Hugging Face's logo

ppriyank
/
CSCI-CC-ReID

Image-to-Image
ReID
CC-ReID
LTCC
PRCC
CCVID
MeVID
Model card Files
xet
ppriyank commited on
Commit
1704cfa
·
verified ·
1 Parent(s): 4bd9209

Upload train_log.txt

Browse files
Files changed (1) hide show
  1. ccvid-49-1245/train_log.txt +1481 -0
ccvid-49-1245/train_log.txt ADDED
@@ -0,0 +1,1481 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EVA-attribure:
2
+ Why load pretrained weights when training e2e
3
+ NoneType: None
4
+ space(cal_eval=False, config_file='configs/ccvid_eva02_l_cloth.yml', env='nccl', eval=False, local_rank=0, multi_node=False, no_head=False, opts=['DATA.ROOT', '/home/c3-0/datasets/CCVID', 'TRAIN.TRAIN_VIDEO', 'True', 'MODEL.MOTION_LOSS', 'True', 'TRAIN.TEACH1', 'ccvid', 'TEST.WEIGHT', 'logs/CCVID/CCVID_IMG/eva02_l_cloth_best.pth', 'TRAIN.HYBRID', 'True', 'TRAIN.DIR_TEACH1', '/home/c3-0/datasets/CCVID', 'TRAIN.TEACH1_MODEL', 'None', 'TRAIN.TEACH1_LOAD_AS_IMG', 'True', 'TRAIN.TEACH_DATASET_FIX', 'color_adv', 'TRAIN.COLOR_ADV', 'True', 'MODEL.NAME', 'ez_eva02_vid_hybrid_extra', 'TRAIN.COLOR_PROFILE', '49', 'SOLVER.SEED', '1245', 'OUTPUT_DIR', 'ccvid-49-1245', 'SOLVER.MAX_EPOCHS', '100', 'SOLVER.LOG_PERIOD', '800'], resume=True, return_index=False, save5=False)
5
+ EVA-attribure: Loaded configuration file configs/ccvid_eva02_l_cloth.yml
6
+ EVA-attribure:
7
+ MODEL:
8
+ META_DIMS: [ 105, ]
9
+
10
+ DATA:
11
+ DATASET: 'ccvid'
12
+ TEST_BATCH: 100
13
+
14
+ SOLVER:
15
+ CHECKPOINT_PERIOD: 60
16
+ LOG_PERIOD: 50
17
+ EVAL_PERIOD: 2
18
+
19
+ OUTPUT_DIR: ''
20
+
21
+
22
+
23
+ EVA-attribure: Running with config:
24
+ ANALYSIS_STATS: None
25
+ AUG:
26
+ RC_PROB: 0.5
27
+ RE_PROB: 0.5
28
+ RF_PROB: 0.5
29
+ SAMPLING_STRIDE: 4
30
+ SEQ_LEN: 4
31
+ TEMPORAL_SAMPLING_MODE: stride
32
+ AUX_DUMP: None
33
+ DATA:
34
+ ADD_META: False
35
+ AUX_INFO: True
36
+ BATCH_SIZE: 8
37
+ DATASET: ccvid
38
+ DATASET_FIX: None
39
+ DATASET_SAMPLING_PERCENTAGE: None
40
+ DENNIS_MODE: False
41
+ F8: None
42
+ GREY_SCALE: None
43
+ IMG_HEIGHT: 224
44
+ IMG_WIDTH: 224
45
+ MASK_META: False
46
+ META_DIR: PAR_PETA_105.txt
47
+ NUM_INSTANCES: 2
48
+ NUM_WORKERS: 4
49
+ PIN_MEMORY: True
50
+ RANDOM_FRAMES: None
51
+ ROOT: /home/c3-0/datasets/CCVID
52
+ SAMPLER: softmax_triplet
53
+ SAMPLING_PERCENTAGE: None
54
+ TEST_BATCH: 100
55
+ GRAD_CAM: None
56
+ MODEL:
57
+ ADD_META: True
58
+ ATT_AS_INPUT: None
59
+ ATT_DIRECT: None
60
+ Adapter: None
61
+ CLOTH_EMBED: None
62
+ CLOTH_ONLY: True
63
+ CLOTH_XISHU: 3
64
+ COS_LAYER: False
65
+ DEVICE: cuda
66
+ DEVICE_ID: 0
67
+ DIST_TRAIN: True
68
+ EMBED_DIM: 1024
69
+ EXTRA_DIM: 12288
70
+ EXTRA_DISENTANGLE: None
71
+ ID_LOSS_TYPE: softmax
72
+ ID_LOSS_WEIGHT: 1.0
73
+ IF_LABELSMOOTH: off
74
+ IF_WITH_CENTER: no
75
+ Joint: None
76
+ MASKED_SEP_ATTN: None
77
+ MASK_META: False
78
+ META_DIMS: [105]
79
+ METRIC_LOSS_TYPE: triplet
80
+ MOTION_LOSS: True
81
+ NAME: ez_eva02_vid_hybrid_extra
82
+ NO_MARGIN: True
83
+ PRETRAIN: True
84
+ RETURN_EARLY: None
85
+ SPATIAL_AVG: None
86
+ TEMPORAL_AVG: None
87
+ TIM_DIM: 4
88
+ TRIPLET_LOSS_WEIGHT: 1.0
89
+ TYPE: eva02_cloth
90
+ UNIFIED_DIST: None
91
+ OUTPUT_DIR: Dump/ccvid-49-1245
92
+ SOLVER:
93
+ BASE_LR: 2e-05
94
+ BIAS_LR_FACTOR: 2
95
+ CENTER_LOSS_WEIGHT: 0.0005
96
+ CENTER_LR: 0.5
97
+ CHECKPOINT_PERIOD: 60
98
+ COSINE_MARGIN: 0.5
99
+ COSINE_SCALE: 30
100
+ EVAL_PERIOD: 2
101
+ GAMMA: 0.1
102
+ LARGE_FC_LR: False
103
+ LOG_PERIOD: 800
104
+ MARGIN: 0.3
105
+ MAX_EPOCHS: 100
106
+ MOMENTUM: 0.9
107
+ OPTIMIZER_NAME: SGD
108
+ SEED: 1245
109
+ STEPS: (40, 60)
110
+ WARMUP_EPOCHS: 20
111
+ WARMUP_FACTOR: 0.01
112
+ WARMUP_LR: 7.8125e-07
113
+ WARMUP_METHOD: linear
114
+ WEIGHT_DECAY: 0.05
115
+ WEIGHT_DECAY_BIAS: 0.05
116
+ TAG: None
117
+ TENSORBOARD: None
118
+ TEST:
119
+ CONCAT_COLORS: None
120
+ FEAT_NORM: yes
121
+ MODE: None
122
+ TYPE: image_only
123
+ WEIGHT: logs/CCVID/CCVID_IMG/eva02_l_cloth_best.pth
124
+ TRAIN:
125
+ COLOR_ADV: True
126
+ COLOR_LOSS: None
127
+ COLOR_PROFILE: 49
128
+ CONT_ONLY: None
129
+ DEBUG: None
130
+ DIR_TEACH1: /home/c3-0/datasets/CCVID
131
+ E2E: True
132
+ GENDER: None
133
+ HYBRID: True
134
+ LAYER_DISESNTANGLE: None
135
+ PAIR_MSE: None
136
+ POSE: None
137
+ POSE_ONLY: None
138
+ START_EPOCH: 1
139
+ TEACH1: ccvid
140
+ TEACH1_LOAD_AS_IMG: True
141
+ TEACH1_MODEL: None
142
+ TEACH1_MODEL_WT: None
143
+ TEACH1_NUMCLASSES: None
144
+ TEACH_DATASET_FIX: color_adv
145
+ TEACH_METHOD: None
146
+ TRAIN_VIDEO: True
147
+ TRAIN_DUMP: None
148
+ EVA-attribure: => CCVID loaded
149
+ EVA-attribure: Dataset statistics:
150
+ EVA-attribure: ---------------------------------------------
151
+ EVA-attribure: subset | # ids | # tracklets | # clothes
152
+ EVA-attribure: ---------------------------------------------
153
+ EVA-attribure: train | 75 | 948 | 159
154
+ EVA-attribure: train_dense | 75 | 1409 | 159
155
+ EVA-attribure: query | 151 | 834 | 160
156
+ EVA-attribure: gallery | 151 | 1074 | 252
157
+ EVA-attribure: ---------------------------------------------
158
+ EVA-attribure: total | 226 | 2856 | 480
159
+ EVA-attribure: number of images per tracklet: 27 ~ 410, average 121.8
160
+ EVA-attribure: ---------------------------------------------
161
+ EVA-attribure: ---------------------------------------------------------------
162
+ EVA-attribure: Partition | <32 | '32-64' | '64-128' | '128-256' | '>256'
163
+ EVA-attribure: train.txt | 0 | 13799 | 49577 | 55237 | 0
164
+ EVA-attribure: query.txt | 0 | 26702 | 41930 | 48167 | 0
165
+ EVA-attribure: gallery.txt | 0 | 11776 | 32958 | 67687 | 0
166
+ EVA-attribure: ---------------------------------------------------------------
167
+ EVA-attribure: => CCVID loaded
168
+ EVA-attribure: Dataset statistics:
169
+ EVA-attribure: ---------------------------------------------
170
+ EVA-attribure: subset | # ids | # tracklets | # clothes
171
+ EVA-attribure: ---------------------------------------------
172
+ EVA-attribure: train | 75 | 948 | 159
173
+ EVA-attribure: train_dense | 75 | 1409 | 159
174
+ EVA-attribure: query | 151 | 834 | 160
175
+ EVA-attribure: gallery | 151 | 1074 | 252
176
+ EVA-attribure: ---------------------------------------------
177
+ EVA-attribure: total | 226 | 2856 | 480
178
+ EVA-attribure: number of images per tracklet: 27 ~ 410, average 121.8
179
+ EVA-attribure: ---------------------------------------------
180
+ EVA-attribure: ---------------------------------------------------------------
181
+ EVA-attribure: Partition | <32 | '32-64' | '64-128' | '128-256' | '>256'
182
+ EVA-attribure: train.txt | 0 | 13799 | 49577 | 55237 | 0
183
+ EVA-attribure: query.txt | 0 | 26702 | 41930 | 48167 | 0
184
+ EVA-attribure: gallery.txt | 0 | 11776 | 32958 | 67687 | 0
185
+ EVA-attribure: ---------------------------------------------------------------
186
+ EVA-attribure:
187
+ *** Resuming From : logs/CCVID/CCVID_IMG/eva02_l_cloth_best.pth ***
188
+
189
+ EVA-attribure:
190
+ Why load pretrained weights when training e2e
191
+ NoneType: None
192
+ EVA-attribure.train: start training
193
+ EVA-attribure.train: Train Start !!
