OpenIBL
OpenIBL copied to clipboard
yxgeee
reproduction problem
I just run the code without any change and find the initial recall scores is as follow: Recall Scores: top-1 1.2% top-5 4.6% top-10 8.7% then I continue the training process util the generation1 epoch2 with the recall score: Recall Scores: top-1 1.4% top-5 6.3% top-10 11.9%
- Finished generation 1 epoch 2 recall@1: 1.4% recall@5: 6.3% recall@10: 11.9% best@5: 7.7% it seems no obvious promotion, is that right?
It seems totally failed. I did not meet this problem before. Did you load all pre-trained weights as instructed?
It seems totally failed. I did not meet this problem before. Did you load all pre-trained weights as instructed?
yes, i load the pre-trained weights and besides i test the best model on pitts 30k that you provide (https://drive.google.com/drive/folders/1FLjxFhKRO-YJQ6FI-DcCMMHDL2K_Hsof) , the score is : Recall Scores: top-1 1.3% top-5 5.0% top-10 9.7% is there something wrong?
I cannot locate your issue. Maybe provide your full testing log could help.
this is the log of first evaluation of my initial model:
Args:Namespace(arch='vgg16', cache_size=1000, data_dir='/data/yangling/datasets', dataset='pitts', deterministic=False, epochs=5, eval_step=1, features=4096, generations=4, gpu=0, height=480, init_dir='/data/yangling/openIBL/examples/../logs', iters=0, launcher='pytorch', layers='conv5', logs_dir='logs/netVLAD/pitts30k-vgg16/conv5-sare_ind-lr0.001-tuple1-SFRS', loss_type='sare_ind', lr=0.001, margin=0.1, momentum=0.9, neg_num=10, neg_pool=1000, ngpus_per_node=1, nowhiten=False, num_clusters=64, pos_num=10, pos_pool=20, print_freq=10, rank=0, resume='', scale='30k', seed=43, soft_weight=0.5, step_size=5, sync_gather=False, syncbn=True, tcp_port='19294', temperature=[0.07, 0.07, 0.06, 0.05], test_batch_size=8, tuple_size=1, weight_decay=0.001, width=640, workers=2, world_size=1)
Pittsburgh dataset loaded subset | # pids | # images
train_query | 311 | 7320 train_gallery | 417 | 10000 val_query | 319 | 7608 val_gallery | 417 | 10000 test_query | 286 | 6816 test_gallery | 417 | 10000 Loading centroids from /data/yangling/openIBL/examples/../logs/vgg16_pitts_64_desc_cen.hdf5 Loading centroids from /data/yangling/openIBL/examples/../logs/vgg16_pitts_64_desc_cen.hdf5 Test the initial model: Extract Features: [10/2201] Time 0.124 (0.506) Data 0.000 (0.282) Extract Features: [20/2201] Time 0.123 (0.315) Data 0.000 (0.141) Extract Features: [30/2201] Time 0.124 (0.251) Data 0.000 (0.094) Extract Features: [40/2201] Time 0.123 (0.219) Data 0.000 (0.071) Extract Features: [50/2201] Time 0.125 (0.200) Data 0.000 (0.057) Extract Features: [60/2201] Time 0.124 (0.188) Data 0.000 (0.047) Extract Features: [70/2201] Time 0.124 (0.179) Data 0.000 (0.040) Extract Features: [80/2201] Time 0.125 (0.172) Data 0.000 (0.035) Extract Features: [90/2201] Time 0.124 (0.167) Data 0.000 (0.031) Extract Features: [100/2201] Time 0.125 (0.163) Data 0.000 (0.028) Extract Features: [110/2201] Time 0.125 (0.159) Data 0.000 (0.026) Extract Features: [120/2201] Time 0.125 (0.156) Data 0.000 (0.024) Extract Features: [130/2201] Time 0.126 (0.154) Data 0.000 (0.022) Extract Features: [140/2201] Time 0.125 (0.152) Data 0.000 (0.020) Extract Features: [150/2201] Time 0.125 (0.150) Data 0.000 (0.019) Extract Features: [160/2201] Time 0.125 (0.149) Data 0.000 (0.018) Extract Features: [170/2201] Time 0.126 (0.147) Data 0.000 (0.017) Extract Features: [180/2201] Time 0.126 (0.146) Data 0.000 (0.016) Extract Features: [190/2201] Time 0.125 (0.145) Data 0.000 (0.015) Extract Features: [200/2201] Time 0.125 (0.144) Data 0.000 (0.014) Extract Features: [210/2201] Time 0.125 (0.144) Data 0.000 (0.014) Extract Features: [220/2201] Time 0.126 (0.143) Data 0.000 (0.013) Extract Features: [230/2201] Time 0.126 (0.142) Data 0.000 (0.012) Extract Features: [240/2201] Time 0.129 (0.141) Data 0.000 (0.012) Extract Features: [250/2201] Time 0.126 (0.141) Data 0.000 (0.011) Extract Features: [260/2201] Time 0.129 (0.140) Data 0.000 (0.011) Extract Features: [270/2201] Time 0.126 (0.140) Data 0.000 (0.011) Extract Features: [280/2201] Time 0.127 (0.139) Data 0.000 (0.010) Extract Features: [290/2201] Time 0.128 (0.139) Data 0.000 (0.010) Extract Features: [300/2201] Time 0.132 (0.139) Data 0.000 (0.010) Extract Features: [310/2201] Time 0.127 (0.138) Data 0.000 (0.009) Extract Features: [320/2201] Time 0.128 (0.138) Data 0.000 (0.009) Extract Features: [330/2201] Time 0.129 (0.138) Data 0.000 (0.009) Extract Features: [340/2201] Time 0.127 (0.137) Data 0.000 (0.008) Extract Features: [350/2201] Time 0.127 (0.137) Data 0.000 (0.008) Extract Features: [360/2201] Time 0.127 (0.137) Data 0.000 (0.008) Extract Features: [370/2201] Time 0.134 (0.137) Data 0.000 (0.008) Extract Features: [380/2201] Time 0.127 (0.136) Data 0.000 (0.008) Extract Features: [390/2201] Time 0.127 (0.136) Data 0.000 (0.007) Extract Features: [400/2201] Time 0.130 (0.136) Data 0.000 (0.007) Extract Features: [410/2201] Time 0.130 (0.136) Data 0.000 (0.007) Extract Features: [420/2201] Time 0.131 (0.136) Data 0.000 (0.007) Extract Features: [430/2201] Time 0.128 (0.135) Data 0.000 (0.007) Extract Features: [440/2201] Time 0.128 (0.135) Data 0.000 (0.007) Extract Features: [450/2201] Time 0.128 (0.135) Data 0.000 (0.006) Extract Features: [460/2201] Time 0.128 (0.135) Data 0.000 (0.006) Extract Features: [470/2201] Time 0.127 (0.135) Data 0.000 (0.006) Extract Features: [480/2201] Time 0.128 (0.135) Data 0.000 (0.006) Extract Features: [490/2201] Time 0.128 (0.135) Data 0.000 (0.006) Extract Features: [500/2201] Time 0.128 (0.135) Data 0.000 (0.006) Extract Features: [510/2201] Time 0.129 (0.134) Data 0.000 (0.006) Extract Features: [520/2201] Time 0.129 (0.134) Data 0.000 (0.006) Extract Features: [530/2201] Time 0.129 (0.134) Data 0.000 (0.005) Extract Features: [540/2201] Time 0.129 (0.134) Data 0.000 (0.005) Extract Features: [550/2201] Time 0.129 (0.134) Data 0.000 (0.005) Extract Features: [560/2201] Time 0.131 (0.134) Data 0.000 (0.005) Extract Features: [570/2201] Time 0.132 (0.134) Data 0.000 (0.005) Extract Features: [580/2201] Time 0.137 (0.134) Data 0.000 (0.005) Extract Features: [590/2201] Time 0.129 (0.134) Data 0.000 (0.005) Extract Features: [600/2201] Time 0.131 (0.134) Data 0.000 (0.005) Extract Features: [610/2201] Time 0.129 (0.134) Data 0.000 (0.005) Extract Features: [620/2201] Time 0.130 (0.134) Data 0.000 (0.005) Extract Features: [630/2201] Time 0.131 (0.133) Data 0.000 (0.005) Extract Features: [640/2201] Time 0.129 (0.133) Data 0.000 (0.005) Extract Features: [650/2201] Time 0.130 (0.133) Data 0.000 (0.004) Extract Features: [660/2201] Time 0.130 (0.133) Data 0.000 (0.004) Extract Features: [670/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [680/2201] Time 0.133 (0.133) Data 0.000 (0.004) Extract Features: [690/2201] Time 0.142 (0.133) Data 0.000 (0.004) Extract Features: [700/2201] Time 0.128 (0.133) Data 0.000 (0.004) Extract Features: [710/2201] Time 0.128 (0.133) Data 0.000 (0.004) Extract Features: [720/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [730/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [740/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [750/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [760/2201] Time 0.129 (0.133) Data 0.000 (0.004) Extract Features: [770/2201] Time 0.132 (0.133) Data 0.000 (0.004) Extract Features: [780/2201] Time 0.130 (0.133) Data 0.000 (0.004) Extract Features: [790/2201] Time 0.139 (0.133) Data 0.000 (0.004) Extract Features: [800/2201] Time 0.130 (0.133) Data 0.000 (0.004) Extract Features: [810/2201] Time 0.130 (0.133) Data 0.000 (0.004) Extract Features: [820/2201] Time 0.128 (0.133) Data 0.000 (0.004) Extract Features: [830/2201] Time 0.142 (0.133) Data 0.000 (0.004) Extract Features: [840/2201] Time 0.131 (0.133) Data 0.000 (0.003) Extract Features: [850/2201] Time 0.129 (0.133) Data 0.000 (0.003) Extract Features: [860/2201] Time 0.134 (0.133) Data 0.000 (0.003) Extract Features: [870/2201] Time 0.130 (0.133) Data 0.000 (0.003) Extract Features: [880/2201] Time 0.131 (0.133) Data 0.000 (0.003) Extract Features: [890/2201] Time 0.130 (0.133) Data 0.000 (0.003) Extract Features: [900/2201] Time 0.130 (0.133) Data 0.000 (0.003) Extract Features: [910/2201] Time 0.131 (0.133) Data 0.000 (0.003) Extract Features: [920/2201] Time 0.129 (0.133) Data 0.000 (0.003) Extract Features: [930/2201] Time 0.130 (0.133) Data 0.000 (0.003) Extract Features: [940/2201] Time 0.129 (0.133) Data 0.000 (0.003) Extract Features: [950/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [960/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [970/2201] Time 0.133 (0.132) Data 0.000 (0.003) Extract Features: [980/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [990/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1000/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1010/2201] Time 0.129 (0.132) Data 0.000 (0.003) Extract Features: [1020/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1030/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1040/2201] Time 0.134 (0.132) Data 0.000 (0.003) Extract Features: [1050/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1060/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1070/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1080/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1090/2201] Time 0.138 (0.132) Data 0.000 (0.003) Extract Features: [1100/2201] Time 0.137 (0.132) Data 0.000 (0.003) Extract Features: [1110/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1120/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1130/2201] Time 0.130 (0.132) Data 0.000 (0.003) Extract Features: [1140/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1150/2201] Time 0.132 (0.132) Data 0.000 (0.003) Extract Features: [1160/2201] Time 0.131 (0.132) Data 0.000 (0.003) Extract Features: [1170/2201] Time 0.132 (0.132) Data 0.000 (0.003) Extract Features: [1180/2201] Time 0.132 (0.132) Data 0.000 (0.003) Extract Features: [1190/2201] Time 0.148 (0.132) Data 0.000 (0.003) Extract Features: [1200/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1210/2201] Time 0.129 (0.132) Data 0.000 (0.002) Extract Features: [1220/2201] Time 0.130 (0.132) Data 0.000 (0.002) Extract Features: [1230/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1240/2201] Time 0.139 (0.132) Data 0.000 (0.002) Extract Features: [1250/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1260/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1270/2201] Time 0.130 (0.132) Data 0.000 (0.002) Extract Features: [1280/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1290/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1300/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1310/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1320/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1330/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1340/2201] Time 0.134 (0.132) Data 0.000 (0.002) Extract Features: [1350/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1360/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1370/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1380/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1390/2201] Time 0.137 (0.132) Data 0.000 (0.002) Extract Features: [1400/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1410/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1420/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1430/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1440/2201] Time 0.136 (0.132) Data 0.000 (0.002) Extract Features: [1450/2201] Time 0.130 (0.132) Data 0.000 (0.002) Extract Features: [1460/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1470/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1480/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1490/2201] Time 0.130 (0.132) Data 0.000 (0.002) Extract Features: [1500/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1510/2201] Time 0.134 (0.132) Data 0.000 (0.002) Extract Features: [1520/2201] Time 0.130 (0.132) Data 0.000 (0.002) Extract Features: [1530/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1540/2201] Time 0.137 (0.132) Data 0.000 (0.002) Extract Features: [1550/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1560/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1570/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1580/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1590/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1600/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1610/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1620/2201] Time 0.136 (0.132) Data 0.000 (0.002) Extract Features: [1630/2201] Time 0.143 (0.132) Data 0.000 (0.002) Extract Features: [1640/2201] Time 0.134 (0.132) Data 0.000 (0.002) Extract Features: [1650/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1660/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1670/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1680/2201] Time 0.133 (0.132) Data 0.000 (0.002) Extract Features: [1690/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1700/2201] Time 0.136 (0.132) Data 0.000 (0.002) Extract Features: [1710/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1720/2201] Time 0.133 (0.132) Data 0.000 (0.002) Extract Features: [1730/2201] Time 0.134 (0.132) Data 0.000 (0.002) Extract Features: [1740/2201] Time 0.135 (0.132) Data 0.000 (0.002) Extract Features: [1750/2201] Time 0.133 (0.132) Data 0.000 (0.002) Extract Features: [1760/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1770/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1780/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1790/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1800/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1810/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1820/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1830/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1840/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1850/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1860/2201] Time 0.135 (0.132) Data 0.000 (0.002) Extract Features: [1870/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1880/2201] Time 0.133 (0.132) Data 0.000 (0.002) Extract Features: [1890/2201] Time 0.131 (0.132) Data 0.000 (0.002) Extract Features: [1900/2201] Time 0.133 (0.132) Data 0.000 (0.002) Extract Features: [1910/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1920/2201] Time 0.132 (0.132) Data 0.000 (0.002) Extract Features: [1930/2201] Time 0.138 (0.132) Data 0.000 (0.002) Extract Features: [1940/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [1950/2201] Time 0.133 (0.133) Data 0.000 (0.002) Extract Features: [1960/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [1970/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [1980/2201] Time 0.133 (0.133) Data 0.000 (0.002) Extract Features: [1990/2201] Time 0.131 (0.133) Data 0.000 (0.002) Extract Features: [2000/2201] Time 0.133 (0.133) Data 0.000 (0.002) Extract Features: [2010/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [2020/2201] Time 0.136 (0.133) Data 0.000 (0.002) Extract Features: [2030/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [2040/2201] Time 0.133 (0.133) Data 0.000 (0.002) Extract Features: [2050/2201] Time 0.138 (0.133) Data 0.000 (0.002) Extract Features: [2060/2201] Time 0.131 (0.133) Data 0.000 (0.002) Extract Features: [2070/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [2080/2201] Time 0.132 (0.133) Data 0.000 (0.002) Extract Features: [2090/2201] Time 0.133 (0.133) Data 0.000 (0.001) Extract Features: [2100/2201] Time 0.132 (0.133) Data 0.000 (0.001) Extract Features: [2110/2201] Time 0.134 (0.133) Data 0.000 (0.001) Extract Features: [2120/2201] Time 0.133 (0.133) Data 0.000 (0.001) Extract Features: [2130/2201] Time 0.134 (0.133) Data 0.000 (0.001) Extract Features: [2140/2201] Time 0.133 (0.133) Data 0.000 (0.001) Extract Features: [2150/2201] Time 0.132 (0.133) Data 0.000 (0.001) Extract Features: [2160/2201] Time 0.132 (0.133) Data 0.000 (0.001) Extract Features: [2170/2201] Time 0.133 (0.133) Data 0.000 (0.001) Extract Features: [2180/2201] Time 0.136 (0.133) Data 0.000 (0.001) Extract Features: [2190/2201] Time 0.133 (0.133) Data 0.000 (0.001) Extract Features: [2200/2201] Time 0.132 (0.133) Data 0.000 (0.001) gathering features from rank no.0 ===> Start calculating pairwise distances ===> Start calculating recalls Recall Scores: top-1 1.2% top-5 4.6% top-10 8.7% ===> Start extracting features for sorting gallery
I cannot locate your issue. Maybe provide your full testing log could help.
