def eval(net, test_num=10000):
net.eval()
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(opt.img_size, opt.img_size))
img = transforms.Compose([transforms.ToTensor(), caffe_normalize])(img)
return img, boxes, labels
dataset = ListDataset(root=opt.eval_img_root,
list_file=opt.eval_img_list,
transform=transform)
box_coder = SSDBoxCoder(net.module)
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
# with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
# gt_difficults = []
# for line in f.readlines():
# line = line.strip().split()
# d = np.array([int(x) for x in line[1:]])
# gt_difficults.append(d)
nums_img = dataset.__len__()
for i in tqdm(range(nums_img)):
inputs, box_targets, label_targets = dataset.__getitem__(i)
gt_boxes.append(box_targets)
gt_labels.append(label_targets)
inputs = inputs.unsqueeze(0)
with torch.no_grad():
loc_preds, cls_preds = net(Variable(inputs.cuda()))
box_preds, label_preds, score_preds = box_coder.decode(
loc_preds.cpu().data.squeeze(),
F.softmax(cls_preds.squeeze(), dim=1).cpu().data,
score_thresh=0.1)
pred_boxes.append(box_preds)
pred_labels.append(label_preds)
pred_scores.append(score_preds)
aps = (voc_eval(pred_boxes,
pred_labels,
pred_scores,
gt_boxes,
gt_labels,
gt_difficults=None,
iou_thresh=0.5,
use_07_metric=False))
net.train()
return aps
def prepare_data(ssd_box_coder, batch_size, n_workers, img_size, data_paths,
drop_last=False):
trainset = ListDataset(
root=data_paths['vda_root'], list_file=data_paths['vda_list_train'],
transform=partial(
transform_train, ssd_box_coder=ssd_box_coder, img_size=img_size
)
)
testset = ListDataset(
root=data_paths['vda_root'], list_file=data_paths['vda_list_test'],
transform=partial(
transform_test, ssd_box_coder=ssd_box_coder, img_size=img_size
)
)
coco_trainset = ListDataset(
root=data_paths['coco_root'], list_file=data_paths['coco_list_train'],
transform=partial(transform_train_target, img_size=img_size)
)
coco_trainset_full = ListDataset(
root=data_paths['coco_root'], list_file=data_paths['coco_list_train'],
transform=partial(
transform_train, ssd_box_coder=ssd_box_coder, img_size=img_size
)
)
coco_testset = ListDataset(
root=data_paths['coco_root_test'], list_file=data_paths['coco_list_test'],
transform=partial(
transform_test, ssd_box_coder=ssd_box_coder, img_size=img_size
)
)
source_train_loader = torch.utils.data.DataLoader(
trainset, batch_size=batch_size, shuffle=True, num_workers=n_workers, drop_last=drop_last
)
target_train_loader = torch.utils.data.DataLoader(
coco_trainset, batch_size=batch_size, shuffle=True, num_workers=n_workers, drop_last=drop_last
)
target_full_train_loader = torch.utils.data.DataLoader(
coco_trainset_full, batch_size=batch_size, shuffle=True, num_workers=n_workers
)
testloader = torch.utils.data.DataLoader(
testset, batch_size=batch_size, shuffle=False, num_workers=0, drop_last=drop_last
)
cocoloader = torch.utils.data.DataLoader(
coco_testset, batch_size=batch_size, shuffle=False, num_workers=0, drop_last=drop_last
)
vda_test_boxes_labels, coco_test_boxes_labels = get_test_bboxes(
data_paths['vda_list_test'], data_paths['vda_root'],
data_paths['coco_list_test'], data_paths['coco_root_test'],
)
return namedtuple('Data', locals().keys())(**locals())
def box_label_list(root_dir, list_file):
return list(
ListDataset(
root=root_dir,
list_file=list_file,
transform=lambda img, boxes, labels:
(resize(img, boxes, size=(img_size, img_size))[1], labels)))
def eval(net):
net.eval()
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
dataset = ListDataset(root=args.data_root, \
list_file=args.voc07_test,
transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=1,
shuffle=False,
num_workers=8)
box_coder = SSDBoxCoder(net)
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
gt_difficults = []
for line in f.readlines():
line = line.strip().split()
d = np.array([int(x) for x in line[1:]])
gt_difficults.append(d)
for i, (inputs, box_targets, label_targets) in enumerate(dataloader):
print('%d/%d' % (i, len(dataloader)))
gt_boxes.append(box_targets.squeeze(0))
gt_labels.append(label_targets.squeeze(0))
loc_preds, cls_preds = net(Variable(inputs.cuda(), volatile=True))
box_preds, label_preds, score_preds = box_coder.decode(
loc_preds.cpu().data.squeeze(),
F.softmax(cls_preds.squeeze(), dim=1).cpu().data,
score_thresh=0.01)
pred_boxes.append(box_preds)
pred_labels.append(label_preds)
pred_scores.append(score_preds)
aps = (voc_eval(pred_boxes,
pred_labels,
pred_scores,
gt_boxes,
gt_labels,
gt_difficults,
iou_thresh=0.5,
use_07_metric=True))
net.train()
return aps
def evaluate(net, img_dir, list_file, img_size, test_code):
net.cuda()
net.eval()
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
print('Loading dataset..')