194
+ EVA-attribure.train: ==> Teacher Training ....
195
+ EVA-attribure.train: ==> Student Training ....
196
+ EVA-attribure.train: ==> Teacher Training ....
197
+ EVA-attribure.train: ==> Student Training ....
198
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
199
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
200
+ EVA-attribure: Extracting features complete in 12m 27s
201
+ EVA-attribure: Distance computing in 0m 0s
202
+ EVA-attribure: Mean Distance : (834, 1074), [-0.24769731 -0.26944467 -0.2668455 -0.24908061 -0.26834354 -0.2681482
203
+ -0.24310766 -0.2688375 -0.26854113 -0.2640222 ] , Mean Query : tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
204
+ 0.0002], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
205
+ 0.0002], device='cuda:0')
206
+ EVA-attribure: Computing CMC and mAP
207
+ EVA-attribure: Results ---------------------------------------------------
208
+ EVA-attribure: top1:86.0% top5:90.4% top10:92.0% top20:93.6% mAP:85.9%
209
+ EVA-attribure: -----------------------------------------------------------
210
+ EVA-attribure: Using 0m 0s
211
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
212
+ EVA-attribure: Results ---------------------------------------------------
213
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
214
+ EVA-attribure: -----------------------------------------------------------
215
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
216
+ EVA-attribure: Results ---------------------------------------------------
217
+ EVA-attribure: top1:83.9% top5:89.0% top10:91.4% top20:93.3% mAP:84.4%
218
+ EVA-attribure: -----------------------------------------------------------
219
+ EVA-attribure.train: ==> Best Rank-1 83.9%, Best Map 84.4% achieved at epoch 2
220
+ EVA-attribure.train: Saving the model Now ....., 0.8393, 0.8440
221
+ EVA-attribure.train: ==> Teacher Training ....
222
+ EVA-attribure.train: ==> Student Training ....
223
+ EVA-attribure.train: ==> Teacher Training ....
224
+ EVA-attribure.train: ==> Student Training ....
225
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
226
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
227
+ EVA-attribure: Extracting features complete in 12m 26s
228
+ EVA-attribure: Distance computing in 0m 0s
229
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21166998 -0.23439142 -0.22547273 -0.21502177 -0.2328893 -0.22969769
230
+ -0.20922586 -0.23316137 -0.22958338 -0.23157628] , Mean Query : tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
231
+ 0.0002], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
232
+ 0.0002], device='cuda:0')
233
+ EVA-attribure: Computing CMC and mAP
234
+ EVA-attribure: Results ---------------------------------------------------
235
+ EVA-attribure: top1:86.2% top5:91.0% top10:92.3% top20:93.8% mAP:87.1%
236
+ EVA-attribure: -----------------------------------------------------------
237
+ EVA-attribure: Using 0m 0s
238
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
239
+ EVA-attribure: Results ---------------------------------------------------
240
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
241
+ EVA-attribure: -----------------------------------------------------------
242
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
243
+ EVA-attribure: Results ---------------------------------------------------
244
+ EVA-attribure: top1:84.3% top5:89.7% top10:92.0% top20:93.4% mAP:85.7%
245
+ EVA-attribure: -----------------------------------------------------------
246
+ EVA-attribure.train: ==> Best Rank-1 84.3%, Best Map 85.7% achieved at epoch 4
247
+ EVA-attribure.train: Saving the model Now ....., 0.8429, 0.8567
248
+ EVA-attribure.train: ==> Teacher Training ....
249
+ EVA-attribure.train: ==> Student Training ....
250
+ EVA-attribure.train: ==> Teacher Training ....
251
+ EVA-attribure.train: ==> Student Training ....
252
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
253
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
254
+ EVA-attribure: Extracting features complete in 12m 26s
255
+ EVA-attribure: Distance computing in 0m 0s
256
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19961916 -0.21816762 -0.21082126 -0.20270222 -0.21739985 -0.21678731
257
+ -0.19717737 -0.21821561 -0.21664658 -0.21950983] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
258
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
259
+ 0.0001], device='cuda:0')
260
+ EVA-attribure: Computing CMC and mAP
261
+ EVA-attribure: Results ---------------------------------------------------
262
+ EVA-attribure: top1:85.9% top5:90.8% top10:92.1% top20:93.8% mAP:86.7%
263
+ EVA-attribure: -----------------------------------------------------------
264
+ EVA-attribure: Using 0m 0s
265
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
266
+ EVA-attribure: Results ---------------------------------------------------
267
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
268
+ EVA-attribure: -----------------------------------------------------------
269
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
270
+ EVA-attribure: Results ---------------------------------------------------
271
+ EVA-attribure: top1:84.1% top5:89.3% top10:91.7% top20:93.4% mAP:85.3%
272
+ EVA-attribure: -----------------------------------------------------------
273
+ EVA-attribure.train: ==> Teacher Training ....
274
+ EVA-attribure.train: ==> Student Training ....
275
+ EVA-attribure.train: ==> Teacher Training ....
276
+ EVA-attribure.train: ==> Student Training ....
277
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
278
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
279
+ EVA-attribure: Extracting features complete in 12m 25s
280
+ EVA-attribure: Distance computing in 0m 0s
281
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20179103 -0.22236693 -0.21449722 -0.2068867 -0.22198264 -0.22087382
282
+ -0.20244981 -0.22367166 -0.22243965 -0.22283244] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
283
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
284
+ 0.0001], device='cuda:0')
285
+ EVA-attribure: Computing CMC and mAP
286
+ EVA-attribure: Results ---------------------------------------------------
287
+ EVA-attribure: top1:86.9% top5:91.4% top10:92.3% top20:93.8% mAP:88.0%
288
+ EVA-attribure: -----------------------------------------------------------
289
+ EVA-attribure: Using 0m 0s
290
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
291
+ EVA-attribure: Results ---------------------------------------------------
292
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
293
+ EVA-attribure: -----------------------------------------------------------
294
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
295
+ EVA-attribure: Results ---------------------------------------------------
296
+ EVA-attribure: top1:85.3% top5:89.9% top10:92.0% top20:93.4% mAP:86.7%
297
+ EVA-attribure: -----------------------------------------------------------
298
+ EVA-attribure.train: ==> Best Rank-1 85.3%, Best Map 86.7% achieved at epoch 8
299
+ EVA-attribure.train: Saving the model Now ....., 0.8525, 0.8673
300
+ EVA-attribure.train: ==> Teacher Training ....
301
+ EVA-attribure.train: ==> Student Training ....
302
+ EVA-attribure.train: ==> Teacher Training ....
303
+ EVA-attribure.train: ==> Student Training ....
304
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
305
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
306
+ EVA-attribure: Extracting features complete in 12m 24s
307
+ EVA-attribure: Distance computing in 0m 0s
308
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19333716 -0.21258062 -0.20448503 -0.19715479 -0.21089673 -0.21127355
309
+ -0.1949473 -0.21338663 -0.21309488 -0.21168542] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
310
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
311
+ 0.0001], device='cuda:0')
312
+ EVA-attribure: Computing CMC and mAP
313
+ EVA-attribure: Results ---------------------------------------------------
314
+ EVA-attribure: top1:88.0% top5:91.4% top10:92.3% top20:93.9% mAP:88.6%
315
+ EVA-attribure: -----------------------------------------------------------
316
+ EVA-attribure: Using 0m 0s
317
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
318
+ EVA-attribure: Results ---------------------------------------------------
319
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
320
+ EVA-attribure: -----------------------------------------------------------
321
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
322
+ EVA-attribure: Results ---------------------------------------------------
323
+ EVA-attribure: top1:86.5% top5:90.0% top10:92.0% top20:93.5% mAP:87.4%
324
+ EVA-attribure: -----------------------------------------------------------
325
+ EVA-attribure.train: ==> Best Rank-1 86.5%, Best Map 87.4% achieved at epoch 10
326
+ EVA-attribure.train: Saving the model Now ....., 0.8645, 0.8739
327
+ EVA-attribure.train: ==> Teacher Training ....
328
+ EVA-attribure.train: ==> Student Training ....
329
+ EVA-attribure.train: ==> Teacher Training ....
330
+ EVA-attribure.train: ==> Student Training ....
331
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
332
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
333
+ EVA-attribure: Extracting features complete in 12m 25s
334
+ EVA-attribure: Distance computing in 0m 0s
335
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19982438 -0.21536987 -0.20189282 -0.20114404 -0.2131594 -0.20483005
336
+ -0.19934252 -0.21293268 -0.20912991 -0.2121066 ] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
337
+ 0.0002], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
338
+ 0.0002], device='cuda:0')
339
+ EVA-attribure: Computing CMC and mAP
340
+ EVA-attribure: Results ---------------------------------------------------
341
+ EVA-attribure: top1:89.0% top5:91.2% top10:92.6% top20:93.8% mAP:88.5%
342
+ EVA-attribure: -----------------------------------------------------------
343
+ EVA-attribure: Using 0m 0s
344
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
345
+ EVA-attribure: Results ---------------------------------------------------
346
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
347
+ EVA-attribure: -----------------------------------------------------------
348
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
349
+ EVA-attribure: Results ---------------------------------------------------
350
+ EVA-attribure: top1:87.3% top5:89.9% top10:92.2% top20:93.5% mAP:87.2%
351
+ EVA-attribure: -----------------------------------------------------------
352
+ EVA-attribure.train: ==> Best Rank-1 87.3%, Best Map 87.4% achieved at epoch 12
353
+ EVA-attribure.train: Saving the model Now ....., 0.8729, 0.8724
354
+ EVA-attribure.train: ==> Teacher Training ....
355
+ EVA-attribure.train: ==> Student Training ....
356
+ EVA-attribure.train: ==> Teacher Training ....
357
+ EVA-attribure.train: ==> Student Training ....
358
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
359
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
360
+ EVA-attribure: Extracting features complete in 12m 26s
361
+ EVA-attribure: Distance computing in 0m 0s
362
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19286442 -0.20939858 -0.19766313 -0.1939049 -0.20669448 -0.20229875
363
+ -0.19246359 -0.2082923 -0.20357865 -0.20353854] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
364
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
365
+ 0.0001], device='cuda:0')
366
+ EVA-attribure: Computing CMC and mAP
367
+ EVA-attribure: Results ---------------------------------------------------
368
+ EVA-attribure: top1:88.2% top5:91.1% top10:92.1% top20:93.8% mAP:88.5%
369
+ EVA-attribure: -----------------------------------------------------------
370
+ EVA-attribure: Using 0m 0s
371
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
372
+ EVA-attribure: Results ---------------------------------------------------
373
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
374
+ EVA-attribure: -----------------------------------------------------------
375
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
376
+ EVA-attribure: Results ---------------------------------------------------
377
+ EVA-attribure: top1:86.6% top5:89.7% top10:91.7% top20:93.5% mAP:87.2%
378
+ EVA-attribure: -----------------------------------------------------------
379
+ EVA-attribure.train: ==> Teacher Training ....
