Statistics for epoch 35: 948 clusters
Epoch: [35][20/200]	Time 0.204 (0.275)	Data 0.001 (0.063)	Loss 0.246 (0.318)
Epoch: [35][40/200]	Time 0.204 (0.285)	Data 0.001 (0.077)	Loss 1.410 (0.628)
Epoch: [35][60/200]	Time 0.208 (0.287)	Data 0.001 (0.080)	Loss 1.208 (0.939)
Epoch: [35][80/200]	Time 0.203 (0.269)	Data 0.001 (0.063)	Loss 1.840 (1.108)
Epoch: [35][100/200]	Time 0.342 (0.277)	Data 0.138 (0.069)	Loss 1.762 (1.199)
Epoch: [35][120/200]	Time 0.204 (0.278)	Data 0.001 (0.070)	Loss 1.746 (1.280)
Epoch: [35][140/200]	Time 0.206 (0.270)	Data 0.000 (0.062)	Loss 1.728 (1.326)
Epoch: [35][160/200]	Time 0.205 (0.272)	Data 0.001 (0.065)	Loss 1.662 (1.358)
Epoch: [35][180/200]	Time 0.217 (0.274)	Data 0.001 (0.066)	Loss 1.428 (1.382)
Epoch: [35][200/200]	Time 0.205 (0.267)	Data 0.000 (0.060)	Loss 1.279 (1.404)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.324 (0.207)	Data 0.255 (0.134)	
Extract Features: [100/128]	Time 0.068 (0.189)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.620646953582764
==> Statistics for epoch 36: 948 clusters
Epoch: [36][20/200]	Time 0.207 (0.259)	Data 0.001 (0.051)	Loss 0.192 (0.325)
Epoch: [36][40/200]	Time 0.204 (0.267)	Data 0.001 (0.061)	Loss 1.602 (0.682)
Epoch: [36][60/200]	Time 0.205 (0.273)	Data 0.001 (0.067)	Loss 1.241 (0.964)
Epoch: [36][80/200]	Time 0.202 (0.258)	Data 0.001 (0.053)	Loss 2.141 (1.131)
Epoch: [36][100/200]	Time 0.222 (0.263)	Data 0.021 (0.058)	Loss 1.843 (1.232)
Epoch: [36][120/200]	Time 0.206 (0.268)	Data 0.001 (0.062)	Loss 1.359 (1.290)
Epoch: [36][140/200]	Time 0.203 (0.259)	Data 0.000 (0.054)	Loss 1.470 (1.331)
Epoch: [36][160/200]	Time 0.206 (0.264)	Data 0.001 (0.058)	Loss 1.585 (1.361)
Epoch: [36][180/200]	Time 0.204 (0.266)	Data 0.001 (0.060)	Loss 1.871 (1.392)
Epoch: [36][200/200]	Time 0.203 (0.260)	Data 0.000 (0.054)	Loss 1.921 (1.414)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.727 (0.215)	Data 0.657 (0.143)	
Extract Features: [100/128]	Time 0.069 (0.194)	Data 0.000 (0.124)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.68509650230408
==> Statistics for epoch 37: 933 clusters
Epoch: [37][20/200]	Time 0.204 (0.255)	Data 0.001 (0.047)	Loss 0.194 (0.295)
Epoch: [37][40/200]	Time 0.213 (0.278)	Data 0.001 (0.067)	Loss 1.851 (0.625)
Epoch: [37][60/200]	Time 0.201 (0.280)	Data 0.001 (0.070)	Loss 2.075 (0.948)
Epoch: [37][80/200]	Time 0.221 (0.262)	Data 0.016 (0.053)	Loss 1.136 (1.074)
Epoch: [37][100/200]	Time 0.211 (0.269)	Data 0.001 (0.060)	Loss 1.362 (1.177)
Epoch: [37][120/200]	Time 0.203 (0.271)	Data 0.001 (0.063)	Loss 1.764 (1.242)
Epoch: [37][140/200]	Time 0.207 (0.262)	Data 0.000 (0.055)	Loss 1.293 (1.279)
Epoch: [37][160/200]	Time 0.204 (0.266)	Data 0.001 (0.059)	Loss 1.778 (1.321)
Epoch: [37][180/200]	Time 0.215 (0.269)	Data 0.001 (0.062)	Loss 1.924 (1.344)
Epoch: [37][200/200]	Time 0.207 (0.263)	Data 0.000 (0.056)	Loss 1.604 (1.356)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.205)	Data 0.000 (0.132)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.116)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.278796434402466