if you have retested the model rightly, could you please zip your local folder and send it to my email ([email protected])? I would appreciate it.
It shows that resume='', indicating that you did not load any model weights.
It shows that resume='', indicating that you did not load any model weights.
this is my training log, so resume=''
It shows that resume='', indicating that you did not load any model weights.
And this is my testing log, the model is downloaded from your provided url:
Use GPU: 0 for testing, rank no.0 of world_size 1
Args:Namespace(arch='vgg16', data_dir='/data/yangling/test/OpenIBL-0.1.0-beta/examples/data', dataset='pitts', features=4096, gpu=0, height=480, lambda_value=0, launcher='pytorch', logs_dir='/data/yangling/test/OpenIBL-0.1.0-beta/examples/logs', ngpus_per_node=1, nowhiten=False, num_clusters=64, print_freq=10, rank=0, reduction=True, rerank=False, resume='logs/ori/model_best.pth.tar', rr_topk=25, scale='30k', sync_gather=False, tcp_port='5017', test_batch_size=6, vlad=True, width=640, workers=2, world_size=1)
Pittsburgh dataset loaded subset | # pids | # images
train_query | 311 | 7320 train_gallery | 417 | 10000 val_query | 319 | 7608 val_gallery | 417 | 10000 test_query | 286 | 6816 test_gallery | 417 | 10000 => Loaded checkpoint 'logs/ori/model_best.pth.tar' => Start epoch 3 best recall5 94.5% Evaluate on the test set: load PCA parameters... Extract Features: [10/1136] Time 0.100 (0.557) Data 0.000 (0.289) Extract Features: [20/1136] Time 0.108 (0.330) Data 0.000 (0.145) Extract Features: [30/1136] Time 0.109 (0.256) Data 0.000 (0.097) Extract Features: [40/1136] Time 0.111 (0.219) Data 0.000 (0.072) Extract Features: [50/1136] Time 0.110 (0.197) Data 0.000 (0.058) Extract Features: [60/1136] Time 0.110 (0.182) Data 0.000 (0.048) Extract Features: [70/1136] Time 0.101 (0.172) Data 0.000 (0.041) Extract Features: [80/1136] Time 0.100 (0.164) Data 0.000 (0.036) Extract Features: [90/1136] Time 0.109 (0.158) Data 0.000 (0.032) Extract Features: [100/1136] Time 0.152 (0.153) Data 0.000 (0.029) Extract Features: [110/1136] Time 0.107 (0.149) Data 0.000 (0.026) Extract Features: [120/1136] Time 0.102 (0.145) Data 0.000 (0.024) Extract Features: [130/1136] Time 0.103 (0.143) Data 0.000 (0.022) Extract Features: [140/1136] Time 0.122 (0.141) Data 0.000 (0.021) Extract Features: [150/1136] Time 0.120 (0.139) Data 0.000 (0.019) Extract Features: [160/1136] Time 0.108 (0.137) Data 0.000 (0.018) Extract Features: [170/1136] Time 0.156 (0.136) Data 0.000 (0.017) Extract Features: [180/1136] Time 0.130 (0.134) Data 0.000 (0.016) Extract Features: [190/1136] Time 0.155 (0.134) Data 0.000 (0.015) Extract Features: [200/1136] Time 0.120 (0.133) Data 0.000 (0.015) Extract Features: [210/1136] Time 0.112 (0.132) Data 0.000 (0.014) Extract Features: [220/1136] Time 0.120 (0.132) Data 0.000 (0.013) Extract Features: [230/1136] Time 0.110 (0.131) Data 0.000 (0.013) Extract Features: [240/1136] Time 0.121 (0.131) Data 0.000 (0.012) Extract Features: [250/1136] Time 0.110 (0.130) Data 0.000 (0.012) Extract Features: [260/1136] Time 0.111 (0.129) Data 0.000 (0.011) Extract Features: [270/1136] Time 0.112 (0.129) Data 0.000 (0.011) Extract Features: [280/1136] Time 0.111 (0.128) Data 0.000 (0.010) Extract Features: [290/1136] Time 0.111 (0.128) Data 0.000 (0.010) Extract Features: [300/1136] Time 0.112 (0.127) Data 0.000 (0.010) Extract Features: [310/1136] Time 0.122 (0.127) Data 0.000 (0.009) Extract Features: [320/1136] Time 0.122 (0.127) Data 0.000 (0.009) Extract Features: [330/1136] Time 0.124 (0.127) Data 0.000 (0.009) Extract Features: [340/1136] Time 0.111 (0.126) Data 0.000 (0.009) Extract Features: [350/1136] Time 0.111 (0.126) Data 0.000 (0.008) Extract Features: [360/1136] Time 0.127 (0.126) Data 0.000 (0.008) Extract Features: [370/1136] Time 0.129 (0.126) Data 0.000 (0.008) Extract Features: [380/1136] Time 0.111 (0.126) Data 0.000 (0.008) Extract Features: [390/1136] Time 0.111 (0.126) Data 0.000 (0.008) Extract Features: [400/1136] Time 0.113 (0.125) Data 0.000 (0.007) Extract Features: [410/1136] Time 0.113 (0.125) Data 