def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--model_path', help='Path to model', type=str)
parser = parser.parse_args(args)
dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
transform=transforms.Compose([Normalizer(), Resizer()]))
dataset_val.image_ids = dataset_val.image_ids[:50] # TEST
# Create the model
retinanet = model.resnet50(num_classes=dataset_val.num_classes(), pretrained=True)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet.load_state_dict(torch.load(parser.model_path))
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet.load_state_dict(torch.load(parser.model_path))
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = False
retinanet.eval()
retinanet.module.freeze_bn()
coco_eval.evaluate_coco(dataset_val, retinanet)
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--load_model_path',
help='Path to model(.pt file)',
type=str)
parser = parser.parse_args(args)
# dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
# transform=transforms.Compose([Normalizer(), Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
# Create the model
# retinanet = model.resnet50(num_classes=dataset_val.num_classes(), pretrained=True)
retinanet = torch.load(parser.load_model_path)
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = False
retinanet.eval()
retinanet.module.freeze_bn()
coco_eval.evaluate_coco(dataset_val, retinanet)
def Val_Dataset(self, root_dir, coco_dir, img_dir, set_dir):
self.system_dict["dataset"]["val"]["status"] = True
self.system_dict["dataset"]["val"]["root_dir"] = root_dir
self.system_dict["dataset"]["val"]["coco_dir"] = coco_dir
self.system_dict["dataset"]["val"]["img_dir"] = img_dir
self.system_dict["dataset"]["val"]["set_dir"] = set_dir
self.system_dict["local"]["dataset_val"] = CocoDataset(
self.system_dict["dataset"]["val"]["root_dir"] + "/" +
self.system_dict["dataset"]["val"]["coco_dir"],
img_dir=self.system_dict["dataset"]["val"]["img_dir"],
set_dir=self.system_dict["dataset"]["val"]["set_dir"],
transform=transforms.Compose([Normalizer(),
Resizer()]))
self.system_dict["local"]["sampler_val"] = AspectRatioBasedSampler(
self.system_dict["local"]["dataset_val"],
batch_size=self.system_dict["params"]["batch_size"],
drop_last=False)
self.system_dict["local"]["dataloader_val"] = DataLoader(
self.system_dict["local"]["dataset_val"],
num_workers=self.system_dict["params"]["num_workers"],
collate_fn=collater,
batch_sampler=self.system_dict["local"]["sampler_val"])
print('Num validation images: {}'.format(
len(self.system_dict["local"]["dataset_val"])))
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser = parser.parse_args(args)
os.makedirs('after_augmentation_image_sample', exist_ok=True)
set_name = 'test'
dataset_sample = CocoDataset(
parser.coco_path,
set_name=set_name,
# transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
transform=transforms.Compose(
[Normalizer(), AugmenterWithImgaug(),
Resizer()]))
sampler = AspectRatioBasedSampler(dataset_sample,
batch_size=1,
drop_last=False)
dataloader_sample = DataLoader(dataset_sample,
num_workers=1,
collate_fn=collater,
batch_sampler=sampler)
unnormalize = UnNormalizer()
for idx, data in enumerate(dataloader_sample):
img = np.array(255 * unnormalize(data['img'][0, :, :, :])).copy()
img[img < 0] = 0
img[img > 255] = 255
img = np.transpose(img, (1, 2, 0))
img = cv2.cvtColor(img.astype(np.uint8), cv2.COLOR_BGR2RGB)
for annot in data['annot']:
annot = annot[0].data.numpy()
x1 = int(annot[0])
y1 = int(annot[1])
x2 = int(annot[2])
y2 = int(annot[3])
cv2.rectangle(img, (x1, y1), (x2, y2),
color=(0, 0, 255),
thickness=2)
cv2.imwrite(
'D:\\StreetView\\RenHong\\Pytorch_RetinaNet\\after_augmentation_image_sample\\'
+ 'sample_from_({})_'.format(set_name) + str(idx) + '.jpg',
img)
print("finish")
def Train_Dataset(self,
root_dir,
coco_dir,
img_dir,
set_dir,
batch_size=8,
image_size=512,
use_gpu=True,
num_workers=3):
self.system_dict["dataset"]["train"]["root_dir"] = root_dir
self.system_dict["dataset"]["train"]["coco_dir"] = coco_dir
self.system_dict["dataset"]["train"]["img_dir"] = img_dir
self.system_dict["dataset"]["train"]["set_dir"] = set_dir
self.system_dict["params"]["batch_size"] = batch_size
self.system_dict["params"]["image_size"] = image_size
self.system_dict["params"]["use_gpu"] = use_gpu
self.system_dict["params"]["num_workers"] = num_workers
self.system_dict["local"]["dataset_train"] = CocoDataset(
self.system_dict["dataset"]["train"]["root_dir"] + "/" +
self.system_dict["dataset"]["train"]["coco_dir"],
img_dir=self.system_dict["dataset"]["train"]["img_dir"],
set_dir=self.system_dict["dataset"]["train"]["set_dir"],
transform=transforms.Compose(
[Normalizer(), Augmenter(),
Resizer()]))
self.system_dict["local"]["sampler"] = AspectRatioBasedSampler(
self.system_dict["local"]["dataset_train"],
batch_size=self.system_dict["params"]["batch_size"],
drop_last=False)
self.system_dict["local"]["dataloader_train"] = DataLoader(
self.system_dict["local"]["dataset_train"],
num_workers=self.system_dict["params"]["num_workers"],
collate_fn=collater,
batch_sampler=self.system_dict["local"]["sampler"])
print('Num training images: {}'.format(
len(self.system_dict["local"]["dataset_train"])))
def main(args=None):
'''
test.py 会计算原始图片中的box的位置，而visualize.py返回的是resize和padding后的图片boudning box 位置
另外test.py支持对识别结果的保存
'''
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_classes', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--model', help='Path to model (.pt) file.')
