==========
Args:Namespace(dataset='market1501', batch_size=256, workers=4, height=256, width=128, num_instances=8, eps=0.6, eps_gap=0.02, k1=30, k2=6, arch='vit_small', pretrained_path='/home/ma-user/work/Projects/ReIDNets_Checkpoints_TransReID/ViT_Small_P16_2468_406080100_bs96_originaldino_noreid.pth', features=0, dropout=0, momentum=0.2, drop_path_rate=0.3, hw_ratio=2, self_norm=True, conv_stem=True, lr=0.00035, weight_decay=0.0005, optimizer='SGD', epochs=50, iters=200, step_size=20, seed=1, print_freq=20, eval_step=10, temp=0.05, data_dir='/home/ma-user/work/Projects/ReIDNet_Finetune/CContrast/data', logs_dir='log/market/vit_small_cion', pooling_type='gem', use_hard=True)
==========
==> Load unlabeled dataset
=> Market1501 loaded
Dataset statistics:
  ----------------------------------------
  subset   | # ids | # images | # cameras
  ----------------------------------------
  train    |   751 |    12936 |         6
  query    |   750 |     3368 |         6
  gallery  |   751 |    15913 |         6
  ----------------------------------------
Using convolution stem
using drop_out rate is : 0.0
using attn_drop_out rate is : 0.0
using drop_path rate is : 0.3
Convert dino model......
Load 172 / 177 layers.
optimizer: SGD
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.094 (0.445)	Data 0.000 (0.028)	
Computing jaccard distance...
Jaccard distance computing time cost: 22.825605869293213
Clustering criterion: eps: 0.600
==> Statistics for epoch 0: 508 clusters
Epoch: [0][20/200]	Time 0.268 (0.892)	Data 0.002 (0.103)	Loss 5.361 (4.054)
Epoch: [0][40/200]	Time 0.260 (0.619)	Data 0.000 (0.088)	Loss 4.345 (4.379)
Epoch: [0][60/200]	Time 0.262 (0.528)	Data 0.000 (0.083)	Loss 3.335 (4.112)
Epoch: [0][80/200]	Time 0.271 (0.497)	Data 0.001 (0.096)	Loss 3.311 (3.912)
Epoch: [0][100/200]	Time 0.263 (0.465)	Data 0.000 (0.090)	Loss 3.429 (3.763)
Epoch: [0][120/200]	Time 0.264 (0.442)	Data 0.000 (0.085)	Loss 3.153 (3.634)
Epoch: [0][140/200]	Time 0.265 (0.438)	Data 0.001 (0.093)	Loss 2.782 (3.529)
Epoch: [0][160/200]	Time 0.263 (0.426)	Data 0.000 (0.090)	Loss 2.777 (3.435)
Epoch: [0][180/200]	Time 0.266 (0.416)	Data 0.000 (0.088)	Loss 2.682 (3.366)
Epoch: [0][200/200]	Time 0.269 (0.415)	Data 0.001 (0.092)	Loss 2.602 (3.282)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.158)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.557568788528442
==> Statistics for epoch 1: 551 clusters
Epoch: [1][20/200]	Time 0.263 (0.381)	Data 0.001 (0.109)	Loss 2.841 (0.908)
Epoch: [1][40/200]	Time 0.264 (0.363)	Data 0.001 (0.093)	Loss 2.478 (1.713)
Epoch: [1][60/200]	Time 0.267 (0.359)	Data 0.001 (0.087)	Loss 2.851 (1.960)
Epoch: [1][80/200]	Time 0.267 (0.355)	Data 0.000 (0.083)	Loss 2.771 (2.111)
Epoch: [1][100/200]	Time 0.266 (0.354)	Data 0.000 (0.082)	Loss 2.196 (2.197)
Epoch: [1][120/200]	Time 1.723 (0.365)	Data 1.407 (0.092)	Loss 2.400 (2.231)
Epoch: [1][140/200]	Time 0.274 (0.362)	Data 0.001 (0.089)	Loss 2.070 (2.268)
Epoch: [1][160/200]	Time 0.269 (0.361)	Data 0.001 (0.088)	Loss 2.498 (2.283)
Epoch: [1][180/200]	Time 0.266 (0.360)	Data 0.000 (0.086)	Loss 2.205 (2.285)
Epoch: [1][200/200]	Time 0.265 (0.358)	Data 0.000 (0.085)	Loss 2.423 (2.287)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.126 (0.133)	Data 0.030 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.21340274810791
==> Statistics for epoch 2: 543 clusters
Epoch: [2][20/200]	Time 0.269 (0.375)	Data 0.001 (0.102)	Loss 2.222 (0.851)
Epoch: [2][40/200]	Time 0.265 (0.360)	Data 0.001 (0.086)	Loss 2.678 (1.512)
Epoch: [2][60/200]	Time 0.265 (0.353)	Data 0.000 (0.082)	Loss 1.919 (1.761)
Epoch: [2][80/200]	Time 0.265 (0.353)	Data 0.000 (0.081)	Loss 2.391 (1.897)
Epoch: [2][100/200]	Time 0.271 (0.366)	Data 0.001 (0.094)	Loss 2.292 (1.984)
Epoch: [2][120/200]	Time 0.269 (0.363)	Data 0.001 (0.091)	Loss 2.214 (2.010)
Epoch: [2][140/200]	Time 0.268 (0.360)	Data 0.000 (0.088)	Loss 2.043 (2.035)
Epoch: [2][160/200]	Time 0.265 (0.360)	Data 0.000 (0.087)	Loss 2.093 (2.062)
Epoch: [2][180/200]	Time 0.273 (0.367)	Data 0.001 (0.094)	Loss 1.942 (2.070)
Epoch: [2][200/200]	Time 0.270 (0.365)	Data 0.001 (0.092)	Loss 1.688 (2.072)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.118 (0.134)	Data 0.022 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.677563190460205
==> Statistics for epoch 3: 558 clusters
Epoch: [3][20/200]	Time 0.264 (0.381)	Data 0.001 (0.111)	Loss 1.889 (0.634)
Epoch: [3][40/200]	Time 0.267 (0.364)	Data 0.001 (0.094)	Loss 2.147 (1.300)
Epoch: [3][60/200]	Time 0.271 (0.360)	Data 0.001 (0.088)	Loss 2.433 (1.563)
Epoch: [3][80/200]	Time 0.271 (0.355)	Data 0.000 (0.083)	Loss 1.953 (1.707)
Epoch: [3][100/200]	Time 0.267 (0.354)	Data 0.000 (0.081)	Loss 2.189 (1.774)
Epoch: [3][120/200]	Time 1.848 (0.367)	Data 1.564 (0.094)	Loss 1.882 (1.820)
Epoch: [3][140/200]	Time 0.274 (0.365)	Data 0.001 (0.091)	Loss 2.386 (1.840)
Epoch: [3][160/200]	Time 0.267 (0.362)	Data 0.001 (0.089)	Loss 2.020 (1.861)
Epoch: [3][180/200]	Time 0.264 (0.361)	Data 0.000 (0.088)	Loss 2.146 (1.866)
Epoch: [3][200/200]	Time 0.263 (0.360)	Data 0.000 (0.087)	Loss 1.614 (1.864)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.133)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 20.10986089706421
==> Statistics for epoch 4: 541 clusters
Epoch: [4][20/200]	Time 0.271 (0.393)	Data 0.001 (0.114)	Loss 2.022 (0.652)
Epoch: [4][40/200]	Time 0.269 (0.371)	Data 0.001 (0.095)	Loss 1.623 (1.227)
Epoch: [4][60/200]	Time 0.267 (0.364)	Data 0.000 (0.090)	Loss 1.328 (1.433)
Epoch: [4][80/200]	Time 0.270 (0.361)	Data 0.000 (0.087)	Loss 1.773 (1.557)
Epoch: [4][100/200]	Time 0.275 (0.375)	Data 0.001 (0.100)	Loss 2.042 (1.615)
Epoch: [4][120/200]	Time 0.393 (0.371)	Data 0.001 (0.095)	Loss 1.520 (1.658)
Epoch: [4][140/200]	Time 0.271 (0.367)	Data 0.000 (0.092)	Loss 1.610 (1.687)
Epoch: [4][160/200]	Time 0.269 (0.365)	Data 0.000 (0.090)	Loss 1.780 (1.696)
Epoch: [4][180/200]	Time 0.271 (0.372)	Data 0.001 (0.096)	Loss 1.929 (1.718)
Epoch: [4][200/200]	Time 0.265 (0.369)	Data 0.001 (0.094)	Loss 1.476 (1.723)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.135)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.285630464553833
==> Statistics for epoch 5: 553 clusters
Epoch: [5][20/200]	Time 0.264 (0.393)	Data 0.001 (0.115)	Loss 1.361 (0.515)