0.000 (0.007) Extract Features: [420/1136] Time 0.124 (0.125) Data 0.000 (0.007) Extract Features: [430/1136] Time 0.138 (0.125) Data 0.000 (0.007) Extract Features: [440/1136] Time 0.113 (0.125) Data 0.000 (0.007) Extract Features: [450/1136] Time 0.113 (0.125) Data 0.000 (0.007) Extract Features: [460/1136] Time 0.142 (0.125) Data 0.000 (0.006) Extract Features: [470/1136] Time 0.114 (0.125) Data 0.000 (0.006) Extract Features: [480/1136] Time 0.113 (0.125) Data 0.000 (0.006) Extract Features: [490/1136] Time 0.113 (0.125) Data 0.000 (0.006) Extract Features: [500/1136] Time 0.112 (0.125) Data 0.000 (0.006) Extract Features: [510/1136] Time 0.113 (0.125) Data 0.000 (0.006) Extract Features: [520/1136] Time 0.116 (0.125) Data 0.000 (0.006) Extract Features: [530/1136] Time 0.115 (0.125) Data 0.000 (0.006) Extract Features: [540/1136] Time 0.112 (0.124) Data 0.000 (0.006) Extract Features: [550/1136] Time 0.114 (0.124) Data 0.000 (0.005) Extract Features: [560/1136] Time 0.124 (0.124) Data 0.000 (0.005) Extract Features: [570/1136] Time 0.122 (0.124) Data 0.000 (0.005) Extract Features: [580/1136] Time 0.128 (0.124) Data 0.000 (0.005) Extract Features: [590/1136] Time 0.122 (0.124) Data 0.000 (0.005) Extract Features: [600/1136] Time 0.113 (0.124) Data 0.000 (0.005) Extract Features: [610/1136] Time 0.111 (0.124) Data 0.000 (0.005) Extract Features: [620/1136] Time 0.114 (0.123) Data 0.000 (0.005) Extract Features: [630/1136] Time 0.122 (0.123) Data 0.000 (0.005) Extract Features: [640/1136] Time 0.112 (0.123) Data 0.000 (0.005) Extract Features: [650/1136] Time 0.112 (0.123) Data 0.000 (0.005) Extract Features: [660/1136] Time 0.120 (0.123) Data 0.000 (0.005) Extract Features: [670/1136] Time 0.111 (0.123) Data 0.000 (0.004) Extract Features: [680/1136] Time 0.111 (0.123) Data 0.000 (0.004) Extract Features: [690/1136] Time 0.111 (0.123) Data 0.000 (0.004) Extract Features: [700/1136] Time 0.112 (0.123) Data 0.000 (0.004) Extract Features: [710/1136] Time 0.116 (0.122) Data 0.000 (0.004) Extract Features: [720/1136] Time 0.110 (0.122) Data 0.000 (0.004) Extract Features: [730/1136] Time 0.112 (0.122) Data 0.000 (0.004) Extract Features: [740/1136] Time 0.130 (0.122) Data 0.000 (0.004) Extract Features: [750/1136] Time 0.111 (0.122) Data 0.000 (0.004) Extract Features: [760/1136] Time 0.110 (0.122) Data 0.000 (0.004) Extract Features: [770/1136] Time 0.111 (0.122) Data 0.000 (0.004) Extract Features: [780/1136] Time 0.122 (0.122) Data 0.000 (0.004) Extract Features: [790/1136] Time 0.111 (0.122) Data 0.000 (0.004) Extract Features: [800/1136] Time 0.112 (0.121) Data 0.000 (0.004) Extract Features: [810/1136] Time 0.111 (0.121) Data 0.000 (0.004) Extract Features: [820/1136] Time 0.113 (0.121) Data 0.000 (0.004) Extract Features: [830/1136] Time 0.125 (0.121) Data 0.000 (0.004) Extract Features: [840/1136] Time 0.122 (0.121) Data 0.000 (0.004) Extract Features: [850/1136] Time 0.114 (0.121) Data 0.000 (0.004) Extract Features: [860/1136] Time 0.110 (0.121) Data 0.000 (0.004) Extract Features: [870/1136] Time 0.109 (0.121) Data 0.000 (0.003) Extract Features: [880/1136] Time 0.114 (0.121) Data 0.000 (0.003) Extract Features: [890/1136] Time 0.108 (0.121) Data 0.000 (0.003) Extract Features: [900/1136] Time 0.125 (0.121) Data 0.000 (0.003) Extract Features: [910/1136] Time 0.114 (0.121) Data 0.000 (0.003) Extract Features: [920/1136] Time 0.119 (0.121) Data 0.000 (0.003) Extract Features: [930/1136] Time 0.109 (0.121) Data 0.000 (0.003) Extract Features: [940/1136] Time 0.110 (0.120) Data 0.000 (0.003) Extract Features: [950/1136] Time 0.115 (0.120) Data 0.000 (0.003) Extract Features: [960/1136] Time 0.109 (0.120) Data 0.000 (0.003) Extract Features: [970/1136] Time 0.110 (0.120) Data 0.000 (0.003) Extract Features: [980/1136] Time 0.121 (0.120) Data 0.000 (0.003) Extract Features: [990/1136] Time 0.123 (0.120) Data 0.000 (0.003) Extract Features: [1000/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1010/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1020/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1030/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1040/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1050/1136] Time 