dataset = ListDataset(root=img_dir, list_file=list_file,
transform=transform)
if test_code:
dataset.num_imgs = 1
dl = torch.utils.data.DataLoader(dataset, batch_size=1, shuffle=False,
num_workers=2)
box_coder = SSDBoxCoder(net)
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
tqdm_dl = tqdm(dl, desc="Evaluate", ncols=0)
for i, (inputs, box_targets, label_targets) in enumerate(tqdm_dl):
gt_boxes.append(box_targets.squeeze(0))
gt_labels.append(label_targets.squeeze(0))
loc_preds, cls_preds = net(Variable(inputs.cuda(), volatile=True))
box_preds, label_preds, score_preds = box_coder.decode(
loc_preds.cpu().data.squeeze(),
F.softmax(cls_preds.squeeze(), dim=1).cpu().data,
score_thresh=0.01)
pred_boxes.append(box_preds)
pred_labels.append(label_preds)
pred_scores.append(score_preds)
ap_map_dict = voc_eval(pred_boxes, pred_labels, pred_scores, gt_boxes,
gt_labels, iou_thresh=0.5, use_07_metric=False)
return ap_map_dict
return img, boxes, labels
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
enc_boxes, enc_labels = box_coder.encode(boxes, labels)
return img, enc_boxes, enc_labels
trainset = ListDataset(
root='/scratch2/mytmp/render_detection_visda',
list_file='/scratch2/mytmp/render_detection_result/listdataset/visda18'
'-detection-train.txt',
transform=transform_train)
testset = ListDataset(
root='/scratch2/mytmp/render_detection_visda',
list_file='/scratch2/mytmp/render_detection_result/listdataset/visda18'
'-detection-test.txt',
transform=transform_test)
cocoset = ListDataset(
root='/scratch2/data/coco/train2014',
list_file=
'/scratch2/mytmp/render_detection_result/listdataset/coco-train-short.txt',
transform=transform_test)
net.eval()
print('Preparing dataset..')
img_size = 512
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
dataset = ListDataset(root='/home/claude.cy/.data/all_images', \
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=1,
shuffle=False,
num_workers=2)
box_coder = SSDBoxCoder(net)
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
gt_difficults = []
print('Loading image..')
def caffe_normalize(x):
return transforms.Compose([
transforms.Lambda(lambda x:255*x[[2,1,0]]) ,
transforms.Normalize([104,117,123], (1,1,1)), # make it the same as caffe
# bgr and 0-255
])(x)
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(opt.img_size, opt.img_size))
img = transforms.Compose([
transforms.ToTensor(),
caffe_normalize
])(img)
return img, boxes, labels
dataset = ListDataset(root=opt.eval_img_root, list_file=opt.eval_img_list, transform=transform)
box_coder = SSDBoxCoder(net.module)
nums_img = dataset.__len__()
idx = random.randint(0, nums_img)
inputs, _, _ = dataset.__getitem__(idx)
inputs = inputs.unsqueeze(0)
with torch.no_grad():
loc_preds, cls_preds = net(Variable(inputs.cuda()))
boxes, labels, scores = box_coder.decode(
loc_preds.cpu().data.squeeze(), F.softmax(cls_preds.squeeze(), dim=1).cpu().data, score_thresh=0.5)
img = Image.open(opt.eval_img_root + dataset.fnames[idx])
sw = float(img.size[0])/float(opt.img_size)
sh = float(img.size[1])/float(opt.img_size)
boxes = boxes.type(torch.FloatTensor) * torch.tensor([sw, sh, sw, sh])
print('Preparing dataset..')