Epoch: [5][40/200]	Time 0.270 (0.367)	Data 0.001 (0.092)	Loss 1.674 (1.148)
Epoch: [5][60/200]	Time 0.269 (0.359)	Data 0.001 (0.086)	Loss 2.108 (1.342)
Epoch: [5][80/200]	Time 0.267 (0.354)	Data 0.000 (0.080)	Loss 1.540 (1.449)
Epoch: [5][100/200]	Time 0.266 (0.354)	Data 0.000 (0.079)	Loss 1.975 (1.514)
Epoch: [5][120/200]	Time 1.759 (0.366)	Data 1.444 (0.091)	Loss 1.716 (1.563)
Epoch: [5][140/200]	Time 0.266 (0.363)	Data 0.001 (0.088)	Loss 2.388 (1.599)
Epoch: [5][160/200]	Time 0.264 (0.360)	Data 0.000 (0.085)	Loss 2.000 (1.630)
Epoch: [5][180/200]	Time 0.267 (0.358)	Data 0.000 (0.083)	Loss 1.937 (1.639)
Epoch: [5][200/200]	Time 0.265 (0.357)	Data 0.000 (0.083)	Loss 1.693 (1.653)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.138)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.34160351753235
==> Statistics for epoch 6: 551 clusters
Epoch: [6][20/200]	Time 0.269 (0.387)	Data 0.001 (0.109)	Loss 1.736 (0.466)
Epoch: [6][40/200]	Time 0.271 (0.366)	Data 0.001 (0.093)	Loss 1.742 (1.032)
Epoch: [6][60/200]	Time 0.271 (0.362)	Data 0.001 (0.088)	Loss 1.797 (1.253)
Epoch: [6][80/200]	Time 0.267 (0.359)	Data 0.000 (0.085)	Loss 1.812 (1.365)
Epoch: [6][100/200]	Time 0.269 (0.356)	Data 0.000 (0.083)	Loss 2.171 (1.441)
Epoch: [6][120/200]	Time 1.849 (0.368)	Data 1.565 (0.094)	Loss 1.972 (1.464)
Epoch: [6][140/200]	Time 0.274 (0.366)	Data 0.001 (0.092)	Loss 1.561 (1.490)
Epoch: [6][160/200]	Time 0.269 (0.364)	Data 0.000 (0.090)	Loss 1.856 (1.512)
Epoch: [6][180/200]	Time 0.272 (0.363)	Data 0.000 (0.088)	Loss 1.530 (1.535)
Epoch: [6][200/200]	Time 0.269 (0.361)	Data 0.000 (0.087)	Loss 1.612 (1.546)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.142)	Data 0.000 (0.041)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.125882148742676
==> Statistics for epoch 7: 569 clusters
Epoch: [7][20/200]	Time 0.267 (0.381)	Data 0.001 (0.102)	Loss 1.571 (0.433)
Epoch: [7][40/200]	Time 0.266 (0.362)	Data 0.001 (0.086)	Loss 1.764 (0.960)
Epoch: [7][60/200]	Time 0.268 (0.356)	Data 0.001 (0.081)	Loss 1.969 (1.161)
Epoch: [7][80/200]	Time 0.269 (0.353)	Data 0.000 (0.079)	Loss 1.734 (1.266)
Epoch: [7][100/200]	Time 0.263 (0.352)	Data 0.000 (0.078)	Loss 1.729 (1.323)
Epoch: [7][120/200]	Time 1.820 (0.365)	Data 1.538 (0.090)	Loss 1.657 (1.360)
Epoch: [7][140/200]	Time 0.267 (0.361)	Data 0.001 (0.087)	Loss 1.769 (1.387)
Epoch: [7][160/200]	Time 0.270 (0.359)	Data 0.001 (0.085)	Loss 1.489 (1.418)
Epoch: [7][180/200]	Time 0.270 (0.358)	Data 0.001 (0.083)	Loss 1.618 (1.431)
Epoch: [7][200/200]	Time 0.266 (0.357)	Data 0.000 (0.083)	Loss 1.695 (1.444)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.119 (0.140)	Data 0.022 (0.039)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.88998055458069
==> Statistics for epoch 8: 575 clusters
Epoch: [8][20/200]	Time 0.274 (0.387)	Data 0.000 (0.107)	Loss 1.693 (0.440)
Epoch: [8][40/200]	Time 0.281 (0.365)	Data 0.001 (0.089)	Loss 1.621 (0.924)
Epoch: [8][60/200]	Time 0.283 (0.359)	Data 0.001 (0.084)	Loss 1.381 (1.125)
Epoch: [8][80/200]	Time 0.265 (0.356)	Data 0.000 (0.080)	Loss 1.640 (1.227)
Epoch: [8][100/200]	Time 0.266 (0.355)	Data 0.000 (0.079)	Loss 1.749 (1.307)
Epoch: [8][120/200]	Time 1.791 (0.366)	Data 1.475 (0.090)	Loss 1.539 (1.356)
Epoch: [8][140/200]	Time 0.270 (0.363)	Data 0.001 (0.088)	Loss 1.439 (1.365)
Epoch: [8][160/200]	Time 0.266 (0.361)	Data 0.001 (0.086)	Loss 1.230 (1.392)
Epoch: [8][180/200]	Time 0.268 (0.360)	Data 0.001 (0.085)	Loss 1.508 (1.409)
Epoch: [8][200/200]	Time 0.266 (0.359)	Data 0.000 (0.084)	Loss 1.436 (1.410)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.140)	Data 0.000 (0.038)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.246800422668457
==> Statistics for epoch 9: 575 clusters
Epoch: [9][20/200]	Time 0.274 (0.378)	Data 0.001 (0.104)	Loss 1.165 (0.379)
Epoch: [9][40/200]	Time 0.268 (0.366)	Data 0.001 (0.091)	Loss 1.626 (0.892)
Epoch: [9][60/200]	Time 0.282 (0.361)	Data 0.001 (0.085)	Loss 1.755 (1.083)
Epoch: [9][80/200]	Time 0.269 (0.359)	Data 0.000 (0.082)	Loss 1.886 (1.197)
Epoch: [9][100/200]	Time 0.265 (0.356)	Data 0.000 (0.081)	Loss 1.240 (1.252)
Epoch: [9][120/200]	Time 1.816 (0.369)	Data 1.534 (0.093)	Loss 1.454 (1.289)
Epoch: [9][140/200]	Time 0.281 (0.365)	Data 0.001 (0.090)	Loss 1.374 (1.311)
Epoch: [9][160/200]	Time 0.269 (0.364)	Data 0.001 (0.088)	Loss 1.510 (1.326)
Epoch: [9][180/200]	Time 0.268 (0.362)	Data 0.001 (0.086)	Loss 1.423 (1.335)
Epoch: [9][200/200]	Time 0.266 (0.361)	Data 0.000 (0.086)	Loss 1.610 (1.349)
Extract Features: [50/76]	Time 0.098 (0.136)	Data 0.000 (0.034)	
Mean AP: 87.3%

 * Finished epoch   9  model mAP: 87.3%  best: 87.3% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.137)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.10030961036682
==> Statistics for epoch 10: 566 clusters
Epoch: [10][20/200]	Time 0.271 (0.414)	Data 0.001 (0.113)	Loss 1.907 (0.410)
Epoch: [10][40/200]	Time 0.268 (0.378)	Data 0.001 (0.093)	Loss 1.199 (0.860)
Epoch: [10][60/200]	Time 0.275 (0.369)	Data 0.001 (0.087)	Loss 1.048 (1.025)
Epoch: [10][80/200]	Time 0.267 (0.365)	Data 0.000 (0.085)	Loss 1.402 (1.113)
Epoch: [10][100/200]	Time 0.274 (0.362)	Data 0.000 (0.082)	Loss 1.145 (1.135)
Epoch: [10][120/200]	Time 1.799 (0.371)	Data 1.499 (0.093)	Loss 1.562 (1.174)
Epoch: [10][140/200]	Time 0.270 (0.368)	Data 0.001 (0.090)	Loss 1.472 (1.201)
Epoch: [10][160/200]	Time 0.268 (0.366)	Data 0.000 (0.088)	Loss 1.369 (1.227)
Epoch: [10][180/200]	Time 0.266 (0.363)	Data 0.001 (0.086)	Loss 0.962 (1.229)
Epoch: [10][200/200]	Time 0.271 (0.361)	Data 0.000 (0.085)	Loss 0.964 (1.245)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.136)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.870290994644165
==> Statistics for epoch 11: 577 clusters
Epoch: [11][20/200]	Time 0.266 (0.375)	Data 0.001 (0.103)	Loss 1.343 (0.309)
Epoch: [11][40/200]	Time 0.268 (0.360)	Data 0.001 (0.087)	Loss 1.725 (0.769)
Epoch: [11][60/200]	Time 0.266 (0.354)	Data 0.001 (0.081)	Loss 1.704 (0.958)
Epoch: [11][80/200]	Time 0.268 (0.353)	Data 0.001 (0.081)	Loss 1.363 (1.051)
Epoch: [11][100/200]	Time 0.266 (0.352)	Data 0.001 (0.079)	Loss 1.524 (1.108)
Epoch: [11][120/200]	Time 0.269 (0.352)	Data 0.000 (0.079)	Loss 1.082 (1.144)