==> Statistics for epoch 38: 926 clusters
Epoch: [38][20/200]	Time 0.203 (0.266)	Data 0.001 (0.062)	Loss 0.290 (0.348)
Epoch: [38][40/200]	Time 0.210 (0.282)	Data 0.000 (0.078)	Loss 1.389 (0.677)
Epoch: [38][60/200]	Time 0.210 (0.285)	Data 0.001 (0.081)	Loss 1.289 (0.971)
Epoch: [38][80/200]	Time 0.202 (0.267)	Data 0.000 (0.064)	Loss 1.704 (1.116)
Epoch: [38][100/200]	Time 0.208 (0.274)	Data 0.001 (0.068)	Loss 1.809 (1.211)
Epoch: [38][120/200]	Time 0.210 (0.278)	Data 0.001 (0.071)	Loss 1.729 (1.278)
Epoch: [38][140/200]	Time 0.204 (0.270)	Data 0.000 (0.063)	Loss 1.875 (1.317)
Epoch: [38][160/200]	Time 0.369 (0.273)	Data 0.001 (0.065)	Loss 2.198 (1.356)
Epoch: [38][180/200]	Time 0.219 (0.274)	Data 0.000 (0.067)	Loss 1.858 (1.377)
Epoch: [38][200/200]	Time 0.206 (0.275)	Data 0.000 (0.068)	Loss 0.953 (1.392)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.206)	Data 0.000 (0.136)	
Extract Features: [100/128]	Time 0.068 (0.191)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.063241481781006
==> Statistics for epoch 39: 933 clusters
Epoch: [39][20/200]	Time 0.201 (0.270)	Data 0.000 (0.066)	Loss 0.186 (0.294)
Epoch: [39][40/200]	Time 0.204 (0.286)	Data 0.001 (0.079)	Loss 1.538 (0.642)
Epoch: [39][60/200]	Time 0.205 (0.285)	Data 0.001 (0.078)	Loss 1.582 (0.971)
Epoch: [39][80/200]	Time 0.204 (0.266)	Data 0.001 (0.061)	Loss 1.217 (1.103)
Epoch: [39][100/200]	Time 0.204 (0.269)	Data 0.001 (0.064)	Loss 1.877 (1.216)
Epoch: [39][120/200]	Time 0.207 (0.272)	Data 0.001 (0.065)	Loss 1.781 (1.276)
Epoch: [39][140/200]	Time 0.204 (0.264)	Data 0.000 (0.057)	Loss 1.893 (1.317)
Epoch: [39][160/200]	Time 0.212 (0.268)	Data 0.001 (0.061)	Loss 1.619 (1.348)
Epoch: [39][180/200]	Time 0.203 (0.271)	Data 0.001 (0.065)	Loss 1.521 (1.367)
Epoch: [39][200/200]	Time 0.203 (0.264)	Data 0.000 (0.058)	Loss 1.172 (1.383)
Extract Features: [50/367]	Time 0.067 (0.214)	Data 0.000 (0.141)	
Extract Features: [100/367]	Time 0.070 (0.195)	Data 0.001 (0.125)	
Extract Features: [150/367]	Time 0.068 (0.191)	Data 0.000 (0.120)	
Extract Features: [200/367]	Time 0.067 (0.189)	Data 0.000 (0.118)	
Extract Features: [250/367]	Time 0.068 (0.186)	Data 0.000 (0.116)	
Extract Features: [300/367]	Time 0.069 (0.186)	Data 0.000 (0.115)	
Extract Features: [350/367]	Time 0.071 (0.184)	Data 0.000 (0.113)	
Mean AP: 44.1%

 * Finished epoch  39  model mAP: 44.1%  best: 44.1% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.067 (0.202)	Data 0.000 (0.132)	
Extract Features: [100/128]	Time 0.068 (0.188)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.97104811668396
==> Statistics for epoch 40: 917 clusters
Epoch: [40][20/200]	Time 0.207 (0.264)	Data 0.001 (0.056)	Loss 0.231 (0.337)
Epoch: [40][40/200]	Time 0.210 (0.280)	Data 0.001 (0.074)	Loss 1.311 (0.640)
Epoch: [40][60/200]	Time 0.208 (0.285)	Data 0.001 (0.079)	Loss 0.898 (0.903)