img_size = 512
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
dataset = ListDataset(root=args.root / 'voc_all_images',
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=1,
shuffle=False,
num_workers=2)
box_coder = FPNSSDBoxCoder()
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
gt_difficults = []
fill=(123, 116, 103))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/home/liukuang/data/kitti/training/image_2', \
list_file='torchcv/datasets/kitti/kitti12_train.txt', \
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/home/liukuang/data/kitti/training/image_2', \
list_file='torchcv/datasets/kitti/kitti12_val.txt', \
transform=transform_test)
return transforms.Compose([
transforms.Lambda(lambda x: 255 * x[[2, 1, 0]]),
transforms.Normalize([104, 117, 123],
(1, 1, 1)), # make it the same as caffe
# bgr and 0-255
])(x)
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(opt.img_size, opt.img_size))
img = transforms.Compose([transforms.ToTensor(), caffe_normalize])(img)
return img, boxes, labels
dataset = ListDataset(root=opt.eval_img_root,
list_file=opt.eval_img_list,
transform=transform)
box_coder = SSDBoxCoder(net.module)
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
#with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
# gt_difficults = []
# for line in f.readlines():
# line = line.strip().split()
# d = np.array([int(x) for x in line[1:]])
# gt_difficults.append(d)
def main(**kwargs):
opt._parse(kwargs)
vis = Visualizer(env=opt.env)
# Model
print('==> Building model..')
net = DSOD(num_classes=21)
# Dataset
print('==> Preparing dataset..')
box_coder = SSDBoxCoder(net)
trainset = ListDataset(root=opt.train_img_root,
list_file=opt.train_img_list,
transform=Transform(box_coder, True))
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=opt.batch_size,
shuffle=True,
num_workers=opt.num_worker,
pin_memory=True)
net.cuda()
net = torch.nn.DataParallel(net,
device_ids=range(torch.cuda.device_count()))
cudnn.benchmark = True
criterion = SSDLoss(num_classes=21)
optimizer = optim.SGD(net.parameters(),
lr=opt.lr,
momentum=0.9,
weight_decay=1e-4)
best_map_ = 0
best_loss = 1e100
start_epoch = 0
if opt.load_path is not None:
print('==> Resuming from checkpoint..')
checkpoint = torch.load(opt.load_path)
net.load_state_dict(checkpoint['net'])
best_loss = checkpoint['map']
start_epoch = checkpoint['epoch'] + 1
print('start_epoch = ', start_epoch, 'best_loss = ', best_loss)
for epoch in range(start_epoch, start_epoch + 100):
print('\nEpoch: ', epoch)
net.train()
train_loss = 0
optimizer.zero_grad()
ix = 0
for batch_idx, (inputs, loc_targets,
cls_targets) in tqdm(enumerate(trainloader)):
inputs = Variable(inputs.cuda())
loc_targets = Variable(loc_targets.cuda())
cls_targets = Variable(cls_targets.cuda())
loc_preds, cls_preds = net(inputs)
ix += 1
loss = criterion(loc_preds, loc_targets, cls_preds, cls_targets)
loss.backward()
train_loss += loss.data.item()
current_loss = train_loss / (1 + batch_idx)
if (batch_idx + 1) % (opt.iter_size) == 0:
for name, p in net.named_parameters():
p.grad.data.div_(ix)
ix = 0
optimizer.step()
optimizer.zero_grad()
if (batch_idx + 1) % opt.plot_every == 0:
vis.plot('loss', current_loss)
# img = predict(net, box_coder, os.path.join(opt.train_img_root, trainset.fnames[batch_idx]))
# vis.img('predict', np.array(img).transpose(2, 0, 1))
# if os.path.exists(opt.debug_file):
# import ipdb
# ipdb.set_trace()
print('current_loss: ', current_loss, 'best_loss: ', best_loss)
if (epoch + 1) % 20 == 0:
for param_group in optimizer.param_groups:
param_group['lr'] *= 0.1
if (epoch + 1) % opt.save_state_every == 0:
state = {
'net': net.state_dict(),
'map': current_loss,
'epoch': epoch,
}
torch.save(state, opt.checkpoint + '%s.pth' % epoch)
