def train(net):
net.train()
priorbox = PriorBox()
with torch.no_grad():
priors = priorbox.forward()
priors = priors.to(device)
dataloader = DataLoader(VOCDetection(),
batch_size=2,
collate_fn=detection_collate,
num_workers=12)
for epoch in range(1000):
loss_ls, loss_cs = [], []
load_t0 = time.time()
if epoch > 500:
adjust_learning_rate(optimizer, 1e-4)
for images, targets in dataloader:
images = images.to(device)
targets = [anno.to(device) for anno in targets]
out = net(images)
optimizer.zero_grad()
loss_l, loss_c = criterion(out, priors, targets)
loss = 2 * loss_l + loss_c
loss.backward()
optimizer.step()
loss_cs.append(loss_c.item())
loss_ls.append(loss_l.item())
load_t1 = time.time()
print(f'{np.mean(loss_cs)}, {np.mean(loss_ls)} time:{load_t1-load_t0}')
torch.save(net.state_dict(), 'Final_FaceBoxes.pth')
def train(net):
net.train()
priorbox = PriorBox()
with torch.no_grad():
priors = priorbox.forward()
priors = priors.to(device)
dataloader = DataLoader(VOCDetection(), batch_size=2, collate_fn=detection_collate, num_workers=12)
for epoch in range(1000):
loss_ls, loss_cs = [], []
load_t0 = time.time()
if epoch > 500:
adjust_learning_rate(optimizer, 1e-4)
for images, targets in dataloader:
images = images.to(device)
targets = [anno.to(device) for anno in targets]
out = net(images)
optimizer.zero_grad()
loss_l, loss_c = criterion(out, priors, targets)
loss = 2 * loss_l + loss_c
loss.backward()
optimizer.step()
loss_cs.append(loss_c.item())
loss_ls.append(loss_l.item())
load_t1 = time.time()
print(f'{np.mean(loss_cs)}, {np.mean(loss_ls)} time:{load_t1-load_t0}')
torch.save(net.state_dict(), 'Final_FaceBoxes.pth')
def init_priors(self, feature_maps, image_size):
# Hacky key system, but works....
key = ".".join([str(item) for i in range(len(feature_maps)) for item in feature_maps[i]]) + \
"," + ".".join([str(_) for _ in image_size])
if key in self.prior_cache:
return self.prior_cache[key].clone()
priorbox = PriorBox(self.cfg, image_size, feature_maps)
prior = priorbox.forward()
self.prior_cache[key] = prior.clone()
return prior
class SSD(nn.Module):
"""Single Shot Multibox Architecture
The network is composed of a base VGG network followed by the
added multibox conv layers. Each multibox layer branches into
1) conv2d for class conf scores
2) conv2d for localization predictions
3) associated priorbox layer to produce default bounding
boxes specific to the layer's feature map size.
See: https://arxiv.org/pdf/1512.02325.pdf for more details.
Args:
phase: (string) Can be "test" or "train"
size: input image size
base: VGG16 layers for input, size of either 300 or 500
extras: extra layers that feed to multibox loc and conf layers
head: "multibox head" consists of loc and conf conv layers
"""
def __init__(self, phase, size, Basenet, Neck, Head, cfg):
super(SSD, self).__init__()
self.phase = phase
self.cfg = cfg
self.priorbox = PriorBox(self.cfg)
self.priors = self.priorbox.forward()
self.size = size
# SSD network
self.basenet = Basenet
self.neck = Neck
self.head = Head
self.num_classes = cfg['num_classes']
self.softmax = nn.Softmax(dim=-1)
self.detect = Detect(self.num_classes, 0, 200, 0.01, 0.45,
variance=cfg['variance'], nms_kind=cfg['nms_kind'], beta1=cfg['beta1'])
def forward(self, x, phase):
"""Applies network layers and ops on input image(s) x.
Args:
x: input image or batch of images. Shape: [batch,3,300,300].
Return:
Depending on phase:
test:
Variable(tensor) of output class label predictions,
confidence score, and corresponding location predictions for
each object detected. Shape: [batch,topk,7]
train:
list of concat outputs from:
1: confidence layers, Shape: [batch*num_priors,num_classes]
2: localization layers, Shape: [batch,num_priors*4]
3: priorbox layers, Shape: [2,num_priors*4]
"""
x = self.basenet(x)
if self.neck is not None:
x = self.neck(x)
conf, loc = self.head(x)
loc = torch.cat([o.view(o.size(0), -1) for o in loc], 1)
conf = torch.cat([o.view(o.size(0), -1) for o in conf], 1)
if phase == "test":
output = self.detect.trans(
loc.view(loc.size(0), -1, 4), # loc preds
self.softmax(conf.view(conf.size(0), -1,
self.num_classes)), # conf preds
#self.priors.type(type(x.data)) # default boxes
self.priors
)
else:
output = (
loc.view(loc.size(0), -1, 4),
conf.view(conf.size(0), -1, self.num_classes),
self.priors
)
return output
def load_weights(self, base_file):
other, ext = os.path.splitext(base_file)
if ext == '.pkl' or '.pth':
print('Loading weights into state dict...')
self.load_state_dict(torch.load(base_file,
map_location=lambda storage, loc: storage))
print('Finished!')
else:
print('Sorry only .pth and .pkl files supported.')
(show_classes, num_classes, dataset, epoch_size, max_iter, testset) = load_dataset()
print('Loading Network...')
from models.detector import Detector
model = Detector(args.size, num_classes, args.backbone, args.neck)
model.train()
model.cuda()
num_param = sum(p.numel() for p in model.parameters() if p.requires_grad)
print('Total param is : {:e}'.format(num_param))
print('Preparing Optimizer & AnchorBoxes...')
optimizer = optim.SGD(tencent_trick(model), lr=args.lr, momentum=0.9, weight_decay=0.0005)
criterion = MultiBoxLoss(num_classes, mutual_guide=args.mutual_guide)
priorbox = PriorBox(args.base_anchor_size, args.size)
with torch.no_grad():
priors = priorbox.forward()
priors = priors.cuda()
if args.trained_model is not None:
print('loading weights from', args.trained_model)
state_dict = torch.load(args.trained_model)
model.load_state_dict(state_dict, strict=True)
else:
print('Training {}-{} on {} with {} images'.format(args.neck, args.backbone, dataset.name, len(dataset)))
os.makedirs(args.save_folder, exist_ok=True)
epoch = 0
timer = Timer()
for iteration in range(max_iter):
if iteration % epoch_size == 0:
# create batch iterator