Epoch: [40][80/200]	Time 0.207 (0.268)	Data 0.000 (0.063)	Loss 1.433 (1.066)
Epoch: [40][100/200]	Time 0.208 (0.273)	Data 0.001 (0.068)	Loss 1.366 (1.162)
Epoch: [40][120/200]	Time 0.209 (0.276)	Data 0.001 (0.070)	Loss 1.426 (1.227)
Epoch: [40][140/200]	Time 0.201 (0.266)	Data 0.000 (0.061)	Loss 1.494 (1.263)
Epoch: [40][160/200]	Time 0.205 (0.270)	Data 0.001 (0.065)	Loss 2.164 (1.293)
Epoch: [40][180/200]	Time 0.206 (0.274)	Data 0.001 (0.068)	Loss 2.389 (1.321)
Epoch: [40][200/200]	Time 0.206 (0.277)	Data 0.001 (0.070)	Loss 1.807 (1.341)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.207)	Data 0.000 (0.138)	
Extract Features: [100/128]	Time 0.067 (0.187)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.2423620223999
==> Statistics for epoch 41: 933 clusters
Epoch: [41][20/200]	Time 0.207 (0.275)	Data 0.001 (0.069)	Loss 0.172 (0.334)
Epoch: [41][40/200]	Time 0.202 (0.286)	Data 0.000 (0.081)	Loss 1.119 (0.649)
Epoch: [41][60/200]	Time 0.204 (0.289)	Data 0.001 (0.081)	Loss 1.359 (0.957)
Epoch: [41][80/200]	Time 0.204 (0.272)	Data 0.001 (0.064)	Loss 1.709 (1.097)
Epoch: [41][100/200]	Time 0.276 (0.277)	Data 0.075 (0.070)	Loss 1.892 (1.171)
Epoch: [41][120/200]	Time 0.207 (0.280)	Data 0.000 (0.073)	Loss 1.507 (1.245)
Epoch: [41][140/200]	Time 0.205 (0.271)	Data 0.000 (0.064)	Loss 1.812 (1.287)
Epoch: [41][160/200]	Time 0.209 (0.275)	Data 0.000 (0.068)	Loss 1.461 (1.321)
Epoch: [41][180/200]	Time 0.204 (0.277)	Data 0.001 (0.070)	Loss 1.574 (1.348)
Epoch: [41][200/200]	Time 0.251 (0.270)	Data 0.048 (0.064)	Loss 1.619 (1.368)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.117 (0.209)	Data 0.048 (0.136)	
Extract Features: [100/128]	Time 0.068 (0.190)	Data 0.000 (0.118)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.46649241447449
==> Statistics for epoch 42: 931 clusters
Epoch: [42][20/200]	Time 0.203 (0.273)	Data 0.001 (0.067)	Loss 0.404 (0.326)
Epoch: [42][40/200]	Time 0.203 (0.284)	Data 0.001 (0.079)	Loss 1.143 (0.660)
Epoch: [42][60/200]	Time 0.211 (0.285)	Data 0.001 (0.081)	Loss 1.283 (0.944)
Epoch: [42][80/200]	Time 0.206 (0.265)	Data 0.001 (0.061)	Loss 1.390 (1.101)
Epoch: [42][100/200]	Time 0.228 (0.270)	Data 0.022 (0.066)	Loss 1.563 (1.205)
Epoch: [42][120/200]	Time 0.211 (0.274)	Data 0.001 (0.068)	Loss 1.556 (1.265)
Epoch: [42][140/200]	Time 0.205 (0.265)	Data 0.000 (0.060)	Loss 1.540 (1.310)
Epoch: [42][160/200]	Time 0.204 (0.269)	Data 0.000 (0.063)	Loss 1.415 (1.335)
Epoch: [42][180/200]	Time 0.203 (0.272)	Data 0.000 (0.066)	Loss 1.205 (1.355)
Epoch: [42][200/200]	Time 0.201 (0.266)	Data 0.000 (0.060)	Loss 1.675 (1.371)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.205)	Data 0.000 (0.136)	
Extract Features: [100/128]	Time 0.595 (0.192)	Data 0.527 (0.121)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.6195809841156
==> Statistics for epoch 43: 927 clusters
Epoch: [43][20/200]	Time 0.208 (0.261)	Data 0.001 (0.056)	Loss 0.365 (0.315)