if current_loss < best_loss:
best_loss = current_loss
print('saving model at epoch: ', epoch)
state = {
'net': net.state_dict(),
'map': best_loss,
'epoch': epoch,
}
torch.save(state, opt.checkpoint + 'dsod.pth')
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(
root='/data/deva/recaptcha_data/labelled',
list_file=['/data/deva/recaptcha_data/labelled/bbox_dataset_train.txt'],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(
root='/data/deva/recaptcha_data/labelled',
img, boxes = random_paste(img, boxes, max_ratio=4, fill=(123,116,103))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img, boxes, size=(img_size,img_size), random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
# print ("labels: ")
# print (labels.size())
boxes, labels = box_coder.encode(boxes, labels)
# print (labels.size())
return img, boxes, labels
trainset = ListDataset(root='/home/ysdu/hardwareDisk/ysduDir/voc/voc_all_images',
list_file=['torchcv/datasets/voc/voc07_trainval.txt',
'torchcv/datasets/voc/voc12_trainval.txt'],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size,img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/home/ysdu/hardwareDisk/ysduDir/voc/voc_all_images',
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform_test)
def transform_train(img, boxes, labels):
img = random_distort(img)
if random.random() < 0.5:
img, boxes = random_paste(img, boxes, max_ratio=4, fill=(123,116,103))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img, boxes, size=(img_size,img_size), random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/search/odin/liukuang/data/voc_all_images',
list_file=['torchcv/datasets/voc/voc07_trainval.txt',
'torchcv/datasets/voc/voc12_trainval.txt'],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size,img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/search/odin/liukuang/data/voc_all_images',
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform_test)
def main(**kwargs):
opt._parse(kwargs)
vis = Visualizer(env=opt.env)
# Model
print('==> Building model..')
net = DSOD(num_classes=21)
start_epoch = 0 # start from epoch 0 or last epoch
if opt.load_path is not None:
print('==> Resuming from checkpoint..')
checkpoint = torch.load(opt.load_path)
net.load_state_dict(checkpoint['net'])
# Dataset
print('==> Preparing dataset..')
box_coder = SSDBoxCoder(net)
trainset = ListDataset(root=opt.data_root,
list_file=[opt.voc07_trainval, opt.voc12_trainval],
transform=Transform(box_coder, True))
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=opt.batch_size,
shuffle=True,
num_workers=8)
net.cuda()
net = torch.nn.DataParallel(net,
device_ids=range(torch.cuda.device_count()))
cudnn.benchmark = True
criterion = SSDLoss(num_classes=21)
optimizer = optim.SGD(net.parameters(),
lr=opt.lr,
momentum=0.9,
weight_decay=5e-4)
best_map_ = 0
for epoch in range(start_epoch, start_epoch + 200):
print('\nEpoch: %d' % epoch)
net.train()
train_loss = 0
for batch_idx, (inputs, loc_targets,
cls_targets) in tqdm(enumerate(trainloader)):
inputs = Variable(inputs.cuda())
loc_targets = Variable(loc_targets.cuda())
cls_targets = Variable(cls_targets.cuda())
optimizer.zero_grad()
loc_preds, cls_preds = net(inputs)
loss = criterion(loc_preds, loc_targets, cls_preds, cls_targets)
loss.backward()
optimizer.step()
train_loss += loss.data[0]
if (batch_idx + 1) % opt.plot_every == 0:
vis.plot('loss', train_loss / (batch_idx + 1))
img = predict(
net, box_coder,
os.path.join(opt.data_root, trainset.fnames[batch_idx]))
vis.img('predict', np.array(img).transpose(2, 0, 1))
if os.path.exists(opt.debug_file):
import ipdb
ipdb.set_trace()
aps = eval(net.module, test_num=epoch * 100 + 100)
map_ = aps['map']
if map_ > best_map_:
print('Saving..')
