mirrors/deep-person-reid
1
0
Fork 0
mirror of https://github.com/KaiyangZhou/deep-person-reid.git synced 2025-06-03 14:53:23 +08:00
Code Issues Projects Releases Wiki Activity

add ranklogger; update variable names

Browse Source
This commit is contained in:
KaiyangZhou 2018-11-08 21:41:32 +00:00
parent 01fe76cf27
commit 57682b768e
4 changed files with 24 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
train_imgreid_xent.py
View File

@ -19,7 +19,7 @@ from torchreid import models
from torchreid.losses import CrossEntropyLoss, DeepSupervision from torchreid.losses import CrossEntropyLoss, DeepSupervision
from torchreid.utils.iotools import save_checkpoint, check_isfile from torchreid.utils.iotools import save_checkpoint, check_isfile
from torchreid.utils.avgmeter import AverageMeter from torchreid.utils.avgmeter import AverageMeter
from torchreid.utils.logger import Logger from torchreid.utils.loggers import Logger, RankLogger
from torchreid.utils.torchtools import set_bn_to_eval, count_num_param from torchreid.utils.torchtools import set_bn_to_eval, count_num_param
from torchreid.utils.reidtools import visualize_ranked_results from torchreid.utils.reidtools import visualize_ranked_results
from torchreid.eval_metrics import evaluate from torchreid.eval_metrics import evaluate
@ -91,7 +91,7 @@ def main():
if args.evaluate: if args.evaluate:
print("Evaluate only") print("Evaluate only")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
@ -106,6 +106,7 @@ def main():
return return
start_time = time.time() start_time = time.time()
ranklogger = RankLogger(args.source_names, args.target_names)
train_time = 0 train_time = 0
print("==> Start training") print("==> Start training")
@ -130,11 +131,12 @@ def main():
if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch: if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch:
print("==> Test") print("==> Test")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
rank1 = test(model, queryloader, galleryloader, use_gpu) rank1 = test(model, queryloader, galleryloader, use_gpu)
ranklogger.write(name, epoch + 1, rank1)
if use_gpu: if use_gpu:
state_dict = model.module.state_dict() state_dict = model.module.state_dict()
@ -151,6 +153,7 @@ def main():
elapsed = str(datetime.timedelta(seconds=elapsed)) elapsed = str(datetime.timedelta(seconds=elapsed))
train_time = str(datetime.timedelta(seconds=train_time)) train_time = str(datetime.timedelta(seconds=train_time))
print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time)) print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time))
ranklogger.show_summary()
def train(epoch, model, criterion, optimizer, trainloader, use_gpu, freeze_bn=False): def train(epoch, model, criterion, optimizer, trainloader, use_gpu, freeze_bn=False):

9
train_imgreid_xent_htri.py
View File

@ -19,7 +19,7 @@ from torchreid import models
from torchreid.losses import CrossEntropyLoss, TripletLoss, DeepSupervision from torchreid.losses import CrossEntropyLoss, TripletLoss, DeepSupervision
from torchreid.utils.iotools import save_checkpoint, check_isfile from torchreid.utils.iotools import save_checkpoint, check_isfile
from torchreid.utils.avgmeter import AverageMeter from torchreid.utils.avgmeter import AverageMeter
from torchreid.utils.logger import Logger from torchreid.utils.loggers import Logger, RankLogger
from torchreid.utils.torchtools import count_num_param from torchreid.utils.torchtools import count_num_param
from torchreid.utils.reidtools import visualize_ranked_results from torchreid.utils.reidtools import visualize_ranked_results
from torchreid.eval_metrics import evaluate from torchreid.eval_metrics import evaluate
@ -87,7 +87,7 @@ def main():
if args.evaluate: if args.evaluate:
print("Evaluate only") print("Evaluate only")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
@ -102,6 +102,7 @@ def main():
return return
start_time = time.time() start_time = time.time()
ranklogger = RankLogger(args.source_names, args.target_names)
train_time = 0 train_time = 0
print("==> Start training") print("==> Start training")
@ -115,11 +116,12 @@ def main():
if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch: if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch:
print("==> Test") print("==> Test")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
rank1 = test(model, queryloader, galleryloader, use_gpu) rank1 = test(model, queryloader, galleryloader, use_gpu)
ranklogger.write(name, epoch + 1, rank1)
if use_gpu: if use_gpu:
state_dict = model.module.state_dict() state_dict = model.module.state_dict()
@ -136,6 +138,7 @@ def main():
elapsed = str(datetime.timedelta(seconds=elapsed)) elapsed = str(datetime.timedelta(seconds=elapsed))
train_time = str(datetime.timedelta(seconds=train_time)) train_time = str(datetime.timedelta(seconds=train_time))
print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time)) print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time))
ranklogger.show_summary()
def train(epoch, model, criterion_xent, criterion_htri, optimizer, trainloader, use_gpu): def train(epoch, model, criterion_xent, criterion_htri, optimizer, trainloader, use_gpu):

