Upload train_log.txt

CCVID_IMG/train_log.txt

@@ -0,0 +1,722 @@
+EVA-attribure: Saving model in the path :CCVID_IMG
+EVA-attribure: Namespace(config_file='configs/mevid/eva02_l_cloth.yml', eval=False, local_rank=0, multi_node=False, opts=['DATA.ADD_META', 'False', 'DATA.MASK_META', 'False', 'MODEL.DIST_TRAIN', 'True', 'SOLVER.LOG_PERIOD', '50', 'DATA.ROOT', '/home/c3-0/datasets/CCVID/', 'DATA.TEST_BATCH', '500', 'DATA.DATASET', 'ccvid', 'SOLVER.EVAL_PERIOD', '2', 'OUTPUT_DIR', 'CCVID_IMG'], resume=False)
+EVA-attribure: Loaded configuration file configs/mevid/eva02_l_cloth.yml
+EVA-attribure: 
+MODEL:
+  TYPE: eva02_cloth
+  NAME: eva02_l_cloth
+  META_DIMS: [ 105, ]
+  METRIC_LOSS_TYPE: 'triplet'
+  IF_LABELSMOOTH: 'off'
+  IF_WITH_CENTER: 'no'
+  NO_MARGIN: True
+  ADD_META: True
+  CLOTH_ONLY: True
+
+DATA:
+  IMG_HEIGHT: 224
+  IMG_WIDTH: 224
+  DATASET: 'mevid'
+
+SOLVER:
+  OPTIMIZER_NAME: 'SGD'
+  MAX_EPOCHS: 60
+  BASE_LR: 2e-5
+  WARMUP_METHOD: 'linear'
+  LARGE_FC_LR: False
+  CHECKPOINT_PERIOD: 60
+  LOG_PERIOD: 50
+  EVAL_PERIOD: 1
+  WEIGHT_DECAY:  0.05
+  WEIGHT_DECAY_BIAS: 0.05
+  BIAS_LR_FACTOR: 2
+
+TEST:
+  WEIGHT: ''
+  FEAT_NORM: 'yes'
+  TYPE: 'image_only'
+OUTPUT_DIR: ''
+
+
+
+EVA-attribure: Running with config:
+AUG:
+  RC_PROB: 0.5
+  RE_PROB: 0.5
+  RF_PROB: 0.5
+  SAMPLING_STRIDE: 4
+  SEQ_LEN: 8
+  TEMPORAL_SAMPLING_MODE: stride
+DATA:
+  ADD_META: False
+  AUX_INFO: True
+  BATCH_SIZE: 8
+  DATASET: ccvid
+  F8: None
+  IMG_HEIGHT: 224
+  IMG_WIDTH: 224
+  MASK_META: False
+  META_DIR: PAR_PETA_105.txt
+  NUM_INSTANCES: 2
+  NUM_WORKERS: 4
+  PIN_MEMORY: True
+  ROOT: /home/c3-0/datasets/CCVID/
+  SAMPLER: softmax_triplet
+  TEST_BATCH: 500
+MODEL:
+  ADD_META: True
+  Adapter: None
+  CLOTH_ONLY: True
+  CLOTH_XISHU: 3
+  COS_LAYER: False
+  DEVICE: cuda
+  DEVICE_ID: 0
+  DIST_TRAIN: True
+  ID_LOSS_TYPE: softmax
+  ID_LOSS_WEIGHT: 1.0
+  IF_LABELSMOOTH: off
+  IF_WITH_CENTER: no
+  Joint: None
+  MASK_META: False
+  META_DIMS: [105]
+  METRIC_LOSS_TYPE: triplet
+  NAME: eva02_l_cloth
+  NO_MARGIN: True
+  TIM_DIM: 4
+  TRIPLET_LOSS_WEIGHT: 1.0
+  TYPE: eva02_cloth
+OUTPUT_DIR: CCVID_IMG
+SOLVER:
+  BASE_LR: 2e-05
+  BIAS_LR_FACTOR: 2
+  CENTER_LOSS_WEIGHT: 0.0005
+  CENTER_LR: 0.5
+  CHECKPOINT_PERIOD: 60
+  COSINE_MARGIN: 0.5
+  COSINE_SCALE: 30
+  EVAL_PERIOD: 2
+  GAMMA: 0.1
+  LARGE_FC_LR: False
+  LOG_PERIOD: 50
+  MARGIN: 0.3
+  MAX_EPOCHS: 60
+  MOMENTUM: 0.9
+  OPTIMIZER_NAME: SGD
+  SEED: 1234
+  STEPS: (40, 60)
+  WARMUP_EPOCHS: 20
+  WARMUP_FACTOR: 0.01
+  WARMUP_LR: 7.8125e-07
+  WARMUP_METHOD: linear
+  WEIGHT_DECAY: 0.05
+  WEIGHT_DECAY_BIAS: 0.05
+TEST:
+  FEAT_NORM: yes
+  TYPE: image_only
+  WEIGHT: 
+TRAIN:
+  E2E: True
+  START_EPOCH: 1
+  TRAIN_VIDEO: None
+EVA-attribure: => CCVID loaded
+EVA-attribure: Dataset statistics:
+EVA-attribure:   ---------------------------------------------
+EVA-attribure:   subset       | # ids | # tracklets | # clothes
+EVA-attribure:   ---------------------------------------------
+EVA-attribure:   train        |    75 |         948 |       159
+EVA-attribure:   train_dense  |    75 |        1409 |       159
+EVA-attribure:   query        |   151 |         834 |       160
+EVA-attribure:   gallery      |   151 |        1074 |       252
+EVA-attribure:   ---------------------------------------------
+EVA-attribure:   total        |   226 |        2856 |       480
+EVA-attribure:   number of images per tracklet: 27 ~ 410, average 121.8
+EVA-attribure:   ---------------------------------------------
+EVA-attribure:   --------------------------------------------------------------- 
+EVA-attribure:    Partition    |  <32  | '32-64' | '64-128' | '128-256' | '>256' 