Epoch: [11][140/200]	Time 0.265 (0.351)	Data 0.000 (0.077)	Loss 1.310 (1.169)
Epoch: [11][160/200]	Time 0.280 (0.349)	Data 0.002 (0.075)	Loss 0.841 (1.182)
Epoch: [11][180/200]	Time 0.268 (0.350)	Data 0.000 (0.076)	Loss 1.265 (1.191)
Epoch: [11][200/200]	Time 0.277 (0.358)	Data 0.000 (0.083)	Loss 1.397 (1.198)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.24318838119507
==> Statistics for epoch 12: 581 clusters
Epoch: [12][20/200]	Time 0.269 (0.390)	Data 0.001 (0.114)	Loss 1.327 (0.315)
Epoch: [12][40/200]	Time 0.284 (0.368)	Data 0.001 (0.094)	Loss 1.036 (0.784)
Epoch: [12][60/200]	Time 0.264 (0.362)	Data 0.001 (0.086)	Loss 1.362 (0.932)
Epoch: [12][80/200]	Time 0.269 (0.358)	Data 0.001 (0.084)	Loss 1.466 (1.010)
Epoch: [12][100/200]	Time 0.268 (0.356)	Data 0.001 (0.082)	Loss 1.093 (1.051)
Epoch: [12][120/200]	Time 0.267 (0.356)	Data 0.000 (0.081)	Loss 1.003 (1.089)
Epoch: [12][140/200]	Time 0.274 (0.355)	Data 0.000 (0.081)	Loss 1.548 (1.116)
Epoch: [12][160/200]	Time 0.267 (0.354)	Data 0.000 (0.080)	Loss 1.135 (1.134)
Epoch: [12][180/200]	Time 0.268 (0.354)	Data 0.000 (0.080)	Loss 1.117 (1.155)
Epoch: [12][200/200]	Time 0.273 (0.362)	Data 0.001 (0.087)	Loss 1.024 (1.163)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.139)	Data 0.000 (0.039)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.046467542648315
==> Statistics for epoch 13: 581 clusters
Epoch: [13][20/200]	Time 0.274 (0.381)	Data 0.001 (0.100)	Loss 1.395 (0.298)
Epoch: [13][40/200]	Time 0.268 (0.367)	Data 0.001 (0.089)	Loss 1.362 (0.744)
Epoch: [13][60/200]	Time 0.270 (0.361)	Data 0.001 (0.084)	Loss 1.333 (0.923)
Epoch: [13][80/200]	Time 0.271 (0.357)	Data 0.001 (0.081)	Loss 0.909 (1.015)
Epoch: [13][100/200]	Time 0.270 (0.354)	Data 0.001 (0.078)	Loss 1.241 (1.062)
Epoch: [13][120/200]	Time 0.266 (0.352)	Data 0.000 (0.077)	Loss 0.811 (1.085)
Epoch: [13][140/200]	Time 0.369 (0.353)	Data 0.000 (0.077)	Loss 1.353 (1.108)
Epoch: [13][160/200]	Time 0.268 (0.351)	Data 0.000 (0.077)	Loss 1.242 (1.116)
Epoch: [13][180/200]	Time 0.268 (0.351)	Data 0.000 (0.076)	Loss 1.453 (1.126)
Epoch: [13][200/200]	Time 0.366 (0.358)	Data 0.000 (0.083)	Loss 1.346 (1.137)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.135)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.040977954864502
==> Statistics for epoch 14: 577 clusters
Epoch: [14][20/200]	Time 0.271 (0.385)	Data 0.001 (0.112)	Loss 0.935 (0.290)
Epoch: [14][40/200]	Time 0.273 (0.368)	Data 0.001 (0.092)	Loss 1.055 (0.741)
Epoch: [14][60/200]	Time 0.266 (0.362)	Data 0.001 (0.086)	Loss 1.108 (0.892)
Epoch: [14][80/200]	Time 0.267 (0.358)	Data 0.002 (0.082)	Loss 1.636 (0.971)
Epoch: [14][100/200]	Time 0.271 (0.356)	Data 0.001 (0.081)	Loss 1.062 (1.027)
Epoch: [14][120/200]	Time 0.266 (0.357)	Data 0.000 (0.081)	Loss 1.008 (1.048)
Epoch: [14][140/200]	Time 0.268 (0.355)	Data 0.000 (0.080)	Loss 0.839 (1.069)
Epoch: [14][160/200]	Time 0.268 (0.354)	Data 0.000 (0.079)	Loss 1.623 (1.074)
Epoch: [14][180/200]	Time 0.270 (0.354)	Data 0.000 (0.079)	Loss 1.400 (1.089)
Epoch: [14][200/200]	Time 0.269 (0.361)	Data 0.001 (0.086)	Loss 1.159 (1.098)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.140)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.72566246986389
==> Statistics for epoch 15: 578 clusters
Epoch: [15][20/200]	Time 0.268 (0.393)	Data 0.001 (0.118)	Loss 1.268 (0.296)
Epoch: [15][40/200]	Time 0.267 (0.372)	Data 0.001 (0.098)	Loss 1.433 (0.683)
Epoch: [15][60/200]	Time 0.265 (0.364)	Data 0.000 (0.091)	Loss 1.046 (0.825)
Epoch: [15][80/200]	Time 0.267 (0.360)	Data 0.001 (0.088)	Loss 0.904 (0.899)
Epoch: [15][100/200]	Time 0.269 (0.359)	Data 0.001 (0.086)	Loss 1.065 (0.946)
Epoch: [15][120/200]	Time 0.267 (0.357)	Data 0.000 (0.084)	Loss 0.978 (0.971)
Epoch: [15][140/200]	Time 0.268 (0.355)	Data 0.000 (0.082)	Loss 0.833 (1.001)
Epoch: [15][160/200]	Time 0.268 (0.354)	Data 0.000 (0.081)	Loss 1.164 (1.021)
Epoch: [15][180/200]	Time 0.263 (0.354)	Data 0.000 (0.080)	Loss 1.221 (1.037)
Epoch: [15][200/200]	Time 0.269 (0.361)	Data 0.001 (0.087)	Loss 1.047 (1.046)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.131)	Data 0.000 (0.027)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.852370977401733
==> Statistics for epoch 16: 586 clusters
Epoch: [16][20/200]	Time 0.270 (0.378)	Data 0.001 (0.107)	Loss 0.997 (0.234)
Epoch: [16][40/200]	Time 0.267 (0.360)	Data 0.001 (0.089)	Loss 1.020 (0.661)
Epoch: [16][60/200]	Time 0.271 (0.357)	Data 0.001 (0.085)	Loss 1.359 (0.801)
Epoch: [16][80/200]	Time 0.277 (0.354)	Data 0.001 (0.081)	Loss 0.963 (0.866)
Epoch: [16][100/200]	Time 0.263 (0.353)	Data 0.001 (0.080)	Loss 1.188 (0.909)
Epoch: [16][120/200]	Time 0.266 (0.352)	Data 0.000 (0.078)	Loss 0.967 (0.936)
Epoch: [16][140/200]	Time 0.263 (0.352)	Data 0.000 (0.078)	Loss 1.238 (0.958)
Epoch: [16][160/200]	Time 0.266 (0.351)	Data 0.000 (0.077)	Loss 0.672 (0.980)
Epoch: [16][180/200]	Time 0.264 (0.351)	Data 0.000 (0.077)	Loss 1.172 (0.993)
Epoch: [16][200/200]	Time 0.271 (0.358)	Data 0.001 (0.084)	Loss 1.356 (1.001)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.136)	Data 0.000 (0.036)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.002779960632324
==> Statistics for epoch 17: 583 clusters
Epoch: [17][20/200]	Time 0.268 (0.388)	Data 0.001 (0.109)	Loss 1.168 (0.246)
Epoch: [17][40/200]	Time 0.268 (0.371)	Data 0.001 (0.094)	Loss 0.859 (0.639)
Epoch: [17][60/200]	Time 0.270 (0.365)	Data 0.001 (0.088)	Loss 1.151 (0.811)
Epoch: [17][80/200]	Time 0.267 (0.360)	Data 0.001 (0.085)	Loss 1.500 (0.896)
Epoch: [17][100/200]	Time 0.270 (0.357)	Data 0.001 (0.082)	Loss 1.205 (0.934)
Epoch: [17][120/200]	Time 0.265 (0.357)	Data 0.000 (0.082)	Loss 0.896 (0.957)
Epoch: [17][140/200]	Time 0.266 (0.356)	Data 0.000 (0.081)	Loss 1.141 (0.978)
Epoch: [17][160/200]	Time 0.267 (0.355)	Data 0.000 (0.081)	Loss 0.807 (0.994)
Epoch: [17][180/200]	Time 0.266 (0.355)	Data 0.000 (0.081)	Loss 1.308 (1.004)
Epoch: [17][200/200]	Time 0.275 (0.362)	Data 0.001 (0.087)	Loss 0.856 (1.005)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.136)	Data 0.000 (0.036)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.80585527420044
==> Statistics for epoch 18: 586 clusters
Epoch: [18][20/200]	Time 0.269 (0.389)	Data 0.001 (0.111)	Loss 0.996 (0.208)