380
+ EVA-attribure.train: ==> Student Training ....
381
+ EVA-attribure.train: ==> Teacher Training ....
382
+ EVA-attribure.train: ==> Student Training ....
383
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
384
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
385
+ EVA-attribure: Extracting features complete in 12m 25s
386
+ EVA-attribure: Distance computing in 0m 0s
387
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19627945 -0.21161307 -0.19864732 -0.19884369 -0.2117708 -0.20147507
388
+ -0.19633155 -0.2120001 -0.20524237 -0.20525794] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
389
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
390
+ 0.0001], device='cuda:0')
391
+ EVA-attribure: Computing CMC and mAP
392
+ EVA-attribure: Results ---------------------------------------------------
393
+ EVA-attribure: top1:89.3% top5:91.1% top10:92.3% top20:93.9% mAP:89.3%
394
+ EVA-attribure: -----------------------------------------------------------
395
+ EVA-attribure: Using 0m 0s
396
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
397
+ EVA-attribure: Results ---------------------------------------------------
398
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
399
+ EVA-attribure: -----------------------------------------------------------
400
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
401
+ EVA-attribure: Results ---------------------------------------------------
402
+ EVA-attribure: top1:87.9% top5:89.9% top10:91.8% top20:93.8% mAP:88.1%
403
+ EVA-attribure: -----------------------------------------------------------
404
+ EVA-attribure.train: ==> Best Rank-1 87.9%, Best Map 88.1% achieved at epoch 16
405
+ EVA-attribure.train: Saving the model Now ....., 0.8789, 0.8813
406
+ EVA-attribure.train: ==> Teacher Training ....
407
+ EVA-attribure.train: ==> Student Training ....
408
+ EVA-attribure.train: ==> Teacher Training ....
409
+ EVA-attribure.train: ==> Student Training ....
410
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
411
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
412
+ EVA-attribure: Extracting features complete in 12m 13s
413
+ EVA-attribure: Distance computing in 0m 0s
414
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19355991 -0.21260658 -0.20227864 -0.19590545 -0.21110472 -0.20793486
415
+ -0.1935647 -0.21252955 -0.2071554 -0.20187746] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
416
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
417
+ 0.0001], device='cuda:0')
418
+ EVA-attribure: Computing CMC and mAP
419
+ EVA-attribure: Results ---------------------------------------------------
420
+ EVA-attribure: top1:88.6% top5:91.4% top10:92.4% top20:93.8% mAP:88.5%
421
+ EVA-attribure: -----------------------------------------------------------
422
+ EVA-attribure: Using 0m 0s
423
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
424
+ EVA-attribure: Results ---------------------------------------------------
425
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
426
+ EVA-attribure: -----------------------------------------------------------
427
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
428
+ EVA-attribure: Results ---------------------------------------------------
429
+ EVA-attribure: top1:86.9% top5:89.9% top10:91.7% top20:93.5% mAP:87.1%
430
+ EVA-attribure: -----------------------------------------------------------
431
+ EVA-attribure.train: ==> Teacher Training ....
432
+ EVA-attribure.train: ==> Student Training ....
433
+ EVA-attribure.train: ==> Teacher Training ....
434
+ EVA-attribure.train: ==> Student Training ....
435
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
436
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
437
+ EVA-attribure: Extracting features complete in 12m 30s
438
+ EVA-attribure: Distance computing in 0m 0s
439
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19600302 -0.21498318 -0.2040906 -0.19738233 -0.21350768 -0.20958823
440
+ -0.19640255 -0.2141574 -0.20801838 -0.20274547] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
441
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
442
+ 0.0001], device='cuda:0')
443
+ EVA-attribure: Computing CMC and mAP
444
+ EVA-attribure: Results ---------------------------------------------------
445
+ EVA-attribure: top1:89.6% top5:91.5% top10:92.9% top20:94.1% mAP:90.0%
446
+ EVA-attribure: -----------------------------------------------------------
447
+ EVA-attribure: Using 0m 0s
448
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
449
+ EVA-attribure: Results ---------------------------------------------------
450
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
451
+ EVA-attribure: -----------------------------------------------------------
452
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
453
+ EVA-attribure: Results ---------------------------------------------------
454
+ EVA-attribure: top1:88.0% top5:90.2% top10:92.7% top20:94.1% mAP:88.8%
455
+ EVA-attribure: -----------------------------------------------------------
456
+ EVA-attribure.train: ==> Best Rank-1 88.0%, Best Map 88.8% achieved at epoch 20
457
+ EVA-attribure.train: Saving the model Now ....., 0.8801, 0.8881
458
+ EVA-attribure.train: ==> Teacher Training ....
459
+ EVA-attribure.train: ==> Student Training ....
460
+ EVA-attribure.train: ==> Teacher Training ....
461
+ EVA-attribure.train: ==> Student Training ....
462
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
463
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
464
+ EVA-attribure: Extracting features complete in 12m 24s
465
+ EVA-attribure: Distance computing in 0m 0s
466
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19215086 -0.21109633 -0.19980139 -0.19598445 -0.20945704 -0.20509842
467
+ -0.19378528 -0.20958017 -0.20393753 -0.20356773] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
468
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
469
+ 0.0001], device='cuda:0')
470
+ EVA-attribure: Computing CMC and mAP
471
+ EVA-attribure: Results ---------------------------------------------------
472
+ EVA-attribure: top1:89.6% top5:91.2% top10:92.6% top20:94.0% mAP:89.9%
473
+ EVA-attribure: -----------------------------------------------------------
474
+ EVA-attribure: Using 0m 0s
475
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
476
+ EVA-attribure: Results ---------------------------------------------------
477
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
478
+ EVA-attribure: -----------------------------------------------------------
479
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
480
+ EVA-attribure: Results ---------------------------------------------------
481
+ EVA-attribure: top1:88.1% top5:90.0% top10:92.4% top20:94.0% mAP:88.7%
482
+ EVA-attribure: -----------------------------------------------------------
483
+ EVA-attribure.train: ==> Best Rank-1 88.1%, Best Map 88.8% achieved at epoch 22
484
+ EVA-attribure.train: Saving the model Now ....., 0.8813, 0.8870
485
+ EVA-attribure.train: ==> Teacher Training ....
486
+ EVA-attribure.train: ==> Student Training ....
487
+ EVA-attribure.train: ==> Teacher Training ....
488
+ EVA-attribure.train: ==> Student Training ....
489
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
490
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
491
+ EVA-attribure: Extracting features complete in 12m 26s
492
+ EVA-attribure: Distance computing in 0m 0s
493
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19838497 -0.21624474 -0.2050694 -0.20181057 -0.2149725 -0.2098718
494
+ -0.20001005 -0.21549447 -0.20949845 -0.20715332] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
495
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
496
+ 0.0001], device='cuda:0')
497
+ EVA-attribure: Computing CMC and mAP
498
+ EVA-attribure: Results ---------------------------------------------------
499
+ EVA-attribure: top1:88.7% top5:90.9% top10:92.4% top20:93.9% mAP:89.1%
500
+ EVA-attribure: -----------------------------------------------------------
501
+ EVA-attribure: Using 0m 0s
502
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
503
+ EVA-attribure: Results ---------------------------------------------------
504
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
505
+ EVA-attribure: -----------------------------------------------------------
506
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
507
+ EVA-attribure: Results ---------------------------------------------------
508
+ EVA-attribure: top1:87.5% top5:90.2% top10:92.4% top20:93.9% mAP:88.1%
509
+ EVA-attribure: -----------------------------------------------------------
510
+ EVA-attribure.train: ==> Teacher Training ....
511
+ EVA-attribure.train: ==> Student Training ....
512
+ EVA-attribure.train: ==> Teacher Training ....
513
+ EVA-attribure.train: ==> Student Training ....
514
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
515
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
516
+ EVA-attribure: Extracting features complete in 12m 26s
517
+ EVA-attribure: Distance computing in 0m 0s
518
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19233902 -0.21333824 -0.20094402 -0.19722241 -0.21123368 -0.2038171
519
+ -0.19421671 -0.2111488 -0.20538591 -0.20303032] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
520
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
521
+ 0.0001], device='cuda:0')
522
+ EVA-attribure: Computing CMC and mAP
523
+ EVA-attribure: Results ---------------------------------------------------
524
+ EVA-attribure: top1:89.8% top5:91.4% top10:93.0% top20:94.0% mAP:90.5%
525
+ EVA-attribure: -----------------------------------------------------------
526
+ EVA-attribure: Using 0m 0s
527
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
528
+ EVA-attribure: Results ---------------------------------------------------
529
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
530
+ EVA-attribure: -----------------------------------------------------------
531
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
532
+ EVA-attribure: Results ---------------------------------------------------
533
+ EVA-attribure: top1:88.6% top5:90.6% top10:92.9% top20:94.0% mAP:89.4%
534
+ EVA-attribure: -----------------------------------------------------------
535
+ EVA-attribure.train: ==> Best Rank-1 88.6%, Best Map 89.4% achieved at epoch 26
536
+ EVA-attribure.train: Saving the model Now ....., 0.8861, 0.8943
537
+ EVA-attribure.train: ==> Teacher Training ....
538
+ EVA-attribure.train: ==> Student Training ....
539
+ EVA-attribure.train: ==> Teacher Training ....
540
+ EVA-attribure.train: ==> Student Training ....
541
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
542
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
543
+ EVA-attribure: Extracting features complete in 12m 25s
544
+ EVA-attribure: Distance computing in 0m 0s
545
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20289192 -0.22350441 -0.21396028 -0.20664486 -0.22010793 -0.21550402
546
+ -0.20569092 -0.2217946 -0.21677792 -0.20888641] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
547
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
548
+ 0.0001], device='cuda:0')
549
+ EVA-attribure: Computing CMC and mAP
550
+ EVA-attribure: Results ---------------------------------------------------
551
+ EVA-attribure: top1:89.8% top5:91.4% top10:92.9% top20:94.0% mAP:90.4%
552
+ EVA-attribure: -----------------------------------------------------------
553
+ EVA-attribure: Using 0m 0s
554
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
555
+ EVA-attribure: Results ---------------------------------------------------
556
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
557
+ EVA-attribure: -----------------------------------------------------------
558
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
559
+ EVA-attribure: Results ---------------------------------------------------
560
+ EVA-attribure: top1:88.6% top5:90.6% top10:92.9% top20:94.0% mAP:89.4%
561
+ EVA-attribure: -----------------------------------------------------------
562
+ EVA-attribure.train: ==> Teacher Training ....