+EVA-attribure:    train.txt    |     0 |  13799  |   49577  |   55237   |      0 
+EVA-attribure:    query.txt    |     0 |  26702  |   41930  |   48167   |      0 
+EVA-attribure:    gallery.txt  |     0 |  11776  |   32958  |   67687   |      0 
+EVA-attribure:     ---------------------------------------------------------------
+EVA-attribure.train: start training
+EVA-attribure.train: Epoch[1] Iteration[50/59] Loss: 5.108, Acc: 0.032, Base Lr: 1.74e-06
+EVA-attribure.train: Epoch[2] Iteration[50/59] Loss: 4.783, Acc: 0.117, Base Lr: 2.70e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 9s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:79.1% top5:86.6% top10:91.8% top20:95.8% mAP:77.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:98.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:70.5% top5:82.1% top10:89.6% top20:94.4% mAP:68.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 70.5%, Best Map 68.7% achieved at epoch 2
+EVA-attribure.train: Epoch[3] Iteration[50/59] Loss: 4.609, Acc: 0.135, Base Lr: 3.66e-06
+EVA-attribure.train: Epoch[4] Iteration[50/59] Loss: 4.453, Acc: 0.132, Base Lr: 4.63e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:78.8% top5:85.6% top10:91.8% top20:96.4% mAP:78.1%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:99.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:71.9% top5:82.7% top10:90.3% top20:95.0% mAP:71.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 71.9%, Best Map 71.7% achieved at epoch 4
+EVA-attribure.train: Epoch[5] Iteration[50/59] Loss: 4.444, Acc: 0.135, Base Lr: 5.59e-06
+EVA-attribure.train: Epoch[6] Iteration[50/59] Loss: 4.431, Acc: 0.135, Base Lr: 6.55e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 3s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:81.7% top5:86.6% top10:91.2% top20:96.9% mAP:81.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:99.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:76.4% top5:83.8% top10:89.7% top20:95.6% mAP:75.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 76.4%, Best Map 75.8% achieved at epoch 6
+EVA-attribure.train: Epoch[7] Iteration[50/59] Loss: 4.395, Acc: 0.135, Base Lr: 7.51e-06
+EVA-attribure.train: Epoch[8] Iteration[50/59] Loss: 4.333, Acc: 0.135, Base Lr: 8.47e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 3s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:80.5% top5:84.7% top10:89.9% top20:94.7% mAP:79.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:75.8% top5:82.3% top10:88.2% top20:93.4% mAP:75.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[9] Iteration[50/59] Loss: 4.302, Acc: 0.135, Base Lr: 9.43e-06
+EVA-attribure.train: Epoch[10] Iteration[50/59] Loss: 4.271, Acc: 0.135, Base Lr: 1.04e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:81.5% top5:86.1% top10:89.9% top20:94.7% mAP:81.5%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:77.9% top5:84.7% top10:88.5% top20:94.4% mAP:78.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 77.9%, Best Map 78.8% achieved at epoch 10
+EVA-attribure.train: Epoch[11] Iteration[50/59] Loss: 4.209, Acc: 0.140, Base Lr: 1.14e-05
+EVA-attribure.train: Epoch[12] Iteration[50/59] Loss: 4.165, Acc: 0.142, Base Lr: 1.23e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 4s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:84.5% top5:88.8% top10:92.6% top20:94.8% mAP:84.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:81.5% top5:87.1% top10:91.4% top20:94.5% mAP:82.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 81.5%, Best Map 82.2% achieved at epoch 12
+EVA-attribure.train: Epoch[13] Iteration[50/59] Loss: 4.073, Acc: 0.190, Base Lr: 1.33e-05