Epoch: [18][40/200]	Time 0.268 (0.372)	Data 0.001 (0.094)	Loss 1.282 (0.591)
Epoch: [18][60/200]	Time 0.269 (0.363)	Data 0.001 (0.087)	Loss 0.941 (0.744)
Epoch: [18][80/200]	Time 0.269 (0.359)	Data 0.001 (0.083)	Loss 0.855 (0.831)
Epoch: [18][100/200]	Time 0.271 (0.357)	Data 0.001 (0.081)	Loss 0.642 (0.868)
Epoch: [18][120/200]	Time 0.265 (0.357)	Data 0.000 (0.081)	Loss 0.690 (0.897)
Epoch: [18][140/200]	Time 0.267 (0.355)	Data 0.000 (0.080)	Loss 1.159 (0.913)
Epoch: [18][160/200]	Time 0.267 (0.354)	Data 0.000 (0.079)	Loss 0.990 (0.932)
Epoch: [18][180/200]	Time 0.268 (0.353)	Data 0.000 (0.078)	Loss 1.164 (0.938)
Epoch: [18][200/200]	Time 0.280 (0.361)	Data 0.001 (0.086)	Loss 0.947 (0.943)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.136)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.28657603263855
==> Statistics for epoch 19: 582 clusters
Epoch: [19][20/200]	Time 0.271 (0.378)	Data 0.001 (0.097)	Loss 1.040 (0.237)
Epoch: [19][40/200]	Time 0.268 (0.359)	Data 0.001 (0.082)	Loss 1.050 (0.620)
Epoch: [19][60/200]	Time 0.265 (0.352)	Data 0.001 (0.076)	Loss 0.810 (0.759)
Epoch: [19][80/200]	Time 0.267 (0.347)	Data 0.001 (0.072)	Loss 0.872 (0.831)
Epoch: [19][100/200]	Time 0.264 (0.348)	Data 0.000 (0.073)	Loss 1.008 (0.873)
Epoch: [19][120/200]	Time 0.269 (0.347)	Data 0.000 (0.073)	Loss 1.060 (0.897)
Epoch: [19][140/200]	Time 0.264 (0.347)	Data 0.000 (0.073)	Loss 0.876 (0.921)
Epoch: [19][160/200]	Time 0.269 (0.347)	Data 0.000 (0.073)	Loss 0.851 (0.927)
Epoch: [19][180/200]	Time 0.270 (0.348)	Data 0.000 (0.073)	Loss 0.682 (0.932)
Epoch: [19][200/200]	Time 0.280 (0.356)	Data 0.000 (0.081)	Loss 1.061 (0.942)
Extract Features: [50/76]	Time 0.098 (0.133)	Data 0.000 (0.031)	
Mean AP: 89.8%

 * Finished epoch  19  model mAP: 89.8%  best: 89.8% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.135)	Data 0.041 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.217947006225586
==> Statistics for epoch 20: 582 clusters
Epoch: [20][20/200]	Time 0.344 (0.381)	Data 0.001 (0.108)	Loss 0.832 (0.211)
Epoch: [20][40/200]	Time 0.268 (0.364)	Data 0.001 (0.093)	Loss 0.942 (0.529)
Epoch: [20][60/200]	Time 0.267 (0.359)	Data 0.001 (0.087)	Loss 1.088 (0.681)
Epoch: [20][80/200]	Time 0.273 (0.359)	Data 0.001 (0.087)	Loss 0.949 (0.759)
Epoch: [20][100/200]	Time 0.268 (0.356)	Data 0.001 (0.084)	Loss 1.073 (0.815)
Epoch: [20][120/200]	Time 0.268 (0.355)	Data 0.000 (0.083)	Loss 1.051 (0.838)
Epoch: [20][140/200]	Time 0.266 (0.355)	Data 0.000 (0.082)	Loss 1.037 (0.853)
Epoch: [20][160/200]	Time 0.268 (0.354)	Data 0.000 (0.081)	Loss 0.940 (0.861)
Epoch: [20][180/200]	Time 0.261 (0.353)	Data 0.000 (0.081)	Loss 0.965 (0.880)
Epoch: [20][200/200]	Time 0.277 (0.360)	Data 0.001 (0.087)	Loss 0.848 (0.886)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.135)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.496294021606445
==> Statistics for epoch 21: 585 clusters
Epoch: [21][20/200]	Time 0.275 (0.386)	Data 0.000 (0.111)	Loss 1.371 (0.233)
Epoch: [21][40/200]	Time 0.268 (0.368)	Data 0.001 (0.094)	Loss 0.903 (0.568)
Epoch: [21][60/200]	Time 0.267 (0.359)	Data 0.001 (0.087)	Loss 0.764 (0.687)
Epoch: [21][80/200]	Time 0.268 (0.356)	Data 0.001 (0.083)	Loss 0.712 (0.740)
Epoch: [21][100/200]	Time 0.264 (0.354)	Data 0.001 (0.081)	Loss 0.786 (0.772)
Epoch: [21][120/200]	Time 0.267 (0.354)	Data 0.000 (0.080)	Loss 1.010 (0.808)
Epoch: [21][140/200]	Time 0.291 (0.353)	Data 0.000 (0.079)	Loss 1.072 (0.828)
Epoch: [21][160/200]	Time 0.265 (0.352)	Data 0.000 (0.079)	Loss 0.578 (0.836)
Epoch: [21][180/200]	Time 0.273 (0.352)	Data 0.000 (0.079)	Loss 1.040 (0.846)
Epoch: [21][200/200]	Time 0.269 (0.358)	Data 0.001 (0.085)	Loss 0.788 (0.853)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.133)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.993584156036377
==> Statistics for epoch 22: 581 clusters
Epoch: [22][20/200]	Time 0.267 (0.370)	Data 0.001 (0.101)	Loss 0.864 (0.187)
Epoch: [22][40/200]	Time 0.269 (0.353)	Data 0.001 (0.082)	Loss 0.846 (0.542)
Epoch: [22][60/200]	Time 0.264 (0.349)	Data 0.001 (0.078)	Loss 0.908 (0.666)
Epoch: [22][80/200]	Time 0.264 (0.348)	Data 0.001 (0.077)	Loss 1.110 (0.732)
Epoch: [22][100/200]	Time 0.265 (0.346)	Data 0.001 (0.075)	Loss 0.763 (0.759)
Epoch: [22][120/200]	Time 0.264 (0.345)	Data 0.000 (0.073)	Loss 0.870 (0.784)
Epoch: [22][140/200]	Time 0.263 (0.344)	Data 0.000 (0.072)	Loss 0.964 (0.812)
Epoch: [22][160/200]	Time 0.267 (0.344)	Data 0.000 (0.072)	Loss 1.261 (0.828)
Epoch: [22][180/200]	Time 0.266 (0.343)	Data 0.000 (0.071)	Loss 0.794 (0.838)
Epoch: [22][200/200]	Time 0.265 (0.349)	Data 0.001 (0.077)	Loss 0.973 (0.847)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.138)	Data 0.000 (0.037)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.587052822113037
==> Statistics for epoch 23: 586 clusters
Epoch: [23][20/200]	Time 0.280 (0.385)	Data 0.001 (0.110)	Loss 0.748 (0.202)
Epoch: [23][40/200]	Time 0.267 (0.362)	Data 0.001 (0.090)	Loss 0.959 (0.532)
Epoch: [23][60/200]	Time 0.278 (0.353)	Data 0.001 (0.083)	Loss 0.970 (0.653)
Epoch: [23][80/200]	Time 0.265 (0.349)	Data 0.001 (0.078)	Loss 1.396 (0.728)
Epoch: [23][100/200]	Time 0.266 (0.348)	Data 0.001 (0.075)	Loss 0.836 (0.769)
Epoch: [23][120/200]	Time 0.267 (0.346)	Data 0.000 (0.073)	Loss 0.773 (0.800)
Epoch: [23][140/200]	Time 0.265 (0.345)	Data 0.000 (0.073)	Loss 1.170 (0.819)
Epoch: [23][160/200]	Time 0.267 (0.345)	Data 0.000 (0.072)	Loss 0.902 (0.829)
Epoch: [23][180/200]	Time 0.354 (0.344)	Data 0.000 (0.071)	Loss 1.310 (0.834)
Epoch: [23][200/200]	Time 0.278 (0.350)	Data 0.001 (0.077)	Loss 0.662 (0.844)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.131)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.9069561958313
==> Statistics for epoch 24: 588 clusters
Epoch: [24][20/200]	Time 0.267 (0.371)	Data 0.001 (0.098)	Loss 0.684 (0.187)
Epoch: [24][40/200]	Time 0.278 (0.354)	Data 0.001 (0.082)	Loss 1.081 (0.548)
Epoch: [24][60/200]	Time 0.262 (0.350)	Data 0.001 (0.079)	Loss 0.881 (0.683)
Epoch: [24][80/200]	Time 0.265 (0.347)	Data 0.001 (0.074)	Loss 0.908 (0.772)
Epoch: [24][100/200]	Time 0.263 (0.345)	Data 0.001 (0.073)	Loss 0.932 (0.803)
Epoch: [24][120/200]	Time 0.267 (0.343)	Data 0.000 (0.071)	Loss 1.265 (0.823)