Epoch: [43][40/200]	Time 0.207 (0.275)	Data 0.001 (0.069)	Loss 1.740 (0.691)
Epoch: [43][60/200]	Time 0.206 (0.282)	Data 0.001 (0.074)	Loss 1.396 (0.971)
Epoch: [43][80/200]	Time 0.206 (0.264)	Data 0.000 (0.056)	Loss 1.568 (1.111)
Epoch: [43][100/200]	Time 0.207 (0.270)	Data 0.001 (0.063)	Loss 1.466 (1.209)
Epoch: [43][120/200]	Time 0.211 (0.275)	Data 0.001 (0.067)	Loss 1.252 (1.259)
Epoch: [43][140/200]	Time 0.202 (0.265)	Data 0.000 (0.057)	Loss 1.501 (1.303)
Epoch: [43][160/200]	Time 0.209 (0.270)	Data 0.001 (0.062)	Loss 1.385 (1.337)
Epoch: [43][180/200]	Time 0.208 (0.272)	Data 0.001 (0.064)	Loss 1.036 (1.369)
Epoch: [43][200/200]	Time 0.218 (0.274)	Data 0.001 (0.066)	Loss 1.846 (1.387)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.069 (0.213)	Data 0.000 (0.143)	
Extract Features: [100/128]	Time 0.068 (0.191)	Data 0.000 (0.120)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.60680317878723
==> Statistics for epoch 44: 929 clusters
Epoch: [44][20/200]	Time 0.204 (0.255)	Data 0.001 (0.047)	Loss 0.231 (0.309)
Epoch: [44][40/200]	Time 0.203 (0.273)	Data 0.001 (0.066)	Loss 1.822 (0.615)
Epoch: [44][60/200]	Time 0.204 (0.280)	Data 0.001 (0.071)	Loss 1.579 (0.935)
Epoch: [44][80/200]	Time 0.206 (0.262)	Data 0.001 (0.054)	Loss 1.246 (1.083)
Epoch: [44][100/200]	Time 0.203 (0.268)	Data 0.001 (0.060)	Loss 1.871 (1.187)
Epoch: [44][120/200]	Time 0.203 (0.271)	Data 0.001 (0.064)	Loss 1.515 (1.263)
Epoch: [44][140/200]	Time 0.206 (0.264)	Data 0.000 (0.055)	Loss 1.909 (1.307)
Epoch: [44][160/200]	Time 0.202 (0.268)	Data 0.001 (0.060)	Loss 1.318 (1.339)
Epoch: [44][180/200]	Time 0.210 (0.271)	Data 0.001 (0.063)	Loss 1.446 (1.356)
Epoch: [44][200/200]	Time 0.205 (0.266)	Data 0.000 (0.057)	Loss 1.769 (1.376)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.087 (0.206)	Data 0.016 (0.136)	
Extract Features: [100/128]	Time 0.069 (0.188)	Data 0.000 (0.117)	
Computing jaccard distance...
Jaccard distance computing time cost: 59.19433379173279
==> Statistics for epoch 45: 934 clusters
Epoch: [45][20/200]	Time 0.221 (0.271)	Data 0.001 (0.054)	Loss 0.210 (0.314)
Epoch: [45][40/200]	Time 0.203 (0.282)	Data 0.001 (0.068)	Loss 1.724 (0.619)
Epoch: [45][60/200]	Time 0.202 (0.286)	Data 0.000 (0.075)	Loss 1.599 (0.963)
Epoch: [45][80/200]	Time 0.211 (0.267)	Data 0.000 (0.057)	Loss 1.545 (1.136)
Epoch: [45][100/200]	Time 0.207 (0.273)	Data 0.000 (0.062)	Loss 1.575 (1.209)
Epoch: [45][120/200]	Time 0.206 (0.275)	Data 0.001 (0.065)	Loss 2.126 (1.260)
Epoch: [45][140/200]	Time 0.219 (0.266)	Data 0.000 (0.057)	Loss 1.445 (1.306)
Epoch: [45][160/200]	Time 0.205 (0.271)	Data 0.000 (0.061)	Loss 1.677 (1.344)
Epoch: [45][180/200]	Time 0.204 (0.275)	Data 0.000 (0.065)	Loss 1.482 (1.366)
Epoch: [45][200/200]	Time 0.206 (0.268)	Data 0.000 (0.058)	Loss 1.629 (1.381)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.213)	Data 0.000 (0.140)	
Extract Features: [100/128]	Time 0.068 (0.198)	Data 0.000 (0.126)	
Computing jaccard distance...