563
+ EVA-attribure.train: ==> Student Training ....
564
+ EVA-attribure.train: ==> Teacher Training ....
565
+ EVA-attribure.train: ==> Student Training ....
566
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
567
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
568
+ EVA-attribure: Extracting features complete in 12m 25s
569
+ EVA-attribure: Distance computing in 0m 0s
570
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19655281 -0.21759759 -0.20796433 -0.19899338 -0.21425858 -0.21015573
571
+ -0.1995696 -0.21575394 -0.21175976 -0.20245436] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
572
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002,
573
+ 0.0001], device='cuda:0')
574
+ EVA-attribure: Computing CMC and mAP
575
+ EVA-attribure: Results ---------------------------------------------------
576
+ EVA-attribure: top1:89.9% top5:91.5% top10:92.9% top20:94.0% mAP:90.7%
577
+ EVA-attribure: -----------------------------------------------------------
578
+ EVA-attribure: Using 0m 0s
579
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
580
+ EVA-attribure: Results ---------------------------------------------------
581
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
582
+ EVA-attribure: -----------------------------------------------------------
583
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
584
+ EVA-attribure: Results ---------------------------------------------------
585
+ EVA-attribure: top1:88.6% top5:90.9% top10:92.8% top20:94.0% mAP:89.7%
586
+ EVA-attribure: -----------------------------------------------------------
587
+ EVA-attribure.train: ==> Teacher Training ....
588
+ EVA-attribure.train: ==> Student Training ....
589
+ EVA-attribure.train: ==> Teacher Training ....
590
+ EVA-attribure.train: ==> Student Training ....
591
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
592
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
593
+ EVA-attribure: Extracting features complete in 12m 27s
594
+ EVA-attribure: Distance computing in 0m 0s
595
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19762269 -0.22000533 -0.21003297 -0.20183666 -0.21651858 -0.21076883
596
+ -0.20099273 -0.21803768 -0.21248558 -0.20277175] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
597
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
598
+ 0.0001], device='cuda:0')
599
+ EVA-attribure: Computing CMC and mAP
600
+ EVA-attribure: Results ---------------------------------------------------
601
+ EVA-attribure: top1:90.3% top5:91.5% top10:93.0% top20:94.0% mAP:90.7%
602
+ EVA-attribure: -----------------------------------------------------------
603
+ EVA-attribure: Using 0m 0s
604
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
605
+ EVA-attribure: Results ---------------------------------------------------
606
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
607
+ EVA-attribure: -----------------------------------------------------------
608
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
609
+ EVA-attribure: Results ---------------------------------------------------
610
+ EVA-attribure: top1:89.0% top5:90.9% top10:92.8% top20:94.0% mAP:89.7%
611
+ EVA-attribure: -----------------------------------------------------------
612
+ EVA-attribure.train: ==> Best Rank-1 89.0%, Best Map 89.7% achieved at epoch 32
613
+ EVA-attribure.train: Saving the model Now ....., 0.8897, 0.8969
614
+ EVA-attribure.train: ==> Teacher Training ....
615
+ EVA-attribure.train: ==> Student Training ....
616
+ EVA-attribure.train: ==> Teacher Training ....
617
+ EVA-attribure.train: ==> Student Training ....
618
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
619
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
620
+ EVA-attribure: Extracting features complete in 12m 30s
621
+ EVA-attribure: Distance computing in 0m 0s
622
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19917184 -0.22038499 -0.20986092 -0.20218371 -0.21573983 -0.21129692
623
+ -0.20173831 -0.21729952 -0.21326122 -0.20346648] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
624
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
625
+ 0.0001], device='cuda:0')
626
+ EVA-attribure: Computing CMC and mAP
627
+ EVA-attribure: Results ---------------------------------------------------
628
+ EVA-attribure: top1:90.6% top5:91.5% top10:93.2% top20:94.0% mAP:91.2%
629
+ EVA-attribure: -----------------------------------------------------------
630
+ EVA-attribure: Using 0m 0s
631
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
632
+ EVA-attribure: Results ---------------------------------------------------
633
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
634
+ EVA-attribure: -----------------------------------------------------------
635
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
636
+ EVA-attribure: Results ---------------------------------------------------
637
+ EVA-attribure: top1:89.3% top5:90.8% top10:92.9% top20:94.0% mAP:90.2%
638
+ EVA-attribure: -----------------------------------------------------------
639
+ EVA-attribure.train: ==> Best Rank-1 89.3%, Best Map 90.2% achieved at epoch 34
640
+ EVA-attribure.train: Saving the model Now ....., 0.8933, 0.9016
641
+ EVA-attribure.train: ==> Teacher Training ....
642
+ EVA-attribure.train: ==> Student Training ....
643
+ EVA-attribure.train: ==> Teacher Training ....
644
+ EVA-attribure.train: ==> Student Training ....
645
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
646
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
647
+ EVA-attribure: Extracting features complete in 12m 13s
648
+ EVA-attribure: Distance computing in 0m 0s
649
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19988889 -0.22072141 -0.21040423 -0.20169441 -0.2156358 -0.21071933
650
+ -0.20238568 -0.21833114 -0.21507655 -0.20284075] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
651
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
652
+ 0.0001], device='cuda:0')
653
+ EVA-attribure: Computing CMC and mAP
654
+ EVA-attribure: Results ---------------------------------------------------
655
+ EVA-attribure: top1:90.8% top5:91.5% top10:93.0% top20:94.0% mAP:91.3%
656
+ EVA-attribure: -----------------------------------------------------------
657
+ EVA-attribure: Using 0m 0s
658
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
659
+ EVA-attribure: Results ---------------------------------------------------
660
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
661
+ EVA-attribure: -----------------------------------------------------------
662
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
663
+ EVA-attribure: Results ---------------------------------------------------
664
+ EVA-attribure: top1:89.4% top5:90.9% top10:92.9% top20:94.0% mAP:90.3%
665
+ EVA-attribure: -----------------------------------------------------------
666
+ EVA-attribure.train: ==> Best Rank-1 89.4%, Best Map 90.3% achieved at epoch 36
667
+ EVA-attribure.train: Saving the model Now ....., 0.8945, 0.9027
668
+ EVA-attribure.train: ==> Teacher Training ....
669
+ EVA-attribure.train: ==> Student Training ....
670
+ EVA-attribure.train: ==> Teacher Training ....
671
+ EVA-attribure.train: ==> Student Training ....
672
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
673
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
674
+ EVA-attribure: Extracting features complete in 12m 26s
675
+ EVA-attribure: Distance computing in 0m 0s
676
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2007648 -0.22090866 -0.21026638 -0.20235877 -0.21601638 -0.21022667
677
+ -0.20371717 -0.21828362 -0.21468791 -0.20205396] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
678
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
679
+ 0.0001], device='cuda:0')
680
+ EVA-attribure: Computing CMC and mAP
681
+ EVA-attribure: Results ---------------------------------------------------
682
+ EVA-attribure: top1:91.1% top5:91.5% top10:93.2% top20:94.0% mAP:91.7%
683
+ EVA-attribure: -----------------------------------------------------------
684
+ EVA-attribure: Using 0m 0s
685
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
686
+ EVA-attribure: Results ---------------------------------------------------
687
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
688
+ EVA-attribure: -----------------------------------------------------------
689
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
690
+ EVA-attribure: Results ---------------------------------------------------
691
+ EVA-attribure: top1:89.8% top5:91.0% top10:92.9% top20:94.0% mAP:90.6%
692
+ EVA-attribure: -----------------------------------------------------------
693
+ EVA-attribure.train: ==> Best Rank-1 89.8%, Best Map 90.6% achieved at epoch 38
694
+ EVA-attribure.train: Saving the model Now ....., 0.8981, 0.9061
695
+ EVA-attribure.train: ==> Teacher Training ....
696
+ EVA-attribure.train: ==> Student Training ....
697
+ EVA-attribure.train: ==> Teacher Training ....
698
+ EVA-attribure.train: ==> Student Training ....
699
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
700
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
701
+ EVA-attribure: Extracting features complete in 12m 26s
702
+ EVA-attribure: Distance computing in 0m 0s
703
+ EVA-attribure: Mean Distance : (834, 1074), [-0.1991053 -0.21884097 -0.20844916 -0.20183793 -0.21453165 -0.20972642
704
+ -0.20337991 -0.21718732 -0.21245489 -0.20170923] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
705
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
706
+ 0.0001], device='cuda:0')
707
+ EVA-attribure: Computing CMC and mAP
708
+ EVA-attribure: Results ---------------------------------------------------
709
+ EVA-attribure: top1:91.1% top5:91.5% top10:93.6% top20:93.9% mAP:91.8%
710
+ EVA-attribure: -----------------------------------------------------------
711
+ EVA-attribure: Using 0m 0s
712
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
713
+ EVA-attribure: Results ---------------------------------------------------
714
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
715
+ EVA-attribure: -----------------------------------------------------------
716
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
717
+ EVA-attribure: Results ---------------------------------------------------
718
+ EVA-attribure: top1:89.8% top5:90.8% top10:93.4% top20:93.9% mAP:90.7%
719
+ EVA-attribure: -----------------------------------------------------------
720
+ EVA-attribure.train: ==> Teacher Training ....
721
+ EVA-attribure.train: ==> Student Training ....
722
+ EVA-attribure.train: ==> Teacher Training ....
723
+ EVA-attribure.train: ==> Student Training ....
724
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
725
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
726
+ EVA-attribure: Extracting features complete in 12m 26s
727
+ EVA-attribure: Distance computing in 0m 0s
728
+ EVA-attribure: Mean Distance : (834, 1074), [-0.19867983 -0.2212026 -0.21106015 -0.20055479 -0.2153733 -0.21053597
729
+ -0.20293616 -0.21961471 -0.21414019 -0.202192 ] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
730
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
731
+ 0.0001], device='cuda:0')
732
+ EVA-attribure: Computing CMC and mAP
733
+ EVA-attribure: Results ---------------------------------------------------
734
+ EVA-attribure: top1:91.4% top5:92.0% top10:93.6% top20:93.9% mAP:91.7%
735
+ EVA-attribure: -----------------------------------------------------------
736
+ EVA-attribure: Using 0m 0s
737
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
738
+ EVA-attribure: Results ---------------------------------------------------
739
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
740
+ EVA-attribure: -----------------------------------------------------------
741
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
742
+ EVA-attribure: Results ---------------------------------------------------
743
+ EVA-attribure: top1:90.3% top5:91.4% top10:93.5% top20:93.9% mAP:90.7%
744
+ EVA-attribure: -----------------------------------------------------------
745
+ EVA-attribure.train: ==> Best Rank-1 90.3%, Best Map 90.7% achieved at epoch 42
746
+ EVA-attribure.train: Saving the model Now ....., 0.9029, 0.9074
747
+ EVA-attribure.train: ==> Teacher Training ....