Epoch: [24][140/200]	Time 0.267 (0.343)	Data 0.000 (0.071)	Loss 0.965 (0.841)
Epoch: [24][160/200]	Time 0.266 (0.343)	Data 0.000 (0.071)	Loss 1.001 (0.846)
Epoch: [24][180/200]	Time 0.356 (0.342)	Data 0.000 (0.070)	Loss 0.908 (0.849)
Epoch: [24][200/200]	Time 0.269 (0.347)	Data 0.001 (0.075)	Loss 0.686 (0.854)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.130)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.746548414230347
==> Statistics for epoch 25: 585 clusters
Epoch: [25][20/200]	Time 0.265 (0.378)	Data 0.001 (0.103)	Loss 0.925 (0.195)
Epoch: [25][40/200]	Time 0.266 (0.357)	Data 0.001 (0.084)	Loss 0.992 (0.503)
Epoch: [25][60/200]	Time 0.270 (0.349)	Data 0.001 (0.076)	Loss 1.268 (0.657)
Epoch: [25][80/200]	Time 0.268 (0.346)	Data 0.001 (0.075)	Loss 0.842 (0.728)
Epoch: [25][100/200]	Time 0.265 (0.345)	Data 0.001 (0.073)	Loss 1.009 (0.764)
Epoch: [25][120/200]	Time 0.267 (0.345)	Data 0.000 (0.073)	Loss 0.863 (0.787)
Epoch: [25][140/200]	Time 0.263 (0.343)	Data 0.000 (0.072)	Loss 0.871 (0.803)
Epoch: [25][160/200]	Time 0.267 (0.343)	Data 0.000 (0.071)	Loss 0.790 (0.812)
Epoch: [25][180/200]	Time 0.265 (0.342)	Data 0.000 (0.071)	Loss 0.829 (0.823)
Epoch: [25][200/200]	Time 0.269 (0.351)	Data 0.001 (0.079)	Loss 0.725 (0.835)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.134)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.888423919677734
==> Statistics for epoch 26: 586 clusters
Epoch: [26][20/200]	Time 0.274 (0.377)	Data 0.001 (0.099)	Loss 0.841 (0.194)
Epoch: [26][40/200]	Time 0.266 (0.362)	Data 0.001 (0.089)	Loss 0.808 (0.514)
Epoch: [26][60/200]	Time 0.276 (0.359)	Data 0.001 (0.085)	Loss 0.643 (0.653)
Epoch: [26][80/200]	Time 0.268 (0.354)	Data 0.001 (0.081)	Loss 0.892 (0.713)
Epoch: [26][100/200]	Time 0.270 (0.351)	Data 0.001 (0.078)	Loss 1.417 (0.755)
Epoch: [26][120/200]	Time 0.266 (0.349)	Data 0.000 (0.076)	Loss 0.725 (0.770)
Epoch: [26][140/200]	Time 0.266 (0.347)	Data 0.000 (0.074)	Loss 0.822 (0.779)
Epoch: [26][160/200]	Time 0.266 (0.346)	Data 0.000 (0.074)	Loss 0.815 (0.798)
Epoch: [26][180/200]	Time 0.265 (0.345)	Data 0.000 (0.073)	Loss 1.508 (0.816)
Epoch: [26][200/200]	Time 0.271 (0.351)	Data 0.001 (0.078)	Loss 1.023 (0.829)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.135)	Data 0.000 (0.033)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.712029695510864
==> Statistics for epoch 27: 585 clusters
Epoch: [27][20/200]	Time 0.270 (0.369)	Data 0.001 (0.099)	Loss 0.605 (0.165)
Epoch: [27][40/200]	Time 0.266 (0.352)	Data 0.000 (0.081)	Loss 0.970 (0.526)
Epoch: [27][60/200]	Time 0.267 (0.344)	Data 0.001 (0.074)	Loss 0.991 (0.653)
Epoch: [27][80/200]	Time 0.360 (0.343)	Data 0.001 (0.073)	Loss 0.918 (0.730)
Epoch: [27][100/200]	Time 0.266 (0.341)	Data 0.001 (0.071)	Loss 0.715 (0.752)
Epoch: [27][120/200]	Time 0.267 (0.343)	Data 0.000 (0.072)	Loss 0.881 (0.781)
Epoch: [27][140/200]	Time 0.269 (0.344)	Data 0.000 (0.072)	Loss 0.783 (0.810)
Epoch: [27][160/200]	Time 0.266 (0.343)	Data 0.000 (0.072)	Loss 0.690 (0.824)
Epoch: [27][180/200]	Time 0.268 (0.343)	Data 0.000 (0.072)	Loss 0.930 (0.833)
Epoch: [27][200/200]	Time 0.267 (0.349)	Data 0.000 (0.078)	Loss 0.588 (0.836)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.131)	Data 0.000 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.000176906585693
==> Statistics for epoch 28: 586 clusters
Epoch: [28][20/200]	Time 0.264 (0.371)	Data 0.001 (0.094)	Loss 0.971 (0.208)
Epoch: [28][40/200]	Time 0.268 (0.356)	Data 0.001 (0.082)	Loss 1.123 (0.542)
Epoch: [28][60/200]	Time 0.268 (0.351)	Data 0.001 (0.078)	Loss 0.965 (0.662)
Epoch: [28][80/200]	Time 0.267 (0.347)	Data 0.001 (0.075)	Loss 0.771 (0.706)
Epoch: [28][100/200]	Time 0.262 (0.345)	Data 0.001 (0.073)	Loss 1.298 (0.750)
Epoch: [28][120/200]	Time 0.265 (0.345)	Data 0.000 (0.073)	Loss 0.861 (0.775)
Epoch: [28][140/200]	Time 0.267 (0.345)	Data 0.000 (0.073)	Loss 0.919 (0.790)
Epoch: [28][160/200]	Time 0.358 (0.344)	Data 0.000 (0.072)	Loss 1.019 (0.804)
Epoch: [28][180/200]	Time 0.263 (0.342)	Data 0.000 (0.071)	Loss 0.734 (0.812)
Epoch: [28][200/200]	Time 0.267 (0.350)	Data 0.001 (0.078)	Loss 0.942 (0.819)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.114 (0.131)	Data 0.019 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.27605414390564
==> Statistics for epoch 29: 586 clusters
Epoch: [29][20/200]	Time 0.267 (0.380)	Data 0.001 (0.103)	Loss 0.787 (0.195)
Epoch: [29][40/200]	Time 0.289 (0.360)	Data 0.001 (0.085)	Loss 0.539 (0.526)
Epoch: [29][60/200]	Time 0.264 (0.353)	Data 0.001 (0.078)	Loss 1.046 (0.659)
Epoch: [29][80/200]	Time 0.266 (0.348)	Data 0.001 (0.074)	Loss 0.856 (0.707)
Epoch: [29][100/200]	Time 0.268 (0.347)	Data 0.001 (0.073)	Loss 0.842 (0.747)
Epoch: [29][120/200]	Time 0.265 (0.345)	Data 0.000 (0.072)	Loss 1.029 (0.768)
Epoch: [29][140/200]	Time 0.264 (0.344)	Data 0.000 (0.071)	Loss 0.680 (0.782)
Epoch: [29][160/200]	Time 0.268 (0.342)	Data 0.000 (0.070)	Loss 0.737 (0.787)
Epoch: [29][180/200]	Time 0.264 (0.342)	Data 0.000 (0.070)	Loss 0.663 (0.798)
Epoch: [29][200/200]	Time 0.278 (0.348)	Data 0.001 (0.076)	Loss 0.906 (0.804)
Extract Features: [50/76]	Time 0.096 (0.135)	Data 0.000 (0.033)	
Mean AP: 90.5%

 * Finished epoch  29  model mAP: 90.5%  best: 90.5% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.130)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.072445154190063
==> Statistics for epoch 30: 586 clusters
Epoch: [30][20/200]	Time 0.271 (0.371)	Data 0.001 (0.102)	Loss 0.676 (0.173)
Epoch: [30][40/200]	Time 0.273 (0.358)	Data 0.001 (0.086)	Loss 0.899 (0.514)
Epoch: [30][60/200]	Time 0.268 (0.353)	Data 0.001 (0.081)	Loss 0.672 (0.629)
Epoch: [30][80/200]	Time 0.267 (0.350)	Data 0.001 (0.077)	Loss 0.720 (0.699)
Epoch: [30][100/200]	Time 0.268 (0.347)	Data 0.001 (0.075)	Loss 0.627 (0.731)
Epoch: [30][120/200]	Time 0.263 (0.345)	Data 0.000 (0.073)	Loss 0.947 (0.752)
Epoch: [30][140/200]	Time 0.268 (0.343)	Data 0.000 (0.071)	Loss 1.086 (0.782)
Epoch: [30][160/200]	Time 0.268 (0.342)	Data 0.000 (0.070)	Loss 1.364 (0.791)
Epoch: [30][180/200]	Time 0.269 (0.342)	Data 0.000 (0.070)	Loss 0.913 (0.799)
Epoch: [30][200/200]	Time 0.265 (0.349)	Data 0.001 (0.077)	Loss 1.425 (0.813)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.138 (0.134)	Data 0.041 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.04198932647705