9
train_vidreid_xent.py
View File

@ -20,7 +20,7 @@ from torchreid import models
from torchreid.losses import CrossEntropyLoss from torchreid.losses import CrossEntropyLoss
from torchreid.utils.iotools import save_checkpoint, check_isfile from torchreid.utils.iotools import save_checkpoint, check_isfile
from torchreid.utils.avgmeter import AverageMeter from torchreid.utils.avgmeter import AverageMeter
from torchreid.utils.logger import Logger from torchreid.utils.loggers import Logger, RankLogger
from torchreid.utils.torchtools import set_bn_to_eval, count_num_param from torchreid.utils.torchtools import set_bn_to_eval, count_num_param
from torchreid.utils.reidtools import visualize_ranked_results from torchreid.utils.reidtools import visualize_ranked_results
from torchreid.eval_metrics import evaluate from torchreid.eval_metrics import evaluate
@ -92,7 +92,7 @@ def main():
if args.evaluate: if args.evaluate:
print("Evaluate only") print("Evaluate only")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
@ -107,6 +107,7 @@ def main():
return return
start_time = time.time() start_time = time.time()
ranklogger = RankLogger(args.source_names, args.target_names)
train_time = 0 train_time = 0
print("==> Start training") print("==> Start training")
@ -131,11 +132,12 @@ def main():
if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch: if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch:
print("==> Test") print("==> Test")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
rank1 = test(model, queryloader, galleryloader, args.pool, use_gpu) rank1 = test(model, queryloader, galleryloader, args.pool, use_gpu)
ranklogger.write(name, epoch + 1, rank1)
if use_gpu: if use_gpu:
state_dict = model.module.state_dict() state_dict = model.module.state_dict()
@ -152,6 +154,7 @@ def main():
elapsed = str(datetime.timedelta(seconds=elapsed)) elapsed = str(datetime.timedelta(seconds=elapsed))
train_time = str(datetime.timedelta(seconds=train_time)) train_time = str(datetime.timedelta(seconds=train_time))
print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time)) print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time))
ranklogger.show_summary()
def train(epoch, model, criterion, optimizer, trainloader, use_gpu, freeze_bn=False): def train(epoch, model, criterion, optimizer, trainloader, use_gpu, freeze_bn=False):

9
train_vidreid_xent_htri.py
View File

@ -20,7 +20,7 @@ from torchreid import models
from torchreid.losses import CrossEntropyLoss, TripletLoss, DeepSupervision from torchreid.losses import CrossEntropyLoss, TripletLoss, DeepSupervision
from torchreid.utils.iotools import save_checkpoint, check_isfile from torchreid.utils.iotools import save_checkpoint, check_isfile
from torchreid.utils.avgmeter import AverageMeter from torchreid.utils.avgmeter import AverageMeter
from torchreid.utils.logger import Logger from torchreid.utils.loggers import Logger, RankLogger
from torchreid.utils.torchtools import count_num_param from torchreid.utils.torchtools import count_num_param
from torchreid.utils.reidtools import visualize_ranked_results from torchreid.utils.reidtools import visualize_ranked_results
from torchreid.eval_metrics import evaluate from torchreid.eval_metrics import evaluate
@ -89,7 +89,7 @@ def main():
if args.evaluate: if args.evaluate:
print("Evaluate only") print("Evaluate only")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
@ -104,6 +104,7 @@ def main():
return return
start_time = time.time() start_time = time.time()
ranklogger = RankLogger(args.source_names, args.target_names)
train_time = 0 train_time = 0
print("==> Start training") print("==> Start training")
@ -117,11 +118,12 @@ def main():
if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch: if (epoch + 1) > args.start_eval and args.eval_step > 0 and (epoch + 1) % args.eval_step == 0 or (epoch + 1) == args.max_epoch:
print("==> Test") print("==> Test")
for name in args.target: for name in args.target_names:
print("Evaluating {} ...".format(name)) print("Evaluating {} ...".format(name))
queryloader = testloader_dict[name]['query'] queryloader = testloader_dict[name]['query']
galleryloader = testloader_dict[name]['gallery'] galleryloader = testloader_dict[name]['gallery']
rank1 = test(model, queryloader, galleryloader, args.pool, use_gpu) rank1 = test(model, queryloader, galleryloader, args.pool, use_gpu)
ranklogger.write(name, epoch + 1, rank1)
if use_gpu: if use_gpu:
state_dict = model.module.state_dict() state_dict = model.module.state_dict()
@ -138,6 +140,7 @@ def main():
elapsed = str(datetime.timedelta(seconds=elapsed)) elapsed = str(datetime.timedelta(seconds=elapsed))
train_time = str(datetime.timedelta(seconds=train_time)) train_time = str(datetime.timedelta(seconds=train_time))
print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time)) print("Finished. Total elapsed time (h:m:s): {}. Training time (h:m:s): {}.".format(elapsed, train_time))
ranklogger.show_summary()
def train(epoch, model, criterion_xent, criterion_htri, optimizer, trainloader, use_gpu): def train(epoch, model, criterion_xent, criterion_htri, optimizer, trainloader, use_gpu):