parser.add_argument('--resultsavepath', help='path to save detection images')
parser.add_argument('--thresh_score', help="thresh score", type=float, default=.5)
parser = parser.parse_args(args)
# 创建结果的保存路径
if parser.resultsavepath:
os.makedirs(parser.resultsavepath, exist_ok=True)
if parser.dataset == 'coco':
dataset_val = CocoDataset(parser.coco_path, set_name='train2017', transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
#dataset_val = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes, transform=transforms.Compose([Normalizer(), Resizer()])) # 提示错误
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes, transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=1, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=1, collate_fn=collater, batch_sampler=sampler_val)
retinanet = torch.load(parser.model)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda() # 设置为多GPU的并行模式
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.eval()# 设置为评估模式
def draw_caption(image, box, caption):
b = np.array(box).astype(int)
# b[1]-20防止label超过上边界
cv2.putText(image, caption, (b[0], b[1]-10 if b[1]-20>0 else 30), cv2.FONT_HERSHEY_PLAIN, 1, (0, 0, 0), 2)
cv2.putText(image, caption, (b[0], b[1]-10 if b[1]-20>0 else 30), cv2.FONT_HERSHEY_PLAIN, 1, (255, 255, 255), 1)
if parser.resultsavepath:
result_csv = "{}_result.csv".format(os.path.splitext(os.path.basename(parser.csv_val))[0])
result_csv = os.path.join(parser.resultsavepath, result_csv)
print(result_csv)
result_csv_fd = open(result_csv, 'w')
for idx, data in enumerate(dataloader_val):
#print("data shape:", data.shape)
print(data['image_path'])
with torch.no_grad():
st = time.time()
if torch.cuda.is_available():
the_result = retinanet(data['img'].cuda().float())
else:
the_result = retinanet(data['img'].float())
print('Elapsed time: {}'.format(time.time()-st))
for image_index, (scores, classification, transformed_anchors) in enumerate(the_result):
idxs = np.where(scores.cpu()>parser.thresh_score)
image_path = data['image_path'][image_index]
scale = data['scale'][image_index]
img = cv2.imread(image_path)
if idxs[0].shape[0]==0:
result_csv_fd.write("{},,,,,\n".format(image_path))
else:
for j in range(idxs[0].shape[0]):
bbox = transformed_anchors[idxs[0][j], :]
x1 = int(bbox[0]/scale)
y1 = int(bbox[1]/scale)
x2 = int(bbox[2]/scale)
y2 = int(bbox[3]/scale)
label_name = dataset_val.labels[int(classification[idxs[0][j]])]
txt_draw = "%s %.2f" %(label_name, scores[j])
draw_caption(img, (x1, y1, x2, y2), txt_draw)
cv2.rectangle(img, (x1, y1), (x2, y2), color=(0, 0, 255), thickness=2)
if parser.resultsavepath:
result_csv_fd.write("{},{},{},{},{},{}\n".format(image_path,x1,y1,x2,y2,label_name))
print(label_name)
if parser.resultsavepath:
new_dir = os.path.join(parser.resultsavepath, os.path.dirname(image_path))
if not os.path.exists(new_dir):
os.makedirs(new_dir)
new_path = os.path.join(parser.resultsavepath, image_path)
cv2.imwrite(new_path, img)
#new_path = os.path.join(parser.resultsavepath, os.path.basename(image_path))
#cv2.imwrite(new_path, img)
#print("create result image:{} ".format(new_path))
else:
cv2.imshow('img', img)
cv2.waitKey(0)
if parser.resultsavepath:
result_csv_fd.close()
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
# parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--dataset_root',
default='/root/data/VOCdevkit/',
help=
'Dataset root directory path [/root/data/VOCdevkit/, /root/data/coco/, /root/data/FLIR_ADAS]'
)
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--resume',
default=None,
type=str,
help='Checkpoint state_dict file to resume training from')
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--batch_size',
default=16,
type=int,
help='Batch size for training')
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--lr',
'--learning_rate',
default=1e-4,
type=float,
help='initial learning rate')
parser.add_argument('--weight_decay',
default=5e-4,
type=float,
help='Weight decay')
parser.add_argument('-j',
'--workers',
default=4,
type=int,
metavar='N',
help='number of data loading workers (default: 4)')
parser.add_argument("--log",
default=False,
action="store_true",
help="Write log file.")