==> Statistics for epoch 31: 585 clusters
Epoch: [31][20/200]	Time 0.268 (0.387)	Data 0.001 (0.109)	Loss 0.717 (0.198)
Epoch: [31][40/200]	Time 0.271 (0.360)	Data 0.001 (0.087)	Loss 0.812 (0.532)
Epoch: [31][60/200]	Time 0.268 (0.353)	Data 0.001 (0.079)	Loss 0.734 (0.653)
Epoch: [31][80/200]	Time 0.269 (0.350)	Data 0.001 (0.076)	Loss 0.990 (0.719)
Epoch: [31][100/200]	Time 0.269 (0.347)	Data 0.001 (0.074)	Loss 1.017 (0.763)
Epoch: [31][120/200]	Time 0.271 (0.347)	Data 0.000 (0.073)	Loss 0.647 (0.782)
Epoch: [31][140/200]	Time 0.267 (0.346)	Data 0.000 (0.072)	Loss 0.664 (0.790)
Epoch: [31][160/200]	Time 0.262 (0.345)	Data 0.000 (0.071)	Loss 1.007 (0.799)
Epoch: [31][180/200]	Time 0.266 (0.344)	Data 0.000 (0.071)	Loss 1.151 (0.803)
Epoch: [31][200/200]	Time 0.266 (0.350)	Data 0.001 (0.076)	Loss 0.647 (0.810)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.134)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.540706872940063
==> Statistics for epoch 32: 584 clusters
Epoch: [32][20/200]	Time 0.268 (0.379)	Data 0.001 (0.104)	Loss 0.921 (0.190)
Epoch: [32][40/200]	Time 0.378 (0.361)	Data 0.001 (0.087)	Loss 0.855 (0.500)
Epoch: [32][60/200]	Time 0.267 (0.352)	Data 0.001 (0.081)	Loss 1.066 (0.624)
Epoch: [32][80/200]	Time 0.269 (0.348)	Data 0.001 (0.077)	Loss 0.969 (0.697)
Epoch: [32][100/200]	Time 0.381 (0.346)	Data 0.001 (0.074)	Loss 0.821 (0.722)
Epoch: [32][120/200]	Time 0.265 (0.343)	Data 0.000 (0.071)	Loss 1.127 (0.751)
Epoch: [32][140/200]	Time 0.266 (0.342)	Data 0.000 (0.071)	Loss 0.869 (0.765)
Epoch: [32][160/200]	Time 0.264 (0.342)	Data 0.000 (0.070)	Loss 0.869 (0.781)
Epoch: [32][180/200]	Time 0.262 (0.341)	Data 0.000 (0.070)	Loss 0.916 (0.786)
Epoch: [32][200/200]	Time 0.266 (0.347)	Data 0.001 (0.075)	Loss 0.673 (0.796)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.131)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.068936347961426
==> Statistics for epoch 33: 586 clusters
Epoch: [33][20/200]	Time 0.275 (0.375)	Data 0.001 (0.099)	Loss 0.837 (0.182)
Epoch: [33][40/200]	Time 0.267 (0.355)	Data 0.001 (0.081)	Loss 0.931 (0.478)
Epoch: [33][60/200]	Time 0.265 (0.349)	Data 0.001 (0.075)	Loss 1.144 (0.610)
Epoch: [33][80/200]	Time 0.268 (0.345)	Data 0.001 (0.073)	Loss 0.794 (0.672)
Epoch: [33][100/200]	Time 0.269 (0.344)	Data 0.001 (0.073)	Loss 0.910 (0.716)
Epoch: [33][120/200]	Time 0.268 (0.343)	Data 0.000 (0.071)	Loss 0.844 (0.739)
Epoch: [33][140/200]	Time 0.262 (0.343)	Data 0.000 (0.071)	Loss 0.956 (0.759)
Epoch: [33][160/200]	Time 0.265 (0.342)	Data 0.000 (0.071)	Loss 0.674 (0.767)
Epoch: [33][180/200]	Time 0.264 (0.342)	Data 0.000 (0.070)	Loss 0.790 (0.771)
Epoch: [33][200/200]	Time 0.266 (0.348)	Data 0.001 (0.076)	Loss 0.873 (0.786)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.136)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.168987274169922
==> Statistics for epoch 34: 584 clusters
Epoch: [34][20/200]	Time 0.276 (0.380)	Data 0.001 (0.105)	Loss 0.961 (0.207)
Epoch: [34][40/200]	Time 0.269 (0.359)	Data 0.001 (0.087)	Loss 0.985 (0.543)
Epoch: [34][60/200]	Time 0.265 (0.352)	Data 0.001 (0.081)	Loss 0.698 (0.634)
Epoch: [34][80/200]	Time 0.268 (0.349)	Data 0.001 (0.077)	Loss 1.046 (0.692)
Epoch: [34][100/200]	Time 0.272 (0.348)	Data 0.001 (0.076)	Loss 0.860 (0.724)
Epoch: [34][120/200]	Time 0.268 (0.348)	Data 0.000 (0.074)	Loss 0.897 (0.744)
Epoch: [34][140/200]	Time 0.270 (0.347)	Data 0.000 (0.073)	Loss 0.725 (0.761)
Epoch: [34][160/200]	Time 0.268 (0.346)	Data 0.000 (0.072)	Loss 0.717 (0.774)
Epoch: [34][180/200]	Time 0.266 (0.345)	Data 0.000 (0.071)	Loss 0.762 (0.787)
Epoch: [34][200/200]	Time 0.286 (0.351)	Data 0.001 (0.077)	Loss 0.771 (0.794)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.131)	Data 0.000 (0.028)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.729604482650757
==> Statistics for epoch 35: 586 clusters
Epoch: [35][20/200]	Time 0.272 (0.382)	Data 0.001 (0.108)	Loss 0.951 (0.195)
Epoch: [35][40/200]	Time 0.268 (0.365)	Data 0.001 (0.094)	Loss 0.787 (0.510)
Epoch: [35][60/200]	Time 0.267 (0.356)	Data 0.001 (0.086)	Loss 0.850 (0.642)
Epoch: [35][80/200]	Time 0.266 (0.352)	Data 0.001 (0.081)	Loss 0.804 (0.703)
Epoch: [35][100/200]	Time 0.265 (0.350)	Data 0.001 (0.078)	Loss 0.936 (0.735)
Epoch: [35][120/200]	Time 0.268 (0.348)	Data 0.000 (0.075)	Loss 0.944 (0.749)
Epoch: [35][140/200]	Time 0.266 (0.346)	Data 0.000 (0.075)	Loss 0.995 (0.762)
Epoch: [35][160/200]	Time 0.265 (0.346)	Data 0.000 (0.074)	Loss 1.075 (0.778)
Epoch: [35][180/200]	Time 0.268 (0.344)	Data 0.002 (0.072)	Loss 0.689 (0.785)
Epoch: [35][200/200]	Time 0.273 (0.350)	Data 0.001 (0.079)	Loss 0.857 (0.799)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.132)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.990747213363647
==> Statistics for epoch 36: 586 clusters
Epoch: [36][20/200]	Time 0.265 (0.392)	Data 0.001 (0.113)	Loss 0.557 (0.173)
Epoch: [36][40/200]	Time 0.270 (0.372)	Data 0.001 (0.095)	Loss 0.792 (0.498)
Epoch: [36][60/200]	Time 0.263 (0.363)	Data 0.001 (0.087)	Loss 0.617 (0.626)
Epoch: [36][80/200]	Time 0.267 (0.358)	Data 0.001 (0.084)	Loss 0.851 (0.693)
Epoch: [36][100/200]	Time 0.267 (0.356)	Data 0.001 (0.082)	Loss 0.908 (0.722)
Epoch: [36][120/200]	Time 0.267 (0.355)	Data 0.000 (0.081)	Loss 0.719 (0.744)
Epoch: [36][140/200]	Time 0.362 (0.355)	Data 0.000 (0.080)	Loss 0.815 (0.762)
Epoch: [36][160/200]	Time 0.269 (0.353)	Data 0.000 (0.079)	Loss 0.794 (0.773)
Epoch: [36][180/200]	Time 0.267 (0.353)	Data 0.000 (0.079)	Loss 0.788 (0.778)
Epoch: [36][200/200]	Time 0.448 (0.359)	Data 0.001 (0.085)	Loss 0.737 (0.784)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.140)	Data 0.000 (0.040)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.222774028778076
==> Statistics for epoch 37: 586 clusters
Epoch: [37][20/200]	Time 0.264 (0.383)	Data 0.001 (0.107)	Loss 0.766 (0.200)
Epoch: [37][40/200]	Time 0.266 (0.364)	Data 0.001 (0.092)	Loss 0.771 (0.517)
Epoch: [37][60/200]	Time 0.267 (0.360)	Data 0.000 (0.087)	Loss 0.776 (0.632)
Epoch: [37][80/200]	Time 0.271 (0.357)	Data 0.000 (0.085)	Loss 1.085 (0.678)
Epoch: [37][100/200]	Time 0.265 (0.355)	Data 0.000 (0.082)	Loss 0.956 (0.714)
Epoch: [37][120/200]	Time 0.272 (0.352)	Data 0.000 (0.080)	Loss 0.974 (0.732)
