How to train the P-DukeMTMC-reID dataset?

[magicffourier]

How can I train the network to get the same performance like your paper reported. I'm a little confused, need you sincerely help.

[hh23333]

You can first train the PCB baseline on P-DukeMTMC-reID with the script:

python scripts/main.py --root PATH_TO_DATAROOT \
 -s p_duke -t p_duke\
 --save-dir PATH_TO_EXPERIMENT_FOLDER/pduke_PCB\
 -a pcb_p6 --gpu-devices 0 --fixbase-epoch 0\
 --open-layers classifier fc\
 --new-layers classifier em\
 --transforms random_flip\
 --optim sgd --lr 0.02\
 --stepsize 25 50\
 --staged-lr --height 384 --width 192\
 --batch-size 32 --base-lr-mult 0.5

and then Extract the Pose imformation with the similiar method as in openpose_market.sh; Then train the PVPM with:

python scripts/main.py --load-pose --root PATH_TO_DATAROOT
 -s p_duke\
 -t p_duke\
 --save-dir PATH_TO_EXPERIMENT_FOLDER/PVPM\
 -a pose_p6s --gpu-devices 0\
 --fixbase-epoch 30\
 --open-layers pose_subnet\
 --new-layers pose_subnet\
 --transforms random_flip\
 --optim sgd --lr 0.02\
 --stepsize 15 25 --staged-lr\
 --height 384 --width 128\
 --batch-size 32\
 --start-eval 20\
 --eval-freq 10\
 --load-weights PATH_TO_EXPERIMENT_FOLDER/pduke_PCB/model.pth.tar-60\
 --train-sampler RandomIdentitySampler\
 --reg-matching-score-epoch 0\
 --graph-matching
 --max-epoch 30
 --part-score