parser = parser.parse_args(args)
network_name = 'RetinaNet-Res{}'.format(parser.depth)
# print('network_name:', network_name)
net_logger = logging.getLogger('Network Logger')
formatter = logging.Formatter(LOGGING_FORMAT)
streamhandler = logging.StreamHandler()
streamhandler.setFormatter(formatter)
net_logger.addHandler(streamhandler)
if parser.log:
net_logger.setLevel(logging.INFO)
# logging.basicConfig(level=logging.DEBUG, format=LOGGING_FORMAT,
# filename=os.path.join('log', '{}.log'.format(network_name)), filemode='a')
filehandler = logging.FileHandler(os.path.join(
'log', '{}.log'.format(network_name)),
mode='a')
filehandler.setFormatter(formatter)
net_logger.addHandler(filehandler)
net_logger.info('Network Name: {:>20}'.format(network_name))
# Create the data loaders
if parser.dataset == 'coco':
if parser.dataset_root is None:
raise ValueError(
'Must provide --dataset_root when training on COCO,')
dataset_train = CocoDataset(parser.dataset_root,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.dataset_root,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'FLIR':
if parser.dataset_root is None:
raise ValueError(
'Must provide --dataset_root when training on FLIR,')
_scale = 1.2
dataset_train = FLIRDataset(parser.dataset_root,
set_name='train',
transform=transforms.Compose([
Normalizer(),
Augmenter(),
Resizer(min_side=int(512 * _scale),
max_side=int(640 * _scale),
logger=net_logger)
]))
dataset_val = FLIRDataset(parser.dataset_root,
set_name='val',
transform=transforms.Compose([
Normalizer(),
Resizer(min_side=int(512 * _scale),
max_side=int(640 * _scale))
]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be FLIR, COCO or csv), exiting.'
)
# Original RetinaNet code
# sampler = AspectRatioBasedSampler(dataset_train, batch_size=2, drop_last=False)
# dataloader_train = DataLoader(dataset_train, num_workers=3, collate_fn=collater, batch_sampler=sampler)
# if dataset_val is not None:
# sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=1, drop_last=False)
# dataloader_val = DataLoader(dataset_val, num_workers=3, collate_fn=collater, batch_sampler=sampler_val)
dataloader_train = DataLoader(dataset_train,
batch_size=parser.batch_size,
num_workers=parser.workers,
shuffle=True,
collate_fn=collater,
pin_memory=True)
dataloader_val = DataLoader(dataset_val,
batch_size=1,
num_workers=parser.workers,
shuffle=False,
collate_fn=collater,
pin_memory=True)
build_param = {'logger': net_logger}
if parser.resume is not None:
net_logger.info('Loading Checkpoint : {}'.format(parser.resume))
retinanet = torch.load(parser.resume)
s_b = parser.resume.rindex('_')
s_e = parser.resume.rindex('.')
start_epoch = int(parser.resume[s_b + 1:s_e]) + 1
net_logger.info('Continue on {} Epoch'.format(start_epoch))
else:
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True,
**build_param)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True,
**build_param)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True,
**build_param)
elif parser.depth == 101:
retinanet = model.resnet101(
num_classes=dataset_train.num_classes(),
pretrained=True,
**build_param)
elif parser.depth == 152:
retinanet = model.resnet152(
num_classes=dataset_train.num_classes(),
pretrained=True,
**build_param)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
start_epoch = 0
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
net_logger