0.109 (0.120) Data 0.000 (0.003) Extract Features: [1060/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1070/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1080/1136] Time 0.111 (0.120) Data 0.000 (0.003) Extract Features: [1090/1136] Time 0.109 (0.120) Data 0.000 (0.003) Extract Features: [1100/1136] Time 0.110 (0.119) Data 0.000 (0.003) Extract Features: [1110/1136] Time 0.122 (0.119) Data 0.000 (0.003) Extract Features: [1120/1136] Time 0.124 (0.119) Data 0.000 (0.003) Extract Features: [1130/1136] Time 0.122 (0.119) Data 0.000 (0.003) gathering features from rank no.0 load PCA parameters... Extract Features: [10/1667] Time 0.104 (0.375) Data 0.000 (0.270) Extract Features: [20/1667] Time 0.104 (0.239) Data 0.000 (0.135) Extract Features: [30/1667] Time 0.103 (0.194) Data 0.000 (0.090) Extract Features: [40/1667] Time 0.104 (0.171) Data 0.000 (0.068) Extract Features: [50/1667] Time 0.117 (0.160) Data 0.000 (0.054) Extract Features: [60/1667] Time 0.115 (0.152) Data 0.000 (0.045) Extract Features: [70/1667] Time 0.107 (0.147) Data 0.000 (0.039) Extract Features: [80/1667] Time 0.107 (0.143) Data 0.000 (0.034) Extract Features: [90/1667] Time 0.103 (0.139) Data 0.000 (0.030) Extract Features: [100/1667] Time 0.105 (0.145) Data 0.000 (0.037) Extract Features: [110/1667] Time 0.102 (0.147) Data 0.000 (0.039) Extract Features: [120/1667] Time 0.104 (0.150) Data 0.000 (0.042) Extract Features: [130/1667] Time 0.189 (0.152) Data 0.084 (0.044) Extract Features: [140/1667] Time 0.103 (0.152) Data 0.000 (0.045) Extract Features: [150/1667] Time 0.108 (0.149) Data 0.001 (0.042) Extract Features: [160/1667] Time 0.165 (0.152) Data 0.054 (0.044) Extract Features: [170/1667] Time 0.216 (0.152) Data 0.112 (0.044) Extract Features: [180/1667] Time 0.183 (0.156) Data 0.081 (0.048) Extract Features: [190/1667] Time 0.231 (0.158) Data 0.128 (0.050) Extract Features: [200/1667] Time 0.170 (0.158) Data 0.066 (0.051) Extract Features: [210/1667] Time 0.267 (0.159) Data 0.164 (0.052) Extract Features: [220/1667] Time 0.316 (0.161) Data 0.213 (0.054) Extract Features: [230/1667] Time 0.248 (0.161) Data 0.141 (0.055) Extract Features: [240/1667] Time 0.163 (0.162) Data 0.059 (0.055) Extract Features: [250/1667] Time 0.204 (0.163) Data 0.098 (0.056) Extract Features: [260/1667] Time 0.182 (0.161) Data 0.066 (0.055) Extract Features: [270/1667] Time 0.252 (0.162) Data 0.148 (0.055) Extract Features: [280/1667] Time 0.217 (0.162) Data 0.114 (0.056) Extract Features: [290/1667] Time 0.108 (0.160) Data 0.000 (0.054) Extract Features: [300/1667] Time 0.115 (0.159) Data 0.000 (0.052) Extract Features: [310/1667] Time 0.113 (0.158) Data 0.000 (0.050) Extract Features: [320/1667] Time 0.122 (0.156) Data 0.000 (0.049) Extract Features: [330/1667] Time 0.105 (0.155) Data 0.000 (0.047) Extract Features: [340/1667] Time 0.107 (0.154) Data 0.000 (0.046) Extract Features: [350/1667] Time 0.112 (0.152) Data 0.000 (0.045) Extract Features: [360/1667] Time 0.107 (0.151) Data 0.000 (0.043) Extract Features: [370/1667] Time 0.106 (0.150) Data 0.000 (0.042) Extract Features: [380/1667] Time 0.118 (0.149) Data 0.000 (0.041) Extract Features: [390/1667] Time 0.110 (0.148) Data 0.000 (0.040) Extract Features: [400/1667] Time 0.110 (0.147) Data 0.000 (0.039) Extract Features: [410/1667] Time 0.110 (0.146) Data 0.000 (0.038) Extract Features: [420/1667] Time 0.110 (0.145) Data 0.000 (0.037) Extract Features: [430/1667] Time 0.118 (0.145) Data 0.000 (0.036) Extract Features: [440/1667] Time 0.120 (0.144) Data 0.000 (0.036) Extract Features: [450/1667] Time 0.112 (0.143) Data 0.000 (0.035) Extract Features: [460/1667] Time 0.110 (0.142) Data 0.000 (0.034) Extract Features: [470/1667] Time 0.112 (0.142) Data 0.000 (0.033) Extract Features: [480/1667] Time 0.113 (0.141) Data 0.000 (0.033) Extract Features: [490/1667] Time 0.109 (0.141) Data 0.000 (0.032) Extract Features: [500/1667] Time 0.109 (0.140) Data 0.000 (0.031) Extract Features: [510/1667] Time 0.111 (0.140) Data 