state = {
'net': net.state_dict(),
'map': best_map_,
'epoch': epoch,
}
best_map_ = map_
if not os.path.isdir(os.path.dirname(opt.checkpoint)):
os.mkdir(os.path.dirname(opt.checkpoint))
torch.save(state, opt.checkpoint + '/%s.pth' % best_map_)
for ii in range(len(labels_tmp)):
if labels_tmp[ii] != 0:
new_boxes.append(boxes_tmp[ii])
new_labels.append(labels_tmp[ii])
else:
att_box.append(boxes_tmp[ii])
boxes = torch.from_numpy(np.array(new_boxes))
labels = torch.from_numpy(np.array(new_labels))
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels, att_box #boxes_tmp, labels_tmp
trainset = ListDataset(root='/home/user/Mayank/box-attention/data/VOC/VOCdevkit/VOC2012/JPEGImages', \
list_file='torchcv/datasets/voc/voc12_trainval.txt', \
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size,img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485,0.456,0.406),(0.229,0.224,0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/home/user/Mayank/box-attention/data/VOC/VOCdevkit/VOC2012/JPEGImages', \
list_file='torchcv/datasets/voc/voc12_test.txt', \
transform=transform_test)
return img, boxes, labels
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=img_size, max_size=img_size)
img = pad(img, (img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/home/eee/ug/15085073/Projects/GRID/data/images',
list_file='torchcv/datasets/grid/training.txt',
transform=transform_train)
testset = ListDataset(root='/home/eee/ug/15085073/Projects/GRID/data/images',
list_file='torchcv/datasets/grid/validation.txt',
transform=transform_test)
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=4,
shuffle=True,
num_workers=8)
testloader = torch.utils.data.DataLoader(testset,
batch_size=8,
shuffle=False,
num_workers=8)
# Model
def main(**kwargs):
opt._parse(kwargs)
vis = Visualizer(env=opt.env)
# Model
print('==> Building model..')
net = DSOD(num_classes=21)
start_epoch = 0 # start from epoch 0 or last epoch
# Dataset
print('==> Preparing dataset..')
box_coder = SSDBoxCoder(net)
trainset = ListDataset(root=opt.data_root,
list_file=[opt.voc07_trainval, opt.voc12_trainval],
transform=Transform(box_coder, True))
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=opt.batch_size,
shuffle=True,
num_workers=8,
pin_memory=True)
net.cuda()
net = torch.nn.DataParallel(net,
device_ids=range(torch.cuda.device_count()))
cudnn.benchmark = True
if opt.load_path is not None:
print('==> Resuming from checkpoint..')
checkpoint = torch.load(opt.load_path)
net.module.load_state_dict(checkpoint['net'])
criterion = SSDLoss(num_classes=21)
optimizer = optim.SGD(net.parameters(),
lr=opt.lr,
momentum=0.9,
weight_decay=5e-4)
best_map_ = 0
best_loss = 1e100
for epoch in range(start_epoch, start_epoch + 200):
print('\nEpoch: %d' % epoch)
net.train()
train_loss = 0
optimizer.zero_grad()
ix = 0
for batch_idx, (inputs, loc_targets,
cls_targets) in tqdm(enumerate(trainloader)):
inputs = Variable(inputs.cuda())
loc_targets = Variable(loc_targets.cuda())
cls_targets = Variable(cls_targets.cuda())
loc_preds, cls_preds = net(inputs)
ix += 1
loss = criterion(loc_preds, loc_targets, cls_preds, cls_targets)
loss.backward()
train_loss += loss.data[0]
if (batch_idx + 1) % (opt.iter_size) == 0:
# if True:
for name, p in net.named_parameters():
p.grad.data.div_(ix)
ix = 0
optimizer.step()
optimizer.zero_grad()
if (batch_idx + 1) % opt.plot_every == 0:
vis.plot('loss', train_loss / (batch_idx + 1))
img = predict(
net, box_coder,
os.path.join(opt.data_root,
trainset.fnames[batch_idx]))
vis.img('predict', np.array(img).transpose(2, 0, 1))
if os.path.exists(opt.debug_file):
import ipdb
ipdb.set_trace()
# if (epoch+1)%10 == 0 :
# state = {
# 'net': net.module.state_dict(),
# # 'map': best_map_,
# 'epoch': epoch,
# }
# torch.save(state, opt.checkpoint + '/%s.pth' % epoch)
# if (epoch+1) % 30 == 0:
# for param_group in optimizer.param_groups:
# param_group['lr'] *= 0.1
current_loss = train_loss / (1 + batch_idx)
if current_loss < best_loss:
best_loss = current_loss
torch.save(net.module.state_dict(), '/tmp/dsod.pth')
if (epoch + 1) % opt.eval_every == 0:
net.module.load_state_dict(torch.load('/tmp/dsod.pth'))
aps = eval(net.module)
map_ = aps['map']
if map_ > best_map_:
print('Saving..')
