def train(**kwargs):
opt._parse(kwargs)
#opt.lr=0.00002
opt.model_name='PCB'
# set random seed and cudnn benchmark
torch.manual_seed(opt.seed)
os.makedirs(opt.save_dir, exist_ok=True)
use_gpu = torch.cuda.is_available()
sys.stdout = Logger(osp.join(opt.save_dir, 'log_train.txt'))
print('=========user config==========')
pprint(opt._state_dict())
print('============end===============')
if use_gpu:
print('currently using GPU')
cudnn.benchmark = True
torch.cuda.manual_seed_all(opt.seed)
else:
print('currently using cpu')
print('initializing dataset {}'.format(opt.dataset))
dataset = data_manager.init_dataset(name=opt.dataset, mode=opt.mode)
tgt_dataset = data_manager.init_dataset(name=opt.tgt_dataset,mode=opt.mode)
pin_memory = True if use_gpu else False
summary_writer = SummaryWriter(osp.join(opt.save_dir, 'tensorboard_log'))
trainloader = DataLoader(
ImageData(dataset.train, TrainTransform(opt.datatype)),
batch_size=opt.train_batch, num_workers=opt.workers,
pin_memory=pin_memory, drop_last=True
)
tgt_trainloader = DataLoader(
ImageData(tgt_dataset.train, TrainTransform(opt.datatype)),
batch_size=opt.train_batch,num_workers=opt.workers,
pin_memory=pin_memory,drop_last=True
)
tgt_queryloader = DataLoader(
ImageData(tgt_dataset.query, TestTransform(opt.datatype)),
batch_size=opt.test_batch, num_workers=opt.workers,
pin_memory=pin_memory
)
tgt_galleryloader = DataLoader(
ImageData(tgt_dataset.gallery, TestTransform(opt.datatype)),
batch_size=opt.test_batch, num_workers=opt.workers,
pin_memory=pin_memory
)
tgt_queryFliploader = DataLoader(
ImageData(tgt_dataset.query, TestTransform(opt.datatype, True)),
batch_size=opt.test_batch, num_workers=opt.workers,
pin_memory=pin_memory
)
tgt_galleryFliploader = DataLoader(
ImageData(tgt_dataset.gallery, TestTransform(opt.datatype, True)),
batch_size=opt.test_batch, num_workers=opt.workers,
pin_memory=pin_memory
)
print('initializing model ...')
model = PCB(dataset.num_train_pids)
optim_policy = model.get_optim_policy()
start_epoch = opt.start_epoch
if opt.pretrained_model:
checkpoint = torch.load(opt.pretrained_model)
state_dict = checkpoint['state_dict']
# state_dict = {k: v for k, v in state_dict.items() \
# if not ('reduction' in k or 'softmax' in k)}
try:
model.load_state_dict(state_dict, False)
print('load pretrained model ' + opt.pretrained_model)
except:
RuntimeError('please keep the same size with source dataset..')
else:
raise RuntimeError('please load a pre-trained model...')
print('model size: {:.5f}M'.format(sum(p.numel() for p in model.parameters()) / 1e6))
if use_gpu:
model = nn.DataParallel(model).cuda()
reid_evaluator = ResNetEvaluator(model)
if opt.evaluate:
print('transfer directly....... ')
reid_evaluator.evaluate(tgt_queryloader, tgt_galleryloader,
tgt_queryFliploader, tgt_galleryFliploader, re_ranking=opt.re_ranking, savefig=opt.savefig)
return
#xent_criterion = CrossEntropyLabelSmooth(dataset.num_train_pids)
embedding_criterion = SelfTraining_TripletLoss(margin=0.5,num_instances=4)
# def criterion(triplet_y, softmax_y, labels):
# losses = [embedding_criterion(output, labels)[0] for output in triplet_y] + \
# [xent_criterion(output, labels) for output in softmax_y]
# loss = sum(losses)
# return loss
def criterion(triplet_y, softmax_y, labels):
#losses = [torch.sum(torch.stack([xent_criterion(logits, labels) for logits in softmax_y]))]
losses = [torch.sum(torch.stack([embedding_criterion(output,labels) for output in triplet_y]))]
loss = sum(losses)
return loss
# get optimizer
if opt.optim == "sgd":
optimizer = torch.optim.SGD(optim_policy, lr=opt.lr, momentum=0.9, weight_decay=opt.weight_decay)
else:
optimizer = torch.optim.Adam(optim_policy, lr=opt.lr, weight_decay=opt.weight_decay)
# get trainer and evaluator
reid_trainer = PCBTrainer(opt, model, optimizer, criterion, summary_writer)
def adjust_lr(optimizer, ep):
if ep < 50:
lr = opt.lr * (ep // 5 + 1)
elif ep < 200:
lr = opt.lr*10
elif ep < 300:
lr = opt.lr
else:
lr = opt.lr*0.1
for p in optimizer.param_groups:
p['lr'] = lr
# start training
best_rank1 = opt.best_rank
best_epoch = 0
print('transfer directly.....')