Epoch: [37][140/200]	Time 0.268 (0.352)	Data 0.000 (0.079)	Loss 0.880 (0.744)
Epoch: [37][160/200]	Time 0.270 (0.351)	Data 0.000 (0.078)	Loss 1.105 (0.763)
Epoch: [37][180/200]	Time 0.271 (0.351)	Data 0.000 (0.078)	Loss 0.951 (0.774)
Epoch: [37][200/200]	Time 0.270 (0.358)	Data 0.001 (0.085)	Loss 0.904 (0.781)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.137)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.89085626602173
==> Statistics for epoch 38: 585 clusters
Epoch: [38][20/200]	Time 0.272 (0.377)	Data 0.001 (0.104)	Loss 1.021 (0.181)
Epoch: [38][40/200]	Time 0.272 (0.356)	Data 0.001 (0.084)	Loss 0.894 (0.492)
Epoch: [38][60/200]	Time 0.269 (0.354)	Data 0.000 (0.081)	Loss 1.162 (0.615)
Epoch: [38][80/200]	Time 0.270 (0.354)	Data 0.000 (0.080)	Loss 1.184 (0.685)
Epoch: [38][100/200]	Time 0.273 (0.352)	Data 0.001 (0.078)	Loss 1.015 (0.732)
Epoch: [38][120/200]	Time 0.268 (0.352)	Data 0.000 (0.078)	Loss 0.801 (0.757)
Epoch: [38][140/200]	Time 0.267 (0.351)	Data 0.000 (0.077)	Loss 0.856 (0.769)
Epoch: [38][160/200]	Time 0.271 (0.350)	Data 0.000 (0.076)	Loss 0.839 (0.781)
Epoch: [38][180/200]	Time 0.264 (0.350)	Data 0.000 (0.076)	Loss 0.682 (0.782)
Epoch: [38][200/200]	Time 0.266 (0.356)	Data 0.000 (0.082)	Loss 0.986 (0.789)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.136)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.109950304031372
==> Statistics for epoch 39: 585 clusters
Epoch: [39][20/200]	Time 0.271 (0.383)	Data 0.001 (0.114)	Loss 0.856 (0.175)
Epoch: [39][40/200]	Time 0.274 (0.362)	Data 0.001 (0.089)	Loss 0.814 (0.483)
Epoch: [39][60/200]	Time 0.268 (0.359)	Data 0.000 (0.086)	Loss 0.986 (0.607)
Epoch: [39][80/200]	Time 0.270 (0.355)	Data 0.001 (0.082)	Loss 0.668 (0.665)
Epoch: [39][100/200]	Time 0.267 (0.355)	Data 0.001 (0.081)	Loss 0.963 (0.704)
Epoch: [39][120/200]	Time 0.272 (0.354)	Data 0.000 (0.081)	Loss 1.084 (0.721)
Epoch: [39][140/200]	Time 0.266 (0.353)	Data 0.000 (0.079)	Loss 0.804 (0.732)
Epoch: [39][160/200]	Time 0.268 (0.353)	Data 0.000 (0.079)	Loss 0.488 (0.744)
Epoch: [39][180/200]	Time 0.271 (0.352)	Data 0.000 (0.079)	Loss 0.869 (0.754)
Epoch: [39][200/200]	Time 0.270 (0.359)	Data 0.001 (0.085)	Loss 0.879 (0.765)
Extract Features: [50/76]	Time 0.097 (0.140)	Data 0.000 (0.038)	
Mean AP: 90.6%

 * Finished epoch  39  model mAP: 90.6%  best: 90.6% *

==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.136)	Data 0.000 (0.034)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.675817012786865
==> Statistics for epoch 40: 587 clusters
Epoch: [40][20/200]	Time 0.269 (0.383)	Data 0.001 (0.107)	Loss 0.760 (0.168)
Epoch: [40][40/200]	Time 0.275 (0.363)	Data 0.001 (0.091)	Loss 0.734 (0.494)
Epoch: [40][60/200]	Time 0.267 (0.358)	Data 0.001 (0.085)	Loss 0.930 (0.601)
Epoch: [40][80/200]	Time 0.272 (0.354)	Data 0.001 (0.080)	Loss 0.907 (0.674)
Epoch: [40][100/200]	Time 0.271 (0.352)	Data 0.001 (0.079)	Loss 0.640 (0.711)
Epoch: [40][120/200]	Time 0.264 (0.351)	Data 0.000 (0.079)	Loss 0.790 (0.729)
Epoch: [40][140/200]	Time 0.265 (0.352)	Data 0.000 (0.078)	Loss 0.845 (0.735)
Epoch: [40][160/200]	Time 0.266 (0.351)	Data 0.000 (0.079)	Loss 0.829 (0.755)
Epoch: [40][180/200]	Time 0.267 (0.351)	Data 0.000 (0.078)	Loss 0.678 (0.766)
Epoch: [40][200/200]	Time 0.270 (0.358)	Data 0.001 (0.085)	Loss 0.713 (0.770)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.132)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.357850790023804
==> Statistics for epoch 41: 587 clusters
Epoch: [41][20/200]	Time 0.269 (0.385)	Data 0.001 (0.105)	Loss 0.892 (0.200)
Epoch: [41][40/200]	Time 0.280 (0.366)	Data 0.000 (0.088)	Loss 1.053 (0.492)
Epoch: [41][60/200]	Time 0.373 (0.357)	Data 0.001 (0.080)	Loss 0.800 (0.600)
Epoch: [41][80/200]	Time 0.262 (0.351)	Data 0.000 (0.076)	Loss 0.551 (0.664)
Epoch: [41][100/200]	Time 0.269 (0.349)	Data 0.000 (0.074)	Loss 1.060 (0.694)
Epoch: [41][120/200]	Time 0.269 (0.348)	Data 0.000 (0.074)	Loss 0.691 (0.716)
Epoch: [41][140/200]	Time 0.266 (0.348)	Data 0.000 (0.074)	Loss 0.945 (0.732)
Epoch: [41][160/200]	Time 0.266 (0.348)	Data 0.000 (0.074)	Loss 1.116 (0.750)
Epoch: [41][180/200]	Time 0.268 (0.348)	Data 0.000 (0.074)	Loss 0.762 (0.760)
Epoch: [41][200/200]	Time 0.269 (0.355)	Data 0.000 (0.081)	Loss 0.732 (0.770)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.139)	Data 0.000 (0.038)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.721734762191772
==> Statistics for epoch 42: 584 clusters
Epoch: [42][20/200]	Time 0.269 (0.384)	Data 0.001 (0.106)	Loss 0.888 (0.196)
Epoch: [42][40/200]	Time 0.269 (0.366)	Data 0.001 (0.090)	Loss 0.609 (0.486)
Epoch: [42][60/200]	Time 0.266 (0.355)	Data 0.001 (0.081)	Loss 0.713 (0.582)
Epoch: [42][80/200]	Time 0.269 (0.350)	Data 0.001 (0.077)	Loss 0.774 (0.663)
Epoch: [42][100/200]	Time 0.269 (0.348)	Data 0.001 (0.074)	Loss 0.897 (0.693)
Epoch: [42][120/200]	Time 0.264 (0.347)	Data 0.000 (0.073)	Loss 0.902 (0.723)
Epoch: [42][140/200]	Time 0.265 (0.345)	Data 0.000 (0.072)	Loss 0.859 (0.748)
Epoch: [42][160/200]	Time 0.266 (0.344)	Data 0.000 (0.071)	Loss 0.949 (0.762)
Epoch: [42][180/200]	Time 0.361 (0.343)	Data 0.000 (0.070)	Loss 0.840 (0.774)
Epoch: [42][200/200]	Time 0.264 (0.348)	Data 0.001 (0.075)	Loss 0.982 (0.776)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.136 (0.132)	Data 0.040 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.676389694213867
==> Statistics for epoch 43: 586 clusters
Epoch: [43][20/200]	Time 0.266 (0.366)	Data 0.001 (0.094)	Loss 0.925 (0.162)
Epoch: [43][40/200]	Time 0.266 (0.349)	Data 0.001 (0.080)	Loss 1.000 (0.456)
Epoch: [43][60/200]	Time 0.280 (0.346)	Data 0.001 (0.077)	Loss 0.740 (0.576)
Epoch: [43][80/200]	Time 0.270 (0.344)	Data 0.001 (0.074)	Loss 0.825 (0.641)
Epoch: [43][100/200]	Time 0.264 (0.343)	Data 0.001 (0.072)	Loss 0.915 (0.684)
Epoch: [43][120/200]	Time 0.267 (0.340)	Data 0.000 (0.070)	Loss 0.755 (0.706)
Epoch: [43][140/200]	Time 0.265 (0.340)	Data 0.000 (0.070)	Loss 0.930 (0.734)
Epoch: [43][160/200]	Time 0.264 (0.340)	Data 0.000 (0.070)	Loss 1.059 (0.751)
Epoch: [43][180/200]	Time 0.262 (0.339)	Data 0.000 (0.069)	Loss 0.693 (0.769)
Epoch: [43][200/200]	Time 0.267 (0.346)	Data 0.000 (0.075)	Loss 0.836 (0.773)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.096 (0.133)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.037235021591187