Jaccard distance computing time cost: 62.69163990020752
==> Statistics for epoch 46: 920 clusters
Epoch: [46][20/200]	Time 0.205 (0.264)	Data 0.001 (0.058)	Loss 0.208 (0.298)
Epoch: [46][40/200]	Time 0.206 (0.273)	Data 0.001 (0.068)	Loss 1.541 (0.657)
Epoch: [46][60/200]	Time 0.203 (0.281)	Data 0.001 (0.075)	Loss 1.119 (0.934)
Epoch: [46][80/200]	Time 0.206 (0.265)	Data 0.000 (0.057)	Loss 2.279 (1.087)
Epoch: [46][100/200]	Time 0.206 (0.271)	Data 0.001 (0.064)	Loss 1.460 (1.184)
Epoch: [46][120/200]	Time 0.211 (0.274)	Data 0.001 (0.067)	Loss 1.837 (1.247)
Epoch: [46][140/200]	Time 0.202 (0.264)	Data 0.000 (0.057)	Loss 1.459 (1.279)
Epoch: [46][160/200]	Time 0.203 (0.267)	Data 0.001 (0.061)	Loss 1.469 (1.325)
Epoch: [46][180/200]	Time 0.207 (0.270)	Data 0.001 (0.063)	Loss 2.048 (1.342)
Epoch: [46][200/200]	Time 0.375 (0.274)	Data 0.001 (0.066)	Loss 1.694 (1.354)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.309 (0.211)	Data 0.240 (0.141)	
Extract Features: [100/128]	Time 0.068 (0.193)	Data 0.000 (0.122)	
Computing jaccard distance...
Jaccard distance computing time cost: 58.99518275260925
==> Statistics for epoch 47: 943 clusters
Epoch: [47][20/200]	Time 0.206 (0.254)	Data 0.001 (0.047)	Loss 0.488 (0.339)
Epoch: [47][40/200]	Time 0.206 (0.274)	Data 0.001 (0.067)	Loss 1.706 (0.658)
Epoch: [47][60/200]	Time 0.203 (0.278)	Data 0.001 (0.069)	Loss 1.324 (0.950)
Epoch: [47][80/200]	Time 0.207 (0.262)	Data 0.001 (0.054)	Loss 1.869 (1.094)
Epoch: [47][100/200]	Time 0.217 (0.268)	Data 0.016 (0.060)	Loss 1.514 (1.201)
Epoch: [47][120/200]	Time 0.206 (0.272)	Data 0.001 (0.065)	Loss 1.599 (1.264)
Epoch: [47][140/200]	Time 0.206 (0.263)	Data 0.000 (0.056)	Loss 1.708 (1.309)
Epoch: [47][160/200]	Time 0.221 (0.267)	Data 0.000 (0.060)	Loss 1.717 (1.344)
Epoch: [47][180/200]	Time 0.203 (0.270)	Data 0.001 (0.063)	Loss 1.873 (1.374)
Epoch: [47][200/200]	Time 0.202 (0.263)	Data 0.000 (0.056)	Loss 1.579 (1.399)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.068 (0.212)	Data 0.001 (0.140)	
Extract Features: [100/128]	Time 0.074 (0.189)	Data 0.006 (0.119)	
Computing jaccard distance...