reid_evaluator.evaluate(tgt_queryloader, tgt_galleryloader,
tgt_queryFliploader, tgt_galleryFliploader, re_ranking=opt.re_ranking, savefig=opt.savefig)
for iter_n in range(start_epoch,opt.max_epoch):
if opt.lambda_value == 0:
source_features = 0
else:
# get source datas' feature
print('Iteration {}: Extracting Source Dataset Features...'.format(iter_n + 1))
source_features, _ = extract_pcb_features(model, trainloader)
# extract training images' features
print('Iteration {}: Extracting Target Dataset Features...'.format(iter_n + 1))
target_features, _ = extract_pcb_features(model, tgt_trainloader)
# synchronization feature order with dataset.train
# calculate distance and rerank result
print('Calculating feature distances...')
target_features = target_features.numpy()
rerank_dist = re_ranking(
source_features, target_features, lambda_value=opt.lambda_value)
if iter_n == 0:
# DBSCAN cluster
tri_mat = np.triu(rerank_dist, 1) # tri_mat.dim=2 取上三角
tri_mat = tri_mat[np.nonzero(tri_mat)] # tri_mat.dim=1
tri_mat = np.sort(tri_mat, axis=None)
top_num = np.round(opt.rho * tri_mat.size).astype(int)
eps = tri_mat[:top_num].mean() # DBSCAN聚类半径
print('eps in cluster: {:.3f}'.format(eps))
cluster = DBSCAN(eps=eps, min_samples=4, metric='precomputed', n_jobs=8)
# select & cluster images as training set of this epochs
print('Clustering and labeling...')
labels = cluster.fit_predict(rerank_dist)
del(rerank_dist)
del(source_features)
del(target_features)
try:
gc.collect()
except:
print('cannot collect')
num_ids = len(set(labels)) - 1
print('Iteration {} have {} training ids'.format(iter_n + 1, num_ids))
# generate new dataset
new_dataset = []
for (fname, _, _), label in zip(tgt_dataset.train, labels):
if label == -1:
continue
# dont need to change codes in trainer.py _parsing_input function and sampler function after add 0
new_dataset.append((fname, label, 0))
print('Iteration {} have {} training images'.format(iter_n + 1, len(new_dataset)))
selftrain_loader = DataLoader(
ImageData(new_dataset, TrainTransform(opt.datatype)),
sampler=RandomIdentitySampler(new_dataset, opt.num_instances),
batch_size=opt.train_batch, num_workers=opt.workers,
pin_memory=pin_memory, drop_last=True
)
# train model with new generated dataset
trainer = PCBTrainer(opt, model, optimizer, criterion, summary_writer)
reid_evaluator = ResNetEvaluator(model)
# Start training
for epoch in range(opt.selftrain_iterations):
trainer.train(epoch, selftrain_loader)
# skip if not save model
if opt.eval_step > 0 and (iter_n + 1) % opt.eval_step == 0 or (iter_n + 1) == opt.max_epoch:
# just avoid out of memory during eval,and can't save the model
if use_gpu:
state_dict = model.module.state_dict()
else:
state_dict = model.state_dict()
save_checkpoint({'state_dict': state_dict, 'epoch': iter_n + 1},
is_best=0, save_dir=opt.save_dir,
filename='checkpoint_ep' + str(iter_n + 1) + '.pth.tar')
if (iter_n + 1) % (opt.eval_step*4) == 0:
if opt.mode == 'class':
rank1 = test(model, tgt_queryloader)
else:
rank1 = reid_evaluator.evaluate(tgt_queryloader, tgt_galleryloader, tgt_queryFliploader,
tgt_galleryFliploader)
is_best = rank1 > best_rank1
if is_best:
best_rank1 = rank1
best_epoch = iter_n + 1
if use_gpu:
state_dict = model.module.state_dict()
else:
state_dict = model.state_dict()
if is_best:
save_checkpoint({'state_dict': state_dict, 'epoch': iter_n + 1},
is_best=is_best, save_dir=opt.save_dir,
filename='checkpoint_ep' + str(iter_n + 1) + '.pth.tar')
print('Best rank-1 {:.1%}, achived at epoch {}'.format(best_rank1, best_epoch))
def train(**kwargs):
opt._parse(kwargs)
opt.model_name = 'bfe_test'