==> Statistics for epoch 44: 586 clusters
Epoch: [44][20/200]	Time 0.269 (0.371)	Data 0.001 (0.099)	Loss 0.944 (0.199)
Epoch: [44][40/200]	Time 0.269 (0.354)	Data 0.001 (0.084)	Loss 1.072 (0.510)
Epoch: [44][60/200]	Time 0.267 (0.349)	Data 0.001 (0.077)	Loss 0.907 (0.611)
Epoch: [44][80/200]	Time 0.271 (0.347)	Data 0.001 (0.075)	Loss 0.890 (0.670)
Epoch: [44][100/200]	Time 0.263 (0.344)	Data 0.001 (0.072)	Loss 0.519 (0.700)
Epoch: [44][120/200]	Time 0.267 (0.343)	Data 0.000 (0.071)	Loss 0.764 (0.727)
Epoch: [44][140/200]	Time 0.268 (0.342)	Data 0.000 (0.069)	Loss 1.034 (0.746)
Epoch: [44][160/200]	Time 0.268 (0.342)	Data 0.000 (0.069)	Loss 0.752 (0.761)
Epoch: [44][180/200]	Time 0.266 (0.341)	Data 0.000 (0.068)	Loss 0.952 (0.771)
Epoch: [44][200/200]	Time 0.268 (0.348)	Data 0.001 (0.075)	Loss 0.723 (0.780)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.132)	Data 0.000 (0.030)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.227222442626953
==> Statistics for epoch 45: 587 clusters
Epoch: [45][20/200]	Time 0.274 (0.377)	Data 0.001 (0.106)	Loss 0.759 (0.177)
Epoch: [45][40/200]	Time 0.375 (0.357)	Data 0.001 (0.085)	Loss 0.810 (0.481)
Epoch: [45][60/200]	Time 0.265 (0.349)	Data 0.001 (0.079)	Loss 0.751 (0.590)
Epoch: [45][80/200]	Time 0.264 (0.347)	Data 0.001 (0.077)	Loss 0.984 (0.660)
Epoch: [45][100/200]	Time 0.417 (0.346)	Data 0.001 (0.074)	Loss 0.915 (0.689)
Epoch: [45][120/200]	Time 0.264 (0.344)	Data 0.000 (0.073)	Loss 0.911 (0.719)
Epoch: [45][140/200]	Time 0.269 (0.344)	Data 0.000 (0.073)	Loss 0.880 (0.744)
Epoch: [45][160/200]	Time 0.265 (0.344)	Data 0.000 (0.072)	Loss 0.906 (0.759)
Epoch: [45][180/200]	Time 0.264 (0.343)	Data 0.000 (0.071)	Loss 0.883 (0.770)
Epoch: [45][200/200]	Time 0.272 (0.349)	Data 0.001 (0.077)	Loss 0.894 (0.777)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.097 (0.133)	Data 0.000 (0.032)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.880117654800415
==> Statistics for epoch 46: 586 clusters
Epoch: [46][20/200]	Time 0.289 (0.374)	Data 0.001 (0.101)	Loss 0.744 (0.176)
Epoch: [46][40/200]	Time 0.269 (0.357)	Data 0.001 (0.083)	Loss 0.636 (0.472)
Epoch: [46][60/200]	Time 0.273 (0.358)	Data 0.001 (0.084)	Loss 0.843 (0.602)
Epoch: [46][80/200]	Time 0.270 (0.356)	Data 0.001 (0.081)	Loss 0.816 (0.667)
Epoch: [46][100/200]	Time 0.270 (0.353)	Data 0.001 (0.079)	Loss 0.747 (0.695)
Epoch: [46][120/200]	Time 0.269 (0.352)	Data 0.000 (0.078)	Loss 0.731 (0.716)
Epoch: [46][140/200]	Time 0.267 (0.350)	Data 0.000 (0.076)	Loss 1.110 (0.730)
Epoch: [46][160/200]	Time 0.263 (0.349)	Data 0.000 (0.075)	Loss 0.942 (0.741)
Epoch: [46][180/200]	Time 0.267 (0.348)	Data 0.000 (0.074)	Loss 0.697 (0.749)
Epoch: [46][200/200]	Time 0.274 (0.354)	Data 0.000 (0.080)	Loss 0.735 (0.756)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.098 (0.133)	Data 0.000 (0.031)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.22758460044861
==> Statistics for epoch 47: 586 clusters
Epoch: [47][20/200]	Time 0.269 (0.372)	Data 0.001 (0.094)	Loss 0.507 (0.154)
Epoch: [47][40/200]	Time 0.268 (0.354)	Data 0.001 (0.082)	Loss 1.125 (0.472)
Epoch: [47][60/200]	Time 0.268 (0.349)	Data 0.001 (0.076)	Loss 0.867 (0.549)
Epoch: [47][80/200]	Time 0.265 (0.346)	Data 0.001 (0.073)	Loss 0.723 (0.617)
Epoch: [47][100/200]	Time 0.268 (0.344)	Data 0.001 (0.071)	Loss 0.975 (0.667)
Epoch: [47][120/200]	Time 0.270 (0.344)	Data 0.000 (0.071)	Loss 0.961 (0.697)
Epoch: [47][140/200]	Time 0.263 (0.343)	Data 0.000 (0.071)	Loss 0.626 (0.725)
Epoch: [47][160/200]	Time 0.265 (0.343)	Data 0.000 (0.070)	Loss 0.767 (0.747)
Epoch: [47][180/200]	Time 0.266 (0.343)	Data 0.000 (0.070)	Loss 0.757 (0.763)
Epoch: [47][200/200]	Time 0.273 (0.349)	Data 0.001 (0.076)	Loss 0.979 (0.771)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.134 (0.130)	Data 0.038 (0.029)	
Computing jaccard distance...
Jaccard distance computing time cost: 19.01333498954773
==> Statistics for epoch 48: 587 clusters
Epoch: [48][20/200]	Time 0.266 (0.370)	Data 0.001 (0.100)	Loss 0.742 (0.159)
Epoch: [48][40/200]	Time 0.266 (0.354)	Data 0.001 (0.081)	Loss 0.845 (0.467)
Epoch: [48][60/200]	Time 0.269 (0.349)	Data 0.001 (0.076)	Loss 0.670 (0.603)
Epoch: [48][80/200]	Time 0.269 (0.346)	Data 0.001 (0.074)	Loss 0.993 (0.681)
Epoch: [48][100/200]	Time 0.267 (0.343)	Data 0.001 (0.072)	Loss 0.950 (0.701)
Epoch: [48][120/200]	Time 0.265 (0.343)	Data 0.000 (0.071)	Loss 0.691 (0.717)
Epoch: [48][140/200]	Time 0.268 (0.343)	Data 0.000 (0.070)	Loss 0.944 (0.729)
Epoch: [48][160/200]	Time 0.356 (0.342)	Data 0.000 (0.070)	Loss 0.804 (0.739)
Epoch: [48][180/200]	Time 0.262 (0.342)	Data 0.000 (0.070)	Loss 0.672 (0.753)
Epoch: [48][200/200]	Time 0.268 (0.348)	Data 0.000 (0.076)	Loss 0.891 (0.768)
==> Create pseudo labels for unlabeled data
Extract Features: [50/51]	Time 0.095 (0.137)	Data 0.000 (0.035)	
Computing jaccard distance...
Jaccard distance computing time cost: 18.839455604553223
==> Statistics for epoch 49: 586 clusters
Epoch: [49][20/200]	Time 0.286 (0.378)	Data 0.001 (0.106)	Loss 0.631 (0.155)
Epoch: [49][40/200]	Time 0.265 (0.361)	Data 0.001 (0.088)	Loss 0.858 (0.468)
Epoch: [49][60/200]	Time 0.269 (0.351)	Data 0.001 (0.079)	Loss 0.866 (0.601)
Epoch: [49][80/200]	Time 0.266 (0.347)	Data 0.001 (0.074)	Loss 0.914 (0.659)
Epoch: [49][100/200]	Time 0.265 (0.346)	Data 0.001 (0.073)	Loss 0.922 (0.696)
Epoch: [49][120/200]	Time 0.263 (0.344)	Data 0.000 (0.072)	Loss 0.757 (0.720)
Epoch: [49][140/200]	Time 0.263 (0.343)	Data 0.000 (0.071)	Loss 0.807 (0.737)
Epoch: [49][160/200]	Time 0.269 (0.343)	Data 0.000 (0.071)	Loss 0.864 (0.746)
Epoch: [49][180/200]	Time 0.267 (0.342)	Data 0.000 (0.070)	Loss 1.076 (0.754)
Epoch: [49][200/200]	Time 0.273 (0.350)	Data 0.001 (0.077)	Loss 0.767 (0.759)
Extract Features: [50/76]	Time 0.098 (0.137)	Data 0.000 (0.035)	
Mean AP: 90.6%

 * Finished epoch  49  model mAP: 90.6%  best: 90.6% *

==> Test with the best model:
=> Loaded checkpoint 'log/market/vit_small_cion/model_best.pth.tar'
Extract Features: [50/76]	Time 0.097 (0.136)	Data 0.000 (0.034)	
Mean AP: 90.6%
CMC Scores:
  top-1          95.6%
  top-5          98.3%
  top-10         99.0%
Total running time:  1:25:52.988658