748
+ EVA-attribure.train: ==> Student Training ....
749
+ EVA-attribure.train: ==> Teacher Training ....
750
+ EVA-attribure.train: ==> Student Training ....
751
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
752
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
753
+ EVA-attribure: Extracting features complete in 12m 25s
754
+ EVA-attribure: Distance computing in 0m 0s
755
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20258223 -0.22301075 -0.21122538 -0.20378838 -0.21744205 -0.21160859
756
+ -0.20683615 -0.22129561 -0.21502475 -0.20208798] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
757
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
758
+ 0.0001], device='cuda:0')
759
+ EVA-attribure: Computing CMC and mAP
760
+ EVA-attribure: Results ---------------------------------------------------
761
+ EVA-attribure: top1:91.1% top5:91.5% top10:93.0% top20:93.9% mAP:91.8%
762
+ EVA-attribure: -----------------------------------------------------------
763
+ EVA-attribure: Using 0m 0s
764
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
765
+ EVA-attribure: Results ---------------------------------------------------
766
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
767
+ EVA-attribure: -----------------------------------------------------------
768
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
769
+ EVA-attribure: Results ---------------------------------------------------
770
+ EVA-attribure: top1:90.3% top5:91.0% top10:92.9% top20:93.9% mAP:90.9%
771
+ EVA-attribure: -----------------------------------------------------------
772
+ EVA-attribure.train: ==> Teacher Training ....
773
+ EVA-attribure.train: ==> Student Training ....
774
+ EVA-attribure.train: ==> Teacher Training ....
775
+ EVA-attribure.train: ==> Student Training ....
776
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
777
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
778
+ EVA-attribure: Extracting features complete in 12m 26s
779
+ EVA-attribure: Distance computing in 0m 0s
780
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20427069 -0.22599487 -0.21405391 -0.20579067 -0.22111861 -0.21213984
781
+ -0.20856202 -0.22375625 -0.21731643 -0.20318936] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
782
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
783
+ 0.0001], device='cuda:0')
784
+ EVA-attribure: Computing CMC and mAP
785
+ EVA-attribure: Results ---------------------------------------------------
786
+ EVA-attribure: top1:91.0% top5:92.0% top10:93.4% top20:93.9% mAP:91.8%
787
+ EVA-attribure: -----------------------------------------------------------
788
+ EVA-attribure: Using 0m 0s
789
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
790
+ EVA-attribure: Results ---------------------------------------------------
791
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
792
+ EVA-attribure: -----------------------------------------------------------
793
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
794
+ EVA-attribure: Results ---------------------------------------------------
795
+ EVA-attribure: top1:89.8% top5:91.4% top10:93.3% top20:93.9% mAP:90.8%
796
+ EVA-attribure: -----------------------------------------------------------
797
+ EVA-attribure.train: ==> Teacher Training ....
798
+ EVA-attribure.train: ==> Student Training ....
799
+ EVA-attribure.train: ==> Teacher Training ....
800
+ EVA-attribure.train: ==> Student Training ....
801
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
802
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
803
+ EVA-attribure: Extracting features complete in 12m 26s
804
+ EVA-attribure: Distance computing in 0m 0s
805
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20003039 -0.22018051 -0.20772567 -0.20193002 -0.21584944 -0.2053007
806
+ -0.2051253 -0.2184594 -0.21093611 -0.20070086] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
807
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
808
+ 0.0001], device='cuda:0')
809
+ EVA-attribure: Computing CMC and mAP
810
+ EVA-attribure: Results ---------------------------------------------------
811
+ EVA-attribure: top1:91.1% top5:92.0% top10:93.0% top20:93.9% mAP:91.5%
812
+ EVA-attribure: -----------------------------------------------------------
813
+ EVA-attribure: Using 0m 0s
814
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
815
+ EVA-attribure: Results ---------------------------------------------------
816
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
817
+ EVA-attribure: -----------------------------------------------------------
818
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
819
+ EVA-attribure: Results ---------------------------------------------------
820
+ EVA-attribure: top1:90.2% top5:91.5% top10:92.9% top20:93.9% mAP:90.6%
821
+ EVA-attribure: -----------------------------------------------------------
822
+ EVA-attribure.train: ==> Teacher Training ....
823
+ EVA-attribure.train: ==> Student Training ....
824
+ EVA-attribure.train: ==> Teacher Training ....
825
+ EVA-attribure.train: ==> Student Training ....
826
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
827
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
828
+ EVA-attribure: Extracting features complete in 12m 26s
829
+ EVA-attribure: Distance computing in 0m 0s
830
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2071178 -0.22493245 -0.21208054 -0.20860605 -0.2206089 -0.21014346
831
+ -0.21173076 -0.22269449 -0.21604745 -0.20719759] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
832
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
833
+ 0.0001], device='cuda:0')
834
+ EVA-attribure: Computing CMC and mAP
835
+ EVA-attribure: Results ---------------------------------------------------
836
+ EVA-attribure: top1:91.0% top5:92.0% top10:93.6% top20:94.0% mAP:91.3%
837
+ EVA-attribure: -----------------------------------------------------------
838
+ EVA-attribure: Using 0m 0s
839
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
840
+ EVA-attribure: Results ---------------------------------------------------
841
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
842
+ EVA-attribure: -----------------------------------------------------------
843
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
844
+ EVA-attribure: Results ---------------------------------------------------
845
+ EVA-attribure: top1:89.9% top5:91.4% top10:93.5% top20:94.0% mAP:90.4%
846
+ EVA-attribure: -----------------------------------------------------------
847
+ EVA-attribure.train: ==> Teacher Training ....
848
+ EVA-attribure.train: ==> Student Training ....
849
+ EVA-attribure.train: ==> Teacher Training ....
850
+ EVA-attribure.train: ==> Student Training ....
851
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
852
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
853
+ EVA-attribure: Extracting features complete in 12m 25s
854
+ EVA-attribure: Distance computing in 0m 0s
855
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21078369 -0.23149595 -0.21767756 -0.2127909 -0.22736692 -0.21571292
856
+ -0.21581937 -0.23031485 -0.22148167 -0.21195483] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
857
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
858
+ 0.0001], device='cuda:0')
859
+ EVA-attribure: Computing CMC and mAP
860
+ EVA-attribure: Results ---------------------------------------------------
861
+ EVA-attribure: top1:90.3% top5:92.0% top10:93.6% top20:93.9% mAP:91.5%
862
+ EVA-attribure: -----------------------------------------------------------
863
+ EVA-attribure: Using 0m 0s
864
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
865
+ EVA-attribure: Results ---------------------------------------------------
866
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
867
+ EVA-attribure: -----------------------------------------------------------
868
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
869
+ EVA-attribure: Results ---------------------------------------------------
870
+ EVA-attribure: top1:89.2% top5:91.4% top10:93.5% top20:93.9% mAP:90.5%
871
+ EVA-attribure: -----------------------------------------------------------
872
+ EVA-attribure.train: ==> Teacher Training ....
873
+ EVA-attribure.train: ==> Student Training ....
874
+ EVA-attribure.train: ==> Teacher Training ....
875
+ EVA-attribure.train: ==> Student Training ....
876
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
877
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
878
+ EVA-attribure: Extracting features complete in 12m 25s
879
+ EVA-attribure: Distance computing in 0m 0s
880
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20944323 -0.23074512 -0.21660303 -0.21139698 -0.22588632 -0.21373416
881
+ -0.2139053 -0.22900234 -0.22124623 -0.21061112] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
882
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
883
+ 0.0001], device='cuda:0')
884
+ EVA-attribure: Computing CMC and mAP
885
+ EVA-attribure: Results ---------------------------------------------------
886
+ EVA-attribure: top1:90.2% top5:92.1% top10:93.8% top20:93.9% mAP:91.4%
887
+ EVA-attribure: -----------------------------------------------------------
888
+ EVA-attribure: Using 0m 0s
889
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
890
+ EVA-attribure: Results ---------------------------------------------------
891
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
892
+ EVA-attribure: -----------------------------------------------------------
893
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
894
+ EVA-attribure: Results ---------------------------------------------------
895
+ EVA-attribure: top1:89.1% top5:91.6% top10:93.6% top20:93.9% mAP:90.4%
896
+ EVA-attribure: -----------------------------------------------------------
897
+ EVA-attribure.train: ==> Teacher Training ....
898
+ EVA-attribure.train: ==> Student Training ....
899
+ EVA-attribure.train: ==> Teacher Training ....
900
+ EVA-attribure.train: ==> Student Training ....
901
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
902
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
903
+ EVA-attribure: Extracting features complete in 12m 25s
904
+ EVA-attribure: Distance computing in 0m 0s
905
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2046912 -0.22591269 -0.21127653 -0.20628245 -0.22144523 -0.21064247
906
+ -0.20895186 -0.2241107 -0.21602699 -0.2071529 ] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
907
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
908
+ 0.0001], device='cuda:0')
909
+ EVA-attribure: Computing CMC and mAP
910
+ EVA-attribure: Results ---------------------------------------------------
911
+ EVA-attribure: top1:90.9% top5:91.8% top10:93.8% top20:93.9% mAP:91.4%
912
+ EVA-attribure: -----------------------------------------------------------
913
+ EVA-attribure: Using 0m 0s
914
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
915
+ EVA-attribure: Results ---------------------------------------------------
916
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
917
+ EVA-attribure: -----------------------------------------------------------
918
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
919
+ EVA-attribure: Results ---------------------------------------------------
920
+ EVA-attribure: top1:90.0% top5:91.4% top10:93.6% top20:93.9% mAP:90.6%
921
+ EVA-attribure: -----------------------------------------------------------
922
+ EVA-attribure.train: ==> Teacher Training ....
923
+ EVA-attribure.train: ==> Student Training ....
924
+ EVA-attribure.train: ==> Teacher Training ....
925
+ EVA-attribure.train: ==> Student Training ....
926
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
927
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
928
+ EVA-attribure: Extracting features complete in 12m 27s
929
+ EVA-attribure: Distance computing in 0m 0s
930
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2141051 -0.2363185 -0.2223109 -0.21658073 -0.23195533 -0.22128704
931
+ -0.21918157 -0.23607329 -0.22682977 -0.21587715] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
932
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
933
+ 0.0001], device='cuda:0')
934
+ EVA-attribure: Computing CMC and mAP
935
+ EVA-attribure: Results ---------------------------------------------------
936
+ EVA-attribure: top1:90.8% top5:92.0% top10:93.8% top20:94.0% mAP:91.5%
937
+ EVA-attribure: -----------------------------------------------------------
938
+ EVA-attribure: Using 0m 0s
939
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
940
+ EVA-attribure: Results ---------------------------------------------------
941
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
942
+ EVA-attribure: -----------------------------------------------------------
943
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
944
+ EVA-attribure: Results ---------------------------------------------------
945
+ EVA-attribure: top1:89.9% top5:91.5% top10:93.6% top20:94.0% mAP:90.6%
946
+ EVA-attribure: -----------------------------------------------------------
947
+ EVA-attribure.train: ==> Teacher Training ....