Jaccard distance computing time cost: 61.68555665016174
==> Statistics for epoch 48: 929 clusters
Epoch: [48][20/200]	Time 0.199 (0.257)	Data 0.001 (0.052)	Loss 0.216 (0.308)
Epoch: [48][40/200]	Time 0.202 (0.272)	Data 0.001 (0.068)	Loss 1.839 (0.619)
Epoch: [48][60/200]	Time 0.203 (0.277)	Data 0.001 (0.073)	Loss 1.487 (0.962)
Epoch: [48][80/200]	Time 0.202 (0.260)	Data 0.000 (0.056)	Loss 1.715 (1.121)
Epoch: [48][100/200]	Time 0.204 (0.264)	Data 0.001 (0.060)	Loss 1.706 (1.211)
Epoch: [48][120/200]	Time 0.206 (0.268)	Data 0.001 (0.063)	Loss 1.770 (1.270)
Epoch: [48][140/200]	Time 0.207 (0.261)	Data 0.000 (0.055)	Loss 1.369 (1.310)
Epoch: [48][160/200]	Time 0.203 (0.265)	Data 0.001 (0.059)	Loss 1.776 (1.335)
Epoch: [48][180/200]	Time 0.203 (0.269)	Data 0.001 (0.063)	Loss 1.929 (1.359)
Epoch: [48][200/200]	Time 0.205 (0.264)	Data 0.000 (0.058)	Loss 1.413 (1.378)
==> Create pseudo labels for unlabeled data
Extract Features: [50/128]	Time 0.070 (0.207)	Data 0.000 (0.137)	
Extract Features: [100/128]	Time 0.069 (0.190)	Data 0.000 (0.119)	
Computing jaccard distance...
Jaccard distance computing time cost: 60.69307279586792
==> Statistics for epoch 49: 934 clusters
Epoch: [49][20/200]	Time 0.202 (0.264)	Data 0.001 (0.058)	Loss 0.283 (0.307)
Epoch: [49][40/200]	Time 0.203 (0.277)	Data 0.001 (0.070)	Loss 1.351 (0.659)
Epoch: [49][60/200]	Time 0.204 (0.282)	Data 0.001 (0.075)	Loss 1.435 (0.952)
Epoch: [49][80/200]	Time 0.201 (0.263)	Data 0.001 (0.057)	Loss 1.545 (1.110)
Epoch: [49][100/200]	Time 0.222 (0.270)	Data 0.016 (0.062)	Loss 1.337 (1.193)
Epoch: [49][120/200]	Time 0.204 (0.274)	Data 0.001 (0.066)	Loss 1.030 (1.240)
Epoch: [49][140/200]	Time 0.208 (0.264)	Data 0.000 (0.057)	Loss 1.519 (1.291)
Epoch: [49][160/200]	Time 0.429 (0.270)	Data 0.001 (0.062)	Loss 1.472 (1.321)
Epoch: [49][180/200]	Time 0.203 (0.272)	Data 0.001 (0.064)	Loss 1.468 (1.343)
Epoch: [49][200/200]	Time 0.204 (0.267)	Data 0.000 (0.058)	Loss 1.834 (1.367)
Extract Features: [50/367]	Time 0.069 (0.214)	Data 0.000 (0.144)	
Extract Features: [100/367]	Time 0.068 (0.197)	Data 0.000 (0.126)	
Extract Features: [150/367]	Time 0.068 (0.193)	Data 0.000 (0.122)	
Extract Features: [200/367]	Time 0.076 (0.189)	Data 0.000 (0.118)	
Extract Features: [250/367]	Time 0.068 (0.187)	Data 0.000 (0.116)	
Extract Features: [300/367]	Time 0.069 (0.185)	Data 0.000 (0.114)	
Extract Features: [350/367]	Time 0.069 (0.184)	Data 0.000 (0.114)	
Mean AP: 44.3%

 * Finished epoch  49  model mAP: 44.3%  best: 44.3% *

==> Test with the best model:
=> Loaded checkpoint '.log/market2msmt/vit_tiny_cion/model_best.pth.tar'
Extract Features: [50/367]	Time 0.168 (0.215)	Data 0.100 (0.142)	
Extract Features: [100/367]	Time 0.069 (0.197)	Data 0.000 (0.125)	
Extract Features: [150/367]	Time 0.170 (0.191)	Data 0.101 (0.120)	
Extract Features: [200/367]	Time 0.069 (0.187)	Data 0.000 (0.116)	
Extract Features: [250/367]	Time 0.199 (0.186)	Data 0.130 (0.115)	
Extract Features: [300/367]	Time 0.068 (0.184)	Data 0.000 (0.113)	
Extract Features: [350/367]	Time 0.220 (0.184)	Data 0.151 (0.113)	
Mean AP: 44.3%
CMC Scores:
  top-1          69.3%
  top-5          80.3%
  top-10         84.1%
Total running time:  2:31:46.194094