+EVA-attribure.train: Epoch[14] Iteration[50/59] Loss: 3.943, Acc: 0.327, Base Lr: 1.42e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.6% top5:90.8% top10:93.6% top20:94.2% mAP:87.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:84.9% top5:89.6% top10:93.2% top20:93.9% mAP:85.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 84.9%, Best Map 85.8% achieved at epoch 14
+EVA-attribure.train: Epoch[15] Iteration[50/59] Loss: 3.781, Acc: 0.462, Base Lr: 1.52e-05
+EVA-attribure.train: Epoch[16] Iteration[50/59] Loss: 3.685, Acc: 0.615, Base Lr: 1.62e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 5s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:88.1% top5:91.1% top10:93.3% top20:94.1% mAP:88.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.4% top5:89.6% top10:92.6% top20:93.8% mAP:86.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 85.4%, Best Map 86.0% achieved at epoch 16
+EVA-attribure.train: Epoch[17] Iteration[50/59] Loss: 3.512, Acc: 0.765, Base Lr: 1.71e-05
+EVA-attribure.train: Epoch[18] Iteration[50/59] Loss: 3.316, Acc: 0.837, Base Lr: 1.81e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 15s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.1% top5:90.4% top10:92.4% top20:93.9% mAP:86.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:83.5% top5:88.8% top10:91.2% top20:93.5% mAP:84.5%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[19] Iteration[50/59] Loss: 3.201, Acc: 0.857, Base Lr: 1.90e-05
+EVA-attribure.train: Epoch[20] Iteration[50/59] Loss: 2.973, Acc: 0.917, Base Lr: 1.55e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 4s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.5% top5:90.8% top10:92.4% top20:93.8% mAP:87.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.5% top5:89.8% top10:92.0% top20:93.4% mAP:85.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 85.5%, Best Map 86.0% achieved at epoch 20
+EVA-attribure.train: Epoch[21] Iteration[50/59] Loss: 2.860, Acc: 0.917, Base Lr: 1.51e-05
+EVA-attribure.train: Epoch[22] Iteration[50/59] Loss: 2.712, Acc: 0.942, Base Lr: 1.47e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.9% top5:90.5% top10:92.4% top20:93.9% mAP:87.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.0% top5:89.6% top10:92.0% top20:93.5% mAP:85.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 86.0%, Best Map 86.0% achieved at epoch 22
+EVA-attribure.train: Epoch[23] Iteration[50/59] Loss: 2.565, Acc: 0.960, Base Lr: 1.42e-05
+EVA-attribure.train: Epoch[24] Iteration[50/59] Loss: 2.414, Acc: 0.962, Base Lr: 1.38e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 1s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.5% top5:90.8% top10:92.4% top20:93.6% mAP:87.7%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.7% top5:89.8% top10:92.1% top20:93.3% mAP:86.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[25] Iteration[50/59] Loss: 2.416, Acc: 0.955, Base Lr: 1.33e-05
+EVA-attribure.train: Epoch[26] Iteration[50/59] Loss: 2.227, Acc: 0.967, Base Lr: 1.29e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:88.6% top5:91.1% top10:92.4% top20:93.9% mAP:88.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.3% top5:90.0% top10:92.1% top20:93.5% mAP:86.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: ==> Best Rank-1 86.3%, Best Map 86.6% achieved at epoch 26
+EVA-attribure.train: Epoch[27] Iteration[50/59] Loss: 2.106, Acc: 0.977, Base Lr: 1.24e-05
+EVA-attribure.train: Epoch[28] Iteration[50/59] Loss: 2.121, Acc: 0.970, Base Lr: 1.19e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 1s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.8% top5:91.2% top10:92.4% top20:93.8% mAP:88.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.3% top5:90.2% top10:92.0% top20:93.4% mAP:86.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[29] Iteration[50/59] Loss: 1.925, Acc: 0.980, Base Lr: 1.15e-05