948
+ EVA-attribure.train: ==> Student Training ....
949
+ EVA-attribure.train: ==> Teacher Training ....
950
+ EVA-attribure.train: ==> Student Training ....
951
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
952
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
953
+ EVA-attribure: Extracting features complete in 12m 27s
954
+ EVA-attribure: Distance computing in 0m 0s
955
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20744948 -0.22740301 -0.21397954 -0.20913096 -0.22360489 -0.21176414
956
+ -0.21196672 -0.22697778 -0.2185743 -0.2083378 ] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
957
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
958
+ 0.0001], device='cuda:0')
959
+ EVA-attribure: Computing CMC and mAP
960
+ EVA-attribure: Results ---------------------------------------------------
961
+ EVA-attribure: top1:91.1% top5:92.0% top10:93.8% top20:94.0% mAP:91.5%
962
+ EVA-attribure: -----------------------------------------------------------
963
+ EVA-attribure: Using 0m 0s
964
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
965
+ EVA-attribure: Results ---------------------------------------------------
966
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
967
+ EVA-attribure: -----------------------------------------------------------
968
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
969
+ EVA-attribure: Results ---------------------------------------------------
970
+ EVA-attribure: top1:90.3% top5:91.5% top10:93.6% top20:94.0% mAP:90.6%
971
+ EVA-attribure: -----------------------------------------------------------
972
+ EVA-attribure.train: ==> Teacher Training ....
973
+ EVA-attribure.train: ==> Student Training ....
974
+ EVA-attribure.train: ==> Teacher Training ....
975
+ EVA-attribure.train: ==> Student Training ....
976
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
977
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
978
+ EVA-attribure: Extracting features complete in 12m 26s
979
+ EVA-attribure: Distance computing in 0m 0s
980
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21010579 -0.23006411 -0.21551466 -0.21144603 -0.22609787 -0.21209757
981
+ -0.2137248 -0.22884774 -0.21981306 -0.21033832] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
982
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
983
+ 0.0001], device='cuda:0')
984
+ EVA-attribure: Computing CMC and mAP
985
+ EVA-attribure: Results ---------------------------------------------------
986
+ EVA-attribure: top1:90.5% top5:92.0% top10:93.8% top20:93.9% mAP:91.0%
987
+ EVA-attribure: -----------------------------------------------------------
988
+ EVA-attribure: Using 0m 0s
989
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
990
+ EVA-attribure: Results ---------------------------------------------------
991
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
992
+ EVA-attribure: -----------------------------------------------------------
993
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
994
+ EVA-attribure: Results ---------------------------------------------------
995
+ EVA-attribure: top1:89.7% top5:91.5% top10:93.6% top20:93.9% mAP:90.2%
996
+ EVA-attribure: -----------------------------------------------------------
997
+ EVA-attribure.train: ==> Teacher Training ....
998
+ EVA-attribure.train: ==> Student Training ....
999
+ EVA-attribure.train: ==> Teacher Training ....
1000
+ EVA-attribure.train: ==> Student Training ....
1001
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1002
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1003
+ EVA-attribure: Extracting features complete in 12m 26s
1004
+ EVA-attribure: Distance computing in 0m 0s
1005
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21132931 -0.23147546 -0.21606588 -0.21318837 -0.22740005 -0.2147094
1006
+ -0.21508133 -0.2301819 -0.22096486 -0.21176468] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1007
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1008
+ 0.0001], device='cuda:0')
1009
+ EVA-attribure: Computing CMC and mAP
1010
+ EVA-attribure: Results ---------------------------------------------------
1011
+ EVA-attribure: top1:90.5% top5:92.0% top10:93.8% top20:93.9% mAP:90.8%
1012
+ EVA-attribure: -----------------------------------------------------------
1013
+ EVA-attribure: Using 0m 0s
1014
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1015
+ EVA-attribure: Results ---------------------------------------------------
1016
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1017
+ EVA-attribure: -----------------------------------------------------------
1018
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1019
+ EVA-attribure: Results ---------------------------------------------------
1020
+ EVA-attribure: top1:89.8% top5:91.5% top10:93.6% top20:93.9% mAP:90.0%
1021
+ EVA-attribure: -----------------------------------------------------------
1022
+ EVA-attribure.train: ==> Teacher Training ....
1023
+ EVA-attribure.train: ==> Student Training ....
1024
+ EVA-attribure.train: ==> Teacher Training ....
1025
+ EVA-attribure.train: ==> Student Training ....
1026
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1027
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1028
+ EVA-attribure: Extracting features complete in 12m 26s
1029
+ EVA-attribure: Distance computing in 0m 0s
1030
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20947438 -0.2314307 -0.21530941 -0.2111264 -0.22629781 -0.21199468
1031
+ -0.21365239 -0.22995219 -0.21955213 -0.21001677] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1032
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1033
+ 0.0001], device='cuda:0')
1034
+ EVA-attribure: Computing CMC and mAP
1035
+ EVA-attribure: Results ---------------------------------------------------
1036
+ EVA-attribure: top1:90.3% top5:92.1% top10:93.9% top20:94.0% mAP:90.8%
1037
+ EVA-attribure: -----------------------------------------------------------
1038
+ EVA-attribure: Using 0m 0s
1039
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1040
+ EVA-attribure: Results ---------------------------------------------------
1041
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1042
+ EVA-attribure: -----------------------------------------------------------
1043
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1044
+ EVA-attribure: Results ---------------------------------------------------
1045
+ EVA-attribure: top1:89.6% top5:91.6% top10:93.8% top20:94.0% mAP:90.1%
1046
+ EVA-attribure: -----------------------------------------------------------
1047
+ EVA-attribure.train: ==> Teacher Training ....
1048
+ EVA-attribure.train: ==> Student Training ....
1049
+ EVA-attribure.train: ==> Teacher Training ....
1050
+ EVA-attribure.train: ==> Student Training ....
1051
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1052
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1053
+ EVA-attribure: Extracting features complete in 12m 26s
1054
+ EVA-attribure: Distance computing in 0m 0s
1055
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20917112 -0.22893745 -0.21343894 -0.2110756 -0.22453667 -0.21080992
1056
+ -0.21373034 -0.22786516 -0.21855876 -0.20977062] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1057
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1058
+ 0.0001], device='cuda:0')
1059
+ EVA-attribure: Computing CMC and mAP
1060
+ EVA-attribure: Results ---------------------------------------------------
1061
+ EVA-attribure: top1:89.7% top5:91.8% top10:93.9% top20:94.0% mAP:90.3%
1062
+ EVA-attribure: -----------------------------------------------------------
1063
+ EVA-attribure: Using 0m 0s
1064
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1065
+ EVA-attribure: Results ---------------------------------------------------
1066
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1067
+ EVA-attribure: -----------------------------------------------------------
1068
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1069
+ EVA-attribure: Results ---------------------------------------------------
1070
+ EVA-attribure: top1:88.8% top5:91.4% top10:93.8% top20:94.0% mAP:89.5%
1071
+ EVA-attribure: -----------------------------------------------------------
1072
+ EVA-attribure.train: ==> Teacher Training ....
1073
+ EVA-attribure.train: ==> Student Training ....
1074
+ EVA-attribure.train: ==> Teacher Training ....
1075
+ EVA-attribure.train: ==> Student Training ....
1076
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1077
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1078
+ EVA-attribure: Extracting features complete in 12m 26s
1079
+ EVA-attribure: Distance computing in 0m 0s
1080
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21074429 -0.23137325 -0.21573906 -0.21263233 -0.22727495 -0.21380158
1081
+ -0.21490979 -0.23053336 -0.22033644 -0.2118777 ] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1082
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1083
+ 0.0001], device='cuda:0')
1084
+ EVA-attribure: Computing CMC and mAP
1085
+ EVA-attribure: Results ---------------------------------------------------
1086
+ EVA-attribure: top1:90.9% top5:92.1% top10:93.8% top20:93.9% mAP:91.5%
1087
+ EVA-attribure: -----------------------------------------------------------
1088
+ EVA-attribure: Using 0m 0s
1089
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1090
+ EVA-attribure: Results ---------------------------------------------------
1091
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1092
+ EVA-attribure: -----------------------------------------------------------
1093
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1094
+ EVA-attribure: Results ---------------------------------------------------
1095
+ EVA-attribure: top1:90.2% top5:91.6% top10:93.6% top20:93.9% mAP:90.8%
1096
+ EVA-attribure: -----------------------------------------------------------
1097
+ EVA-attribure.train: ==> Teacher Training ....
1098
+ EVA-attribure.train: ==> Student Training ....
1099
+ EVA-attribure.train: ==> Teacher Training ....
1100
+ EVA-attribure.train: ==> Student Training ....
1101
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1102
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1103
+ EVA-attribure: Extracting features complete in 12m 31s
1104
+ EVA-attribure: Distance computing in 0m 0s
1105
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21315248 -0.23214054 -0.21794829 -0.21532787 -0.2284778 -0.21513326
1106
+ -0.21718617 -0.23122525 -0.2228269 -0.21332757] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1107
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1108
+ 0.0001], device='cuda:0')
1109
+ EVA-attribure: Computing CMC and mAP
1110
+ EVA-attribure: Results ---------------------------------------------------
1111
+ EVA-attribure: top1:91.0% top5:92.1% top10:93.8% top20:93.9% mAP:91.2%
1112
+ EVA-attribure: -----------------------------------------------------------
1113
+ EVA-attribure: Using 0m 0s
1114
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1115
+ EVA-attribure: Results ---------------------------------------------------
1116
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1117
+ EVA-attribure: -----------------------------------------------------------
1118
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1119
+ EVA-attribure: Results ---------------------------------------------------
1120
+ EVA-attribure: top1:90.2% top5:91.6% top10:93.6% top20:93.9% mAP:90.4%
1121
+ EVA-attribure: -----------------------------------------------------------
1122
+ EVA-attribure.train: ==> Teacher Training ....
1123
+ EVA-attribure.train: ==> Student Training ....
1124
+ EVA-attribure.train: ==> Teacher Training ....
1125
+ EVA-attribure.train: ==> Student Training ....