0.000 (0.031) Extract Features: [520/1667] Time 0.126 (0.139) Data 0.000 (0.030) Extract Features: [530/1667] Time 0.118 (0.139) Data 0.000 (0.030) Extract Features: [540/1667] Time 0.108 (0.138) Data 0.000 (0.029) Extract Features: [550/1667] Time 0.108 (0.138) Data 0.000 (0.028) Extract Features: [560/1667] Time 0.115 (0.137) Data 0.000 (0.028) Extract Features: [570/1667] Time 0.112 (0.137) Data 0.000 (0.027) Extract Features: [580/1667] Time 0.109 (0.136) Data 0.000 (0.027) Extract Features: [590/1667] Time 0.116 (0.136) Data 0.000 (0.027) Extract Features: [600/1667] Time 0.123 (0.136) Data 0.000 (0.026) Extract Features: [610/1667] Time 0.110 (0.135) Data 0.000 (0.026) Extract Features: [620/1667] Time 0.116 (0.135) Data 0.000 (0.025) Extract Features: [630/1667] Time 0.115 (0.135) Data 0.000 (0.025) Extract Features: [640/1667] Time 0.129 (0.135) Data 0.000 (0.024) Extract Features: [650/1667] Time 0.118 (0.135) Data 0.000 (0.024) Extract Features: [660/1667] Time 0.112 (0.134) Data 0.000 (0.024) Extract Features: [670/1667] Time 0.117 (0.134) Data 0.000 (0.023) Extract Features: [680/1667] Time 0.124 (0.134) Data 0.000 (0.023) Extract Features: [690/1667] Time 0.122 (0.134) Data 0.000 (0.023) Extract Features: [700/1667] Time 0.123 (0.133) Data 0.000 (0.022) Extract Features: [710/1667] Time 0.141 (0.133) Data 0.000 (0.022) Extract Features: [720/1667] Time 0.143 (0.133) Data 0.000 (0.022) Extract Features: [730/1667] Time 0.111 (0.133) Data 0.000 (0.021) Extract Features: [740/1667] Time 0.108 (0.133) Data 0.000 (0.021) Extract Features: [750/1667] Time 0.115 (0.132) Data 0.000 (0.021) Extract Features: [760/1667] Time 0.117 (0.132) Data 0.000 (0.021) Extract Features: [770/1667] Time 0.120 (0.132) Data 0.000 (0.020) Extract Features: [780/1667] Time 0.113 (0.132) Data 0.000 (0.020) Extract Features: [790/1667] Time 0.138 (0.132) Data 0.000 (0.020) Extract Features: [800/1667] Time 0.129 (0.131) Data 0.000 (0.020) Extract Features: [810/1667] Time 0.110 (0.131) Data 0.000 (0.019) Extract Features: [820/1667] Time 0.113 (0.131) Data 0.000 (0.019) Extract Features: [830/1667] Time 0.117 (0.131) Data 0.000 (0.019) Extract Features: [840/1667] Time 0.107 (0.131) Data 0.000 (0.019) Extract Features: [850/1667] Time 0.108 (0.131) Data 0.000 (0.018) Extract Features: [860/1667] Time 0.118 (0.130) Data 0.000 (0.018) Extract Features: [870/1667] Time 0.107 (0.130) Data 0.000 (0.018) Extract Features: [880/1667] Time 0.114 (0.130) Data 0.000 (0.018) Extract Features: [890/1667] Time 0.122 (0.130) Data 0.000 (0.018) Extract Features: [900/1667] Time 0.116 (0.130) Data 0.000 (0.017) Extract Features: [910/1667] Time 0.130 (0.130) Data 0.000 (0.017) Extract Features: [920/1667] Time 0.115 (0.130) Data 0.000 (0.017) Extract Features: [930/1667] Time 0.118 (0.129) Data 0.000 (0.017) Extract Features: [940/1667] Time 0.111 (0.129) Data 0.000 (0.017) Extract Features: [950/1667] Time 0.107 (0.129) Data 0.000 (0.017) Extract Features: [960/1667] Time 0.126 (0.129) Data 0.000 (0.016) Extract Features: [970/1667] Time 0.128 (0.129) Data 0.000 (0.016) Extract Features: [980/1667] Time 0.131 (0.129) Data 0.000 (0.016) Extract Features: [990/1667] Time 0.113 (0.129) Data 0.000 (0.016) Extract Features: [1000/1667] Time 0.110 (0.129) Data 0.000 (0.016) Extract Features: [1010/1667] Time 0.115 (0.129) Data 0.000 (0.016) Extract Features: [1020/1667] Time 0.113 (0.128) Data 0.000 (0.015) Extract Features: [1030/1667] Time 0.111 (0.128) Data 0.000 (0.015) Extract Features: [1040/1667] Time 0.134 (0.128) Data 0.000 (0.015) Extract Features: [1050/1667] Time 0.108 (0.128) Data 0.000 (0.015) Extract Features: [1060/1667] Time 0.110 (0.128) Data 0.000 (0.015) Extract Features: [1070/1667] Time 0.113 (0.128) Data 0.000 (0.015) Extract Features: [1080/1667] Time 0.136 (0.128) Data 0.000 (0.015) Extract Features: [1090/1667] Time 0.112 (0.128) Data 0.000 (0.014) Extract Features: [1100/1667] Time 0.109 (0.128) Data 0.000 (0.014) Extract Features: [1110/1667] Time 0.112 (0.128) Data 0.000 (0.014) Extract Features: [1120/1667] Time 0.141 (0.128) Data 0.000 (0.014) Extract Features: [1130/1667] Time 0.114 (0.127) Data 0.000 (0.014) Extract Features: [1140/1667] Time 0.136 (0.127) Data 0.000 (0.014) Extract Features: [1150/1667] Time 0.119 (0.127) Data 0.000 (0.014) Extract Features: [1160/1667] Time 0.120 (0.127) Data 0.000 (0.014) Extract Features: [1170/1667] Time 0.131 (0.127) Data 0.000 (0.013) Extract Features: [1180/1667] Time 0.118 (0.127) Data 0.000 (0.013) Extract Features: [1190/1667] Time 0.113 (0.127) Data 0.000 (0.013) Extract Features: [1200/1667] Time 0.119 (0.127) Data 0.000 (0.013) Extract Features: [1210/1667] Time 0.108 (0.127) Data 0.000 (0.013) Extract Features: [1220/1667] Time 0.111 (0.127) Data 0.000 (0.013) Extract Features: [1230/1667] Time 0.106 (0.127) Data 0.000 (0.013) Extract Features: [1240/1667] Time 0.122 (0.126) Data 0.000 (0.013) Extract Features: [1250/1667] Time 0.115 (0.126) Data 0.000 (0.013) Extract Features: [1260/1667] Time 0.124 (0.126) Data 0.000 (0.013) Extract Features: [1270/1667] Time 0.106 (0.126) Data 0.000 (0.012) Extract Features: [1280/1667] Time 0.116 (0.126) Data 0.000 (0.012) Extract Features: [1290/1667] Time 0.120 (0.126) Data 0.000 (0.012) Extract Features: [1300/1667] Time 0.117 (0.126) Data 0.000 (0.012) Extract Features: [1310/1667] Time 0.112 (0.126) Data 0.000 (0.012) Extract Features: [1320/1667] Time 0.108 (0.126) Data 0.000 (0.012) Extract Features: [1330/1667] Time 0.123 (0.126) Data 0.000 (0.012) Extract Features: [1340/1667] Time 0.118 (0.126) Data 0.000 (0.012) Extract Features: [1350/1667] Time 0.136 (0.126) Data 0.000 (0.012) Extract Features: [1360/1667] Time 0.107 (0.126) Data 0.000 (0.012) Extract Features: [1370/1667] Time 0.109 (0.125) Data 0.000 (0.012) Extract Features: [1380/1667] Time 0.109 (0.125) Data 0.000 (0.011) Extract Features: [1390/1667] Time 0.110 (0.125) Data 0.000 (0.011) Extract Features: [1400/1667] Time 0.108 (0.125) Data 0.000 (0.011) Extract Features: [1410/1667] Time 0.107 (0.125) Data 0.000 (0.011) Extract Features: [1420/1667] Time 0.119 (0.125) Data 0.000 (0.011) Extract Features: [1430/1667] Time 0.122 (0.125) Data 0.000 (0.011) Extract Features: [1440/1667] Time 0.117 (0.125) Data 0.000 (0.011) Extract Features: [1450/1667] Time 0.108 (0.125) Data 0.000 (0.011) Extract Features: [1460/1667] Time 0.108 (0.125) Data 0.000 (0.011) Extract Features: [1470/1667] Time 0.107 (0.125) Data 0.000 (0.011) Extract Features: [1480/1667] Time 0.106 (0.125) Data 0.000 (0.011) Extract Features: [1490/1667] Time 0.108 (0.124) Data 0.000 (0.011) Extract Features: [1500/1667] Time 0.107 (0.124) Data 0.000 (0.011) Extract Features: [1510/1667] Time 0.129 (0.124) Data 0.000 (0.011) Extract Features: [1520/1667] Time 0.108 (0.124) Data 0.000 (0.010) Extract Features: [1530/1667] Time 0.105 (0.124) Data 0.000 (0.010) Extract Features: [1540/1667] Time 0.109 (0.124) Data 0.000 (0.010) Extract Features: [1550/1667] Time 0.109 (0.124) Data 0.000 (0.010) Extract Features: [1560/1667] Time 0.108 (0.124) Data 0.000 (0.010) Extract Features: [1570/1667] Time 0.105 (0.124) Data 0.000 (0.010) Extract Features: [1580/1667] Time 0.111 (0.124) Data 0.000 (0.010) Extract Features: [1590/1667] Time 0.108 (0.123) Data 0.000 (0.010) Extract Features: [1600/1667] Time 0.110 (0.123) Data 0.000 (0.010) Extract Features: [1610/1667] Time 0.112 (0.123) Data 0.000 (0.010) Extract Features: [1620/1667] Time 0.105 (0.123) Data 0.000 (0.010) Extract Features: [1630/1667] Time 0.112 (0.123) Data 0.000 (0.010) Extract Features: [1640/1667] Time 0.106 (0.123) Data 0.000 (0.010) Extract Features: [1650/1667] Time 0.128 (0.123) Data 0.000 (0.010) Extract Features: [1660/1667] Time 0.114 (0.123) Data 0.000 (0.010) gathering features from rank no.0 ===> Start calculating pairwise distances ===> Start calculating recalls Recall Scores: top-1 1.3% top-5 5.0% top-10 9.7%
It shows that resume='', indicating that you did not load any model weights.
I think whether some default settings about dataset preprocessing are wrong with your current code, because the evaluation results of the models you provide and i train are similarly much lower than you report.