+EVA-attribure.train: Epoch[30] Iteration[50/59] Loss: 1.841, Acc: 0.985, Base Lr: 1.10e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 5s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.2% top5:90.8% top10:92.3% top20:93.6% mAP:87.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:84.3% top5:90.0% top10:92.0% top20:93.5% mAP:85.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[31] Iteration[50/59] Loss: 1.775, Acc: 0.977, Base Lr: 1.05e-05
+EVA-attribure.train: Epoch[32] Iteration[50/59] Loss: 1.748, Acc: 0.977, Base Lr: 1.01e-05
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 12s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.1% top5:90.4% top10:92.3% top20:93.6% mAP:87.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.1% top5:89.6% top10:92.0% top20:93.3% mAP:86.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[33] Iteration[50/59] Loss: 1.672, Acc: 0.980, Base Lr: 9.59e-06
+EVA-attribure.train: Epoch[34] Iteration[50/59] Loss: 1.632, Acc: 0.980, Base Lr: 9.13e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.1% top5:90.5% top10:92.3% top20:93.6% mAP:88.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.0% top5:89.7% top10:92.0% top20:93.5% mAP:86.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[35] Iteration[50/59] Loss: 1.515, Acc: 0.980, Base Lr: 8.67e-06
+EVA-attribure.train: Epoch[36] Iteration[50/59] Loss: 1.474, Acc: 0.975, Base Lr: 8.22e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.7% top5:90.5% top10:92.3% top20:93.4% mAP:87.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:84.8% top5:89.7% top10:92.0% top20:93.0% mAP:86.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[37] Iteration[50/59] Loss: 1.432, Acc: 0.987, Base Lr: 7.77e-06
+EVA-attribure.train: Epoch[38] Iteration[50/59] Loss: 1.374, Acc: 0.982, Base Lr: 7.34e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 1s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.8% top5:90.5% top10:92.3% top20:93.5% mAP:87.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:84.9% top5:89.7% top10:92.0% top20:93.4% mAP:86.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[39] Iteration[50/59] Loss: 1.333, Acc: 0.987, Base Lr: 6.91e-06
+EVA-attribure.train: Epoch[40] Iteration[50/59] Loss: 1.312, Acc: 0.993, Base Lr: 6.50e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.1% top5:90.5% top10:92.2% top20:93.5% mAP:87.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.1% top5:89.7% top10:91.8% top20:93.4% mAP:86.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[41] Iteration[50/59] Loss: 1.303, Acc: 0.990, Base Lr: 6.10e-06
+EVA-attribure.train: Epoch[42] Iteration[50/59] Loss: 1.211, Acc: 0.995, Base Lr: 5.71e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.3% top5:90.8% top10:92.3% top20:93.5% mAP:88.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.5% top5:89.9% top10:92.0% top20:93.5% mAP:87.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[43] Iteration[50/59] Loss: 1.211, Acc: 0.987, Base Lr: 5.34e-06
+EVA-attribure.train: Epoch[44] Iteration[50/59] Loss: 1.173, Acc: 0.982, Base Lr: 4.98e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.2% top5:90.6% top10:92.2% top20:93.6% mAP:88.1%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.1% top5:89.8% top10:91.8% top20:93.6% mAP:86.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[45] Iteration[50/59] Loss: 1.181, Acc: 0.982, Base Lr: 4.64e-06
+EVA-attribure.train: Epoch[46] Iteration[50/59] Loss: 1.162, Acc: 0.977, Base Lr: 4.31e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 16s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.4% top5:90.8% top10:92.2% top20:93.5% mAP:88.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.6% top5:89.9% top10:91.8% top20:93.4% mAP:87.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[47] Iteration[50/59] Loss: 1.156, Acc: 0.970, Base Lr: 4.01e-06