1126
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1127
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1128
+ EVA-attribure: Extracting features complete in 12m 28s
1129
+ EVA-attribure: Distance computing in 0m 0s
1130
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2115355 -0.2308264 -0.2163416 -0.21375622 -0.22679903 -0.21408999
1131
+ -0.21559504 -0.2303026 -0.22116365 -0.2126784 ] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1132
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1133
+ 0.0001], device='cuda:0')
1134
+ EVA-attribure: Computing CMC and mAP
1135
+ EVA-attribure: Results ---------------------------------------------------
1136
+ EVA-attribure: top1:91.1% top5:92.1% top10:93.8% top20:93.9% mAP:91.5%
1137
+ EVA-attribure: -----------------------------------------------------------
1138
+ EVA-attribure: Using 0m 0s
1139
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1140
+ EVA-attribure: Results ---------------------------------------------------
1141
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1142
+ EVA-attribure: -----------------------------------------------------------
1143
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1144
+ EVA-attribure: Results ---------------------------------------------------
1145
+ EVA-attribure: top1:90.4% top5:91.6% top10:93.6% top20:93.9% mAP:90.7%
1146
+ EVA-attribure: -----------------------------------------------------------
1147
+ EVA-attribure.train: ==> Best Rank-1 90.4%, Best Map 90.9% achieved at epoch 74
1148
+ EVA-attribure.train: Saving the model Now ....., 0.9041, 0.9069
1149
+ EVA-attribure.train: ==> Teacher Training ....
1150
+ EVA-attribure.train: ==> Student Training ....
1151
+ EVA-attribure.train: ==> Teacher Training ....
1152
+ EVA-attribure.train: ==> Student Training ....
1153
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1154
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1155
+ EVA-attribure: Extracting features complete in 12m 28s
1156
+ EVA-attribure: Distance computing in 0m 0s
1157
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2147566 -0.23463756 -0.22020487 -0.21780506 -0.23037697 -0.21784124
1158
+ -0.21935055 -0.23452243 -0.22496746 -0.21604049] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1159
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1160
+ 0.0001], device='cuda:0')
1161
+ EVA-attribure: Computing CMC and mAP
1162
+ EVA-attribure: Results ---------------------------------------------------
1163
+ EVA-attribure: top1:91.1% top5:92.1% top10:93.8% top20:93.9% mAP:91.8%
1164
+ EVA-attribure: -----------------------------------------------------------
1165
+ EVA-attribure: Using 0m 0s
1166
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1167
+ EVA-attribure: Results ---------------------------------------------------
1168
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1169
+ EVA-attribure: -----------------------------------------------------------
1170
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1171
+ EVA-attribure: Results ---------------------------------------------------
1172
+ EVA-attribure: top1:90.4% top5:91.7% top10:93.6% top20:93.9% mAP:91.0%
1173
+ EVA-attribure: -----------------------------------------------------------
1174
+ EVA-attribure.train: ==> Teacher Training ....
1175
+ EVA-attribure.train: ==> Student Training ....
1176
+ EVA-attribure.train: ==> Teacher Training ....
1177
+ EVA-attribure.train: ==> Student Training ....
1178
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1179
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1180
+ EVA-attribure: Extracting features complete in 12m 25s
1181
+ EVA-attribure: Distance computing in 0m 0s
1182
+ EVA-attribure: Mean Distance : (834, 1074), [-0.2141483 -0.23439108 -0.21916826 -0.21638437 -0.22984637 -0.21719843
1183
+ -0.21840937 -0.23319286 -0.22443655 -0.21379735] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1184
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1185
+ 0.0001], device='cuda:0')
1186
+ EVA-attribure: Computing CMC and mAP
1187
+ EVA-attribure: Results ---------------------------------------------------
1188
+ EVA-attribure: top1:90.8% top5:92.1% top10:93.8% top20:94.0% mAP:91.5%
1189
+ EVA-attribure: -----------------------------------------------------------
1190
+ EVA-attribure: Using 0m 0s
1191
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1192
+ EVA-attribure: Results ---------------------------------------------------
1193
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1194
+ EVA-attribure: -----------------------------------------------------------
1195
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1196
+ EVA-attribure: Results ---------------------------------------------------
1197
+ EVA-attribure: top1:90.0% top5:91.6% top10:93.6% top20:94.0% mAP:90.7%
1198
+ EVA-attribure: -----------------------------------------------------------
1199
+ EVA-attribure.train: ==> Teacher Training ....
1200
+ EVA-attribure.train: ==> Student Training ....
1201
+ EVA-attribure.train: ==> Teacher Training ....
1202
+ EVA-attribure.train: ==> Student Training ....
1203
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1204
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1205
+ EVA-attribure: Extracting features complete in 12m 27s
1206
+ EVA-attribure: Distance computing in 0m 0s
1207
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21101823 -0.23224232 -0.21760605 -0.2132943 -0.2275472 -0.21568808
1208
+ -0.21514174 -0.23130286 -0.22212155 -0.21181212] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1209
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1210
+ 0.0001], device='cuda:0')
1211
+ EVA-attribure: Computing CMC and mAP
1212
+ EVA-attribure: Results ---------------------------------------------------
1213
+ EVA-attribure: top1:90.9% top5:92.1% top10:93.8% top20:94.0% mAP:91.5%
1214
+ EVA-attribure: -----------------------------------------------------------
1215
+ EVA-attribure: Using 0m 0s
1216
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1217
+ EVA-attribure: Results ---------------------------------------------------
1218
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1219
+ EVA-attribure: -----------------------------------------------------------
1220
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1221
+ EVA-attribure: Results ---------------------------------------------------
1222
+ EVA-attribure: top1:90.2% top5:91.7% top10:93.6% top20:94.0% mAP:90.8%
1223
+ EVA-attribure: -----------------------------------------------------------
1224
+ EVA-attribure.train: ==> Teacher Training ....
1225
+ EVA-attribure.train: ==> Student Training ....
1226
+ EVA-attribure.train: ==> Teacher Training ....
1227
+ EVA-attribure.train: ==> Student Training ....
1228
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1229
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1230
+ EVA-attribure: Extracting features complete in 12m 27s
1231
+ EVA-attribure: Distance computing in 0m 0s
1232
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21280545 -0.2331697 -0.21911798 -0.21471713 -0.22867844 -0.21731086
1233
+ -0.21594621 -0.23236732 -0.22388323 -0.21247986] , Mean Query : tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1234
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1235
+ 0.0001], device='cuda:0')
1236
+ EVA-attribure: Computing CMC and mAP
1237
+ EVA-attribure: Results ---------------------------------------------------
1238
+ EVA-attribure: top1:90.9% top5:92.1% top10:93.8% top20:94.0% mAP:91.8%
1239
+ EVA-attribure: -----------------------------------------------------------
1240
+ EVA-attribure: Using 0m 0s
1241
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1242
+ EVA-attribure: Results ---------------------------------------------------
1243
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1244
+ EVA-attribure: -----------------------------------------------------------
1245
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1246
+ EVA-attribure: Results ---------------------------------------------------
1247
+ EVA-attribure: top1:90.2% top5:91.6% top10:93.6% top20:94.0% mAP:91.0%
1248
+ EVA-attribure: -----------------------------------------------------------
1249
+ EVA-attribure.train: ==> Teacher Training ....
1250
+ EVA-attribure.train: ==> Student Training ....
1251
+ EVA-attribure.train: ==> Teacher Training ....
1252
+ EVA-attribure.train: ==> Student Training ....
1253
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1254
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1255
+ EVA-attribure: Extracting features complete in 12m 27s
1256
+ EVA-attribure: Distance computing in 0m 0s
1257
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20765163 -0.23189156 -0.21707892 -0.21176305 -0.22738832 -0.21494499
1258
+ -0.21272634 -0.23173973 -0.2212931 -0.21157524] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1259
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1260
+ 0.0001], device='cuda:0')
1261
+ EVA-attribure: Computing CMC and mAP
1262
+ EVA-attribure: Results ---------------------------------------------------
1263
+ EVA-attribure: top1:91.1% top5:92.1% top10:93.8% top20:94.0% mAP:91.8%
1264
+ EVA-attribure: -----------------------------------------------------------
1265
+ EVA-attribure: Using 0m 0s
1266
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1267
+ EVA-attribure: Results ---------------------------------------------------
1268
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1269
+ EVA-attribure: -----------------------------------------------------------
1270
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1271
+ EVA-attribure: Results ---------------------------------------------------
1272
+ EVA-attribure: top1:90.6% top5:91.6% top10:93.6% top20:94.0% mAP:91.1%
1273
+ EVA-attribure: -----------------------------------------------------------
1274
+ EVA-attribure.train: ==> Best Rank-1 90.6%, Best Map 91.1% achieved at epoch 84
1275
+ EVA-attribure.train: Saving the model Now ....., 0.9065, 0.9109
1276
+ EVA-attribure.train: ==> Teacher Training ....
1277
+ EVA-attribure.train: ==> Student Training ....
1278
+ EVA-attribure.train: ==> Teacher Training ....
1279
+ EVA-attribure.train: ==> Student Training ....
1280
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1281
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1282
+ EVA-attribure: Extracting features complete in 12m 26s
1283
+ EVA-attribure: Distance computing in 0m 0s
1284
+ EVA-attribure: Mean Distance : (834, 1074), [-0.20871276 -0.23122072 -0.21646637 -0.21211465 -0.22760239 -0.21453595
1285
+ -0.21322015 -0.23008838 -0.22122735 -0.2104189 ] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1286
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1287
+ 0.0001], device='cuda:0')
1288
+ EVA-attribure: Computing CMC and mAP
1289
+ EVA-attribure: Results ---------------------------------------------------
1290
+ EVA-attribure: top1:91.1% top5:92.1% top10:93.8% top20:94.0% mAP:91.6%
1291
+ EVA-attribure: -----------------------------------------------------------
1292
+ EVA-attribure: Using 0m 0s
1293
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1294
+ EVA-attribure: Results ---------------------------------------------------
1295
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1296
+ EVA-attribure: -----------------------------------------------------------
1297
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1298
+ EVA-attribure: Results ---------------------------------------------------
1299
+ EVA-attribure: top1:90.5% top5:91.6% top10:93.6% top20:94.0% mAP:90.9%
1300
+ EVA-attribure: -----------------------------------------------------------
1301
+ EVA-attribure.train: ==> Teacher Training ....
1302
+ EVA-attribure.train: ==> Student Training ....
1303
+ EVA-attribure.train: ==> Teacher Training ....
1304
+ EVA-attribure.train: ==> Student Training ....
1305
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1306
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1307
+ EVA-attribure: Extracting features complete in 12m 33s
1308
+ EVA-attribure: Distance computing in 0m 0s
1309
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21206005 -0.2340149 -0.21801369 -0.21654427 -0.22922502 -0.2152433
1310
+ -0.21575643 -0.23191503 -0.22151464 -0.2141129 ] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1311
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1312
+ 0.0001], device='cuda:0')
1313
+ EVA-attribure: Computing CMC and mAP
1314
+ EVA-attribure: Results ---------------------------------------------------
1315
+ EVA-attribure: top1:91.4% top5:92.2% top10:93.8% top20:94.0% mAP:91.3%
1316
+ EVA-attribure: -----------------------------------------------------------
1317
+ EVA-attribure: Using 0m 0s
1318
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1319
+ EVA-attribure: Results ---------------------------------------------------
1320
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1321
+ EVA-attribure: -----------------------------------------------------------
1322
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1323
+ EVA-attribure: Results ---------------------------------------------------
1324
+ EVA-attribure: top1:90.8% top5:91.8% top10:93.6% top20:94.0% mAP:90.6%
1325
+ EVA-attribure: -----------------------------------------------------------
1326
+ EVA-attribure.train: ==> Best Rank-1 90.8%, Best Map 91.1% achieved at epoch 88
1327
+ EVA-attribure.train: Saving the model Now ....., 0.9077, 0.9058
1328
+ EVA-attribure.train: ==> Teacher Training ....
1329
+ EVA-attribure.train: ==> Student Training ....
1330
+ EVA-attribure.train: ==> Teacher Training ....
1331
+ EVA-attribure.train: ==> Student Training ....
1332
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1333
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1334
+ EVA-attribure: Extracting features complete in 12m 16s
1335
+ EVA-attribure: Distance computing in 0m 0s
1336
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21463989 -0.23406848 -0.21852815 -0.21809904 -0.22983834 -0.21456386
1337
+ -0.21651968 -0.23113817 -0.2204485 -0.21294434] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1338
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1339
+ 0.0001], device='cuda:0')
1340
+ EVA-attribure: Computing CMC and mAP
1341
+ EVA-attribure: Results ---------------------------------------------------
1342
+ EVA-attribure: top1:91.8% top5:92.2% top10:93.6% top20:94.0% mAP:91.6%
1343
+ EVA-attribure: -----------------------------------------------------------
1344
+ EVA-attribure: Using 0m 0s
1345
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1346
+ EVA-attribure: Results ---------------------------------------------------
1347
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1348
+ EVA-attribure: -----------------------------------------------------------
1349
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1350
+ EVA-attribure: Results ---------------------------------------------------
1351
+ EVA-attribure: top1:91.2% top5:91.8% top10:93.5% top20:94.0% mAP:90.8%
1352
+ EVA-attribure: -----------------------------------------------------------
1353
+ EVA-attribure.train: ==> Best Rank-1 91.2%, Best Map 91.1% achieved at epoch 90
1354
+ EVA-attribure.train: Saving the model Now ....., 0.9125, 0.9082
1355
+ EVA-attribure.train: ==> Teacher Training ....
1356
+ EVA-attribure.train: ==> Student Training ....
1357
+ EVA-attribure.train: ==> Teacher Training ....
1358
+ EVA-attribure.train: ==> Student Training ....
1359
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1360
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1361
+ EVA-attribure: Extracting features complete in 12m 27s
1362
+ EVA-attribure: Distance computing in 0m 0s
1363
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21363999 -0.23516048 -0.21698713 -0.21662797 -0.22949347 -0.21558842
1364
+ -0.21720865 -0.23076694 -0.22225036 -0.2149075 ] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1365
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1366
+ 0.0001], device='cuda:0')
1367
+ EVA-attribure: Computing CMC and mAP
1368
+ EVA-attribure: Results ---------------------------------------------------
1369
+ EVA-attribure: top1:91.2% top5:92.6% top10:93.6% top20:94.0% mAP:91.0%
1370
+ EVA-attribure: -----------------------------------------------------------
1371
+ EVA-attribure: Using 0m 0s
1372
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1373
+ EVA-attribure: Results ---------------------------------------------------
1374
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1375
+ EVA-attribure: -----------------------------------------------------------
1376
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1377
+ EVA-attribure: Results ---------------------------------------------------
1378
+ EVA-attribure: top1:90.9% top5:92.2% top10:93.5% top20:94.0% mAP:90.3%
1379
+ EVA-attribure: -----------------------------------------------------------
1380
+ EVA-attribure.train: ==> Teacher Training ....
1381
+ EVA-attribure.train: ==> Student Training ....
1382
+ EVA-attribure.train: ==> Teacher Training ....
1383
+ EVA-attribure.train: ==> Student Training ....
1384
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1385
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1386
+ EVA-attribure: Extracting features complete in 12m 27s
1387
+ EVA-attribure: Distance computing in 0m 0s
1388
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21092741 -0.23491454 -0.21771935 -0.21668677 -0.22978817 -0.21541935
1389
+ -0.2156132 -0.23216116 -0.22115982 -0.21116486] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1390
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1391
+ 0.0001], device='cuda:0')
1392
+ EVA-attribure: Computing CMC and mAP
1393
+ EVA-attribure: Results ---------------------------------------------------
1394
+ EVA-attribure: top1:91.2% top5:92.3% top10:93.6% top20:94.0% mAP:91.1%
1395
+ EVA-attribure: -----------------------------------------------------------
1396
+ EVA-attribure: Using 0m 0s
1397
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1398
+ EVA-attribure: Results ---------------------------------------------------
1399
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1400
+ EVA-attribure: -----------------------------------------------------------
1401
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1402
+ EVA-attribure: Results ---------------------------------------------------
1403
+ EVA-attribure: top1:90.8% top5:92.0% top10:93.5% top20:94.0% mAP:90.4%
1404
+ EVA-attribure: -----------------------------------------------------------
1405
+ EVA-attribure.train: ==> Teacher Training ....
1406
+ EVA-attribure.train: ==> Student Training ....
1407
+ EVA-attribure.train: ==> Teacher Training ....
1408
+ EVA-attribure.train: ==> Student Training ....
1409
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1410
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1411
+ EVA-attribure: Extracting features complete in 12m 26s
1412
+ EVA-attribure: Distance computing in 0m 0s
1413
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21297581 -0.2365449 -0.21866372 -0.2155347 -0.22957788 -0.21461928
1414
+ -0.21455525 -0.23241125 -0.22258326 -0.21336825] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1415
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1416
+ 0.0001], device='cuda:0')
1417
+ EVA-attribure: Computing CMC and mAP
1418
+ EVA-attribure: Results ---------------------------------------------------
1419
+ EVA-attribure: top1:91.4% top5:92.2% top10:93.5% top20:94.0% mAP:91.2%
1420
+ EVA-attribure: -----------------------------------------------------------
1421
+ EVA-attribure: Using 0m 0s
1422
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1423
+ EVA-attribure: Results ---------------------------------------------------
1424
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1425
+ EVA-attribure: -----------------------------------------------------------
1426
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1427
+ EVA-attribure: Results ---------------------------------------------------
1428
+ EVA-attribure: top1:90.8% top5:91.8% top10:93.4% top20:94.0% mAP:90.5%
1429
+ EVA-attribure: -----------------------------------------------------------
1430
+ EVA-attribure.train: ==> Teacher Training ....
1431
+ EVA-attribure.train: ==> Student Training ....
1432
+ EVA-attribure.train: ==> Teacher Training ....
1433
+ EVA-attribure.train: ==> Student Training ....
1434
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1435
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1436
+ EVA-attribure: Extracting features complete in 12m 26s
1437
+ EVA-attribure: Distance computing in 0m 0s
1438
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21381028 -0.23569532 -0.21910676 -0.21504684 -0.2313859 -0.2178408
1439
+ -0.21613993 -0.2334768 -0.22398372 -0.21140188] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1440
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1441
+ 0.0001], device='cuda:0')
1442
+ EVA-attribure: Computing CMC and mAP
1443
+ EVA-attribure: Results ---------------------------------------------------
1444
+ EVA-attribure: top1:90.3% top5:92.1% top10:93.8% top20:93.9% mAP:90.8%
1445
+ EVA-attribure: -----------------------------------------------------------
1446
+ EVA-attribure: Using 0m 0s
1447
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1448
+ EVA-attribure: Results ---------------------------------------------------
1449
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1450
+ EVA-attribure: -----------------------------------------------------------
1451
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1452
+ EVA-attribure: Results ---------------------------------------------------
1453
+ EVA-attribure: top1:89.7% top5:91.6% top10:93.6% top20:93.9% mAP:90.0%
1454
+ EVA-attribure: -----------------------------------------------------------
1455
+ EVA-attribure.train: ==> Teacher Training ....
1456
+ EVA-attribure.train: ==> Student Training ....
1457
+ EVA-attribure.train: ==> Teacher Training ....
1458
+ EVA-attribure.train: ==> Student Training ....
1459
+ EVA-attribure: Extracted features for query set, obtained torch.Size([834, 1024]) matrix
1460
+ EVA-attribure: Extracted features for gallery set, obtained torch.Size([1074, 1024]) matrix
1461
+ EVA-attribure: Extracting features complete in 12m 26s
1462
+ EVA-attribure: Distance computing in 0m 0s
1463
+ EVA-attribure: Mean Distance : (834, 1074), [-0.21488309 -0.23592728 -0.219786 -0.22046727 -0.2320272 -0.21765688
1464
+ -0.21799433 -0.23396699 -0.22374003 -0.21535294] , Mean Query : tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1465
+ 0.0001], device='cuda:0') Mean Gallery: tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002,
1466
+ 0.0001], device='cuda:0')
1467
+ EVA-attribure: Computing CMC and mAP
1468
+ EVA-attribure: Results ---------------------------------------------------
1469
+ EVA-attribure: top1:90.6% top5:92.2% top10:93.8% top20:94.0% mAP:90.5%
1470
+ EVA-attribure: -----------------------------------------------------------
1471
+ EVA-attribure: Using 0m 0s
1472
+ EVA-attribure: Computing CMC and mAP only for the same clothes setting
1473
+ EVA-attribure: Results ---------------------------------------------------
1474
+ EVA-attribure: top1:100.0% top5:100.0% top10:100.0% top20:100.0% mAP:100.0%
1475
+ EVA-attribure: -----------------------------------------------------------
1476
+ EVA-attribure: Computing CMC and mAP only for clothes-changing
1477
+ EVA-attribure: Results ---------------------------------------------------
1478
+ EVA-attribure: top1:90.0% top5:91.8% top10:93.6% top20:94.0% mAP:89.8%
1479
+ EVA-attribure: -----------------------------------------------------------
1480
+ EVA-attribure.train: Training time 13:00:18
1481
+ EVA-attribure.train: ==> Best Rank-1 91.2%, Best Map 91.1% achieved at epoch 90