state = {
'net': net.module.state_dict(),
'map': best_map_,
'epoch': epoch,
}
best_map_ = map_
if not os.path.isdir(os.path.dirname(opt.checkpoint)):
os.mkdir(os.path.dirname(opt.checkpoint))
best_path = opt.checkpoint + '/%s.pth' % best_map_
torch.save(state, best_path)
else:
net.module.load_state_dict(torch.load(best_path)['net'])
for param_group in optimizer.param_groups:
param_group['lr'] *= 0.1
vis.log(
dict(epoch=(epoch + 1),
map=map_,
loss=train_loss / (batch_idx + 1)))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root=args.data_root,
list_file=[args.voc07_trainval, args.voc12_trainval],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=args.batch_size,
from PIL import Image
from torchcv.datasets import ListDataset
from torchcv.visualizations import vis_image
from torchcv.transforms import resize, random_flip, random_crop, random_paste
import torchvision.transforms as transforms
def transform(img, boxes, labels):
img, boxes = random_paste(img, boxes, max_ratio=4, fill=(123, 116, 103))
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img, boxes, size=600, random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.ToTensor()(img)
return img, boxes, labels
dataset = ListDataset(root='/mnt/hgfs/D/mscoco/2017/val2017',
list_file='torchcv/datasets/mscoco/coco17_val.txt',
transform=transform)
img, boxes, labels = dataset[0]
vis_image(img, boxes)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(
root='/home/lyan/Documents/sample_uvb/all_imgs',
list_file=[
'/home/lyan/Documents/torchcv/torchcv/datasets/uvb/uvb_train.txt'
],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
def val_transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(IMG_SIZE, IMG_SIZE))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trn_ds = ListDataset(root=img_dir,
list_file=trn_labels_fpath,
transform=trn_transform,
test_code=args.test_code)
val_ds = ListDataset(root=img_dir_test,
list_file=val_labels_fpath,
transform=val_transform,
test_code=args.test_code)
trn_dl = torch.utils.data.DataLoader(trn_ds,
batch_size=BATCH_SIZE,
shuffle=shuffle,
num_workers=NUM_WORKERS)
val_dl = torch.utils.data.DataLoader(val_ds,
batch_size=BATCH_SIZE,
shuffle=False,
num_workers=NUM_WORKERS)
def calculate_loss(batch_idx, batch, volatile=False):
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root=args.root / 'voc_all_images',
list_file=[
'torchcv/datasets/voc/voc07_trainval.txt',
'torchcv/datasets/voc/voc12_trainval.txt'
],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root=args.root / 'voc_all_images',
return img, boxes, labels
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=img_size, max_size=img_size)
img = pad(img, (img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/home/liukuang/data/coco/train2017',
list_file='torchcv/datasets/mscoco/coco17_train.txt',
transform=transform_train)
testset = ListDataset(root='/home/liukuang/data/coco/val2017',
list_file='torchcv/datasets/mscoco/coco17_val.txt',
transform=transform_test)
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=1,
shuffle=True,
num_workers=8)
testloader = torch.utils.data.DataLoader(testset,
batch_size=8,
shuffle=False,
num_workers=8)
# Model
net.eval()
print('Preparing dataset..')
img_size = 300
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
dataset = ListDataset(root='/search/odin/liukuang/data/voc_all_images/', \
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=1,
shuffle=False,
num_workers=2)
box_coder = SSDBoxCoder()
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
gt_difficults = []
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='data/coco/images/train2017',
list_file=[
'torchcv/datasets/mscoco/coco17_train.txt',
],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/search/odin/liukuang/data/voc_all_images',
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_w, img_h),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
# transforms.Normalize([0.5]*3,[0.5]*3)
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root=path_img, \
list_file='torchcv/datasets/kitti/kitti12_train2.txt', \
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_w, img_h))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
# transforms.Normalize([0.5]*3,[0.5]*3)
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root=path_img, \
list_file='torchcv/datasets/kitti/kitti12_val2.txt', \
net.eval()
print('Preparing dataset..')
img_size = 512
def transform(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
return img, boxes, labels
dataset = ListDataset(root='/home/ysdu/hardwareDisk/ysduDir/voc/voc_all_images', \
list_file='torchcv/datasets/voc/voc07_test.txt',
transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=1,
shuffle=False,
num_workers=2)
box_coder = FPNSSDBoxCoder()
pred_boxes = []
pred_labels = []
pred_scores = []
gt_boxes = []
gt_labels = []
with open('torchcv/datasets/voc/voc07_test_difficult.txt') as f:
gt_difficults = []
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root=IMGS_ROOT,
list_file=[LIST_FILE],
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root=IMGS_ROOT,
list_file=LIST_FILE,