# set random seed and cudnn benchmark
torch.manual_seed(opt.seed)
os.makedirs(opt.save_dir, exist_ok=True)
use_gpu = torch.cuda.is_available()
sys.stdout = Logger(osp.join(opt.save_dir, 'log_train.txt'))
print('=========user config==========')
pprint(opt._state_dict())
print('============end===============')
if use_gpu:
print('currently using GPU')
cudnn.benchmark = True
torch.cuda.manual_seed_all(opt.seed)
else:
print('currently using cpu')
print('initializing dataset {}'.format(opt.dataset))
dataset = data_manager.init_dataset(name=opt.dataset, mode=opt.mode)
pin_memory = True if use_gpu else False
summary_writer = SummaryWriter(osp.join(opt.save_dir, 'tensorboard_log'))
trainloader = DataLoader(ImageData(dataset.train,
TrainTransform(opt.datatype)),
sampler=RandomIdentitySampler(
dataset.train, opt.num_instances),
batch_size=opt.train_batch,
num_workers=opt.workers,
pin_memory=pin_memory,
drop_last=True)
queryloader = DataLoader(ImageData(dataset.query,
TestTransform(opt.datatype)),
batch_size=opt.test_batch,
num_workers=opt.workers,
pin_memory=pin_memory)
galleryloader = DataLoader(ImageData(dataset.gallery,
TestTransform(opt.datatype)),
batch_size=opt.test_batch,
num_workers=opt.workers,
pin_memory=pin_memory)
queryFliploader = DataLoader(ImageData(dataset.query,
TestTransform(opt.datatype, True)),
batch_size=opt.test_batch,
num_workers=opt.workers,
pin_memory=pin_memory)
galleryFliploader = DataLoader(ImageData(dataset.gallery,
TestTransform(opt.datatype,
True)),
batch_size=opt.test_batch,
num_workers=opt.workers,
pin_memory=pin_memory)
print('initializing model ...')
model = BFE(dataset.num_train_pids, 1.0, 0.33)
optim_policy = model.get_optim_policy()
if opt.pretrained_model:
state_dict = torch.load(opt.pretrained_model)['state_dict']
# state_dict = {k: v for k, v in state_dict.items() \
# if not ('reduction' in k or 'softmax' in k)}
model.load_state_dict(state_dict, False)
print('load pretrained model ' + opt.pretrained_model)
print('model size: {:.5f}M'.format(
sum(p.numel() for p in model.parameters()) / 1e6))
if use_gpu:
model = nn.DataParallel(model).cuda()
reid_evaluator = ResNetEvaluator(model)
if opt.evaluate:
reid_evaluator.evaluate(queryloader,
galleryloader,
queryFliploader,
galleryFliploader,
re_ranking=opt.re_ranking,
savefig=opt.savefig)
return
# xent_criterion = nn.CrossEntropyLoss()
xent_criterion = CrossEntropyLabelSmooth(dataset.num_train_pids)
if opt.loss == 'triplet':
embedding_criterion = TripletLoss(opt.margin)
elif opt.loss == 'lifted':
embedding_criterion = LiftedStructureLoss(hard_mining=True)
elif opt.loss == 'weight':
embedding_criterion = Margin()
def criterion(triplet_y, softmax_y, labels):
losses = [embedding_criterion(output, labels)[0] for output in triplet_y] + \
[xent_criterion(output, labels) for output in softmax_y]
loss = sum(losses)
return loss
# get optimizer
if opt.optim == "sgd":
optimizer = torch.optim.SGD(optim_policy,
lr=opt.lr,
momentum=0.9,
weight_decay=opt.weight_decay)
else:
optimizer = torch.optim.Adam(optim_policy,
lr=opt.lr,
weight_decay=opt.weight_decay)
start_epoch = opt.start_epoch
# get trainer and evaluator
reid_trainer = cls_tripletTrainer(opt, model, optimizer, criterion,
summary_writer)
def adjust_lr(optimizer, ep):
if ep < 10:
lr = opt.lr * 0.1 * (ep / 10.0) # warm_up
elif ep < 50:
lr = opt.lr * (ep // 5 + 1)
elif ep < 200:
lr = opt.lr * 10.0
elif ep < 300:
lr = opt.lr
else:
lr = opt.lr * 0.1
for p in optimizer.param_groups:
p['lr'] = lr
# start training
best_rank1 = opt.best_rank
best_epoch = 0
for epoch in range(start_epoch, opt.max_epoch):
if opt.adjust_lr:
adjust_lr(optimizer, epoch + 1)
reid_trainer.train(epoch, trainloader)
# skip if not save model
if opt.eval_step > 0 and (epoch + 1) % opt.eval_step == 0 or (
epoch + 1) == opt.max_epoch:
if opt.mode == 'class':
rank1 = test(model, queryloader)
else:
rank1 = reid_evaluator.evaluate(queryloader, galleryloader,
queryFliploader,
galleryFliploader)
is_best = rank1 > best_rank1
if is_best:
best_rank1 = rank1
best_epoch = epoch + 1
if use_gpu:
state_dict = model.module.state_dict()
else:
state_dict = model.state_dict()
save_checkpoint({
'state_dict': state_dict,
'epoch': epoch + 1
},
is_best=is_best,
save_dir=opt.save_dir,
filename='checkpoint_ep' + str(epoch + 1) +
'.pth.tar')
print('Best rank-1 {:.1%}, achived at epoch {}'.format(
best_rank1, best_epoch))
def train(**kwargs):
opt._parse(kwargs)
os.makedirs(opt.save_dir, exist_ok=True)
use_gpu = torch.cuda.is_available()
sys.stdout = Logger(osp.join(opt.save_dir, 'log_train.txt'))
print('=========user config==========')
pprint(opt._state_dict())
print('============end===============')
if use_gpu:
print('currently using GPU')
cudnn.benchmark = True
torch.cuda.manual_seed_all(opt.seed)
else:
print('currently using cpu')
print('initializing dataset {}'.format(opt.dataset))
dataset = data_manager.init_dataset(name=opt.dataset, use_all=opt.use_all)
summary_writer = SummaryWriter(osp.join(opt.save_dir, 'tensorboard_log'))
# load data
pin_memory = True if use_gpu else False
dataloader = load_data(dataset, pin_memory)
print('initializing model ...')
if opt.loss == 'softmax' or opt.loss == 'softmax_triplet':
model = ResNetBuilder(dataset.num_train_pids, opt.last_stride, True)
elif opt.loss == 'triplet':
model = ResNetBuilder(None, opt.last_stride, True)
if opt.pretrained_model:
if use_gpu:
state_dict = torch.load(opt.pretrained_model)['state_dict']
else:
state_dict = torch.load(opt.pretrained_model,
map_location='cpu')['state_dict']
model.load_state_dict(state_dict, False)
print('load pretrained model ' + opt.pretrained_model)
print('model size: {:.5f}M'.format(
sum(p.numel() for p in model.parameters()) / 1e6))
optim_policy = model.get_optim_policy()
if use_gpu:
model = nn.DataParallel(model).cuda()
reid_evaluator = ResNetEvaluator(model)
if opt.evaluate:
reid_evaluator.evaluate(dataloader['query'],
dataloader['gallery'],
dataloader['queryFlip'],
dataloader['galleryFlip'],
savefig=opt.savefig)
return
criterion = get_loss()
# optimizer
if opt.optim == "sgd":
optimizer = torch.optim.SGD(optim_policy,
lr=opt.lr,
momentum=0.9,
weight_decay=5e-4)
else:
optimizer = torch.optim.Adam(optim_policy,
lr=opt.lr,
weight_decay=5e-4)
scheduler = WarmupMultiStepLR(optimizer, [40, 70], 0.1, 0.01, 10, 'linear')
start_epoch = opt.start_epoch
# get trainer and evaluator
reid_trainer = Trainer(opt, model, optimizer, criterion, summary_writer)
# start training
best_rank1 = opt.best_rank
best_epoch = 0
for epoch in range(start_epoch, opt.max_epoch):
scheduler.step()
reid_trainer.train(epoch, dataloader['train'])
# skip if not save model
if opt.eval_step > 0 and (epoch + 1) % opt.eval_step == 0 or (
epoch + 1) == opt.max_epoch:
rank1 = reid_evaluator.evaluate(dataloader['query'],
dataloader['gallery'],
dataloader['queryFlip'],
dataloader['galleryFlip'])
is_best = rank1 > best_rank1
if is_best:
best_rank1 = rank1
best_epoch = epoch + 1
state_dict = model.state_dict()
save_checkpoint({
'state_dict': state_dict,
'epoch': epoch + 1
},
is_best=is_best,
save_dir=opt.save_dir,
filename='checkpoint_ep' + str(epoch + 1) +
'.pth.tar')
print('Best rank-1 {:.1%}, achived at epoch {}'.format(
best_rank1, best_epoch))