+EVA-attribure.train: Epoch[48] Iteration[50/59] Loss: 1.092, Acc: 0.993, Base Lr: 3.72e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.4% top5:90.8% top10:92.2% top20:93.5% mAP:88.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.6% top5:89.9% top10:91.8% top20:93.4% mAP:86.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[49] Iteration[50/59] Loss: 1.108, Acc: 0.985, Base Lr: 3.45e-06
+EVA-attribure.train: Epoch[50] Iteration[50/59] Loss: 1.095, Acc: 0.987, Base Lr: 3.21e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.2% top5:90.9% top10:92.2% top20:93.6% mAP:88.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.4% top5:90.0% top10:91.8% top20:93.6% mAP:86.9%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[51] Iteration[50/59] Loss: 1.067, Acc: 0.987, Base Lr: 2.98e-06
+EVA-attribure.train: Epoch[52] Iteration[50/59] Loss: 1.080, Acc: 0.982, Base Lr: 2.78e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 10s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.8% top5:90.9% top10:92.2% top20:93.8% mAP:88.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.0% top5:89.9% top10:91.8% top20:93.8% mAP:87.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[53] Iteration[50/59] Loss: 1.107, Acc: 0.982, Base Lr: 2.60e-06
+EVA-attribure.train: Epoch[54] Iteration[50/59] Loss: 1.026, Acc: 0.982, Base Lr: 2.44e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.9% top5:91.0% top10:92.2% top20:93.8% mAP:88.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.0% top5:90.0% top10:91.8% top20:93.8% mAP:87.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[55] Iteration[50/59] Loss: 1.078, Acc: 0.977, Base Lr: 2.31e-06
+EVA-attribure.train: Epoch[56] Iteration[50/59] Loss: 1.023, Acc: 0.993, Base Lr: 2.20e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 15s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.9% top5:91.0% top10:92.2% top20:93.8% mAP:88.8%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.0% top5:90.0% top10:91.8% top20:93.8% mAP:87.4%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[57] Iteration[50/59] Loss: 1.028, Acc: 0.982, Base Lr: 2.11e-06
+EVA-attribure.train: Epoch[58] Iteration[50/59] Loss: 0.999, Acc: 0.990, Base Lr: 2.05e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.8% top5:91.0% top10:92.2% top20:93.6% mAP:88.6%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:86.0% top5:90.0% top10:91.8% top20:93.6% mAP:87.3%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Epoch[59] Iteration[50/59] Loss: 1.057, Acc: 0.995, Base Lr: 2.01e-06
+EVA-attribure.train: Epoch[60] Iteration[50/59] Loss: 1.015, Acc: 0.980, Base Lr: 2.00e-06
+EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
+EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
+EVA-attribure: Extracting features complete in 3m 2s
+EVA-attribure: Distance computing in 0m 0s
+EVA-attribure: Computing CMC and mAP
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:87.6% top5:91.0% top10:92.2% top20:93.6% mAP:88.5%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Using 0m 0s
+EVA-attribure: Computing CMC and mAP only for the same clothes setting
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure: Computing CMC and mAP only for clothes-changing
+EVA-attribure: Results ---------------------------------------------------
+EVA-attribure: top1:85.9% top5:90.0% top10:91.8% top20:93.6% mAP:87.2%
+EVA-attribure: -----------------------------------------------------------
+EVA-attribure.train: Training time 1:51:13
+EVA-attribure.train: ==> Best Rank-1 86.3%, Best Map 87.4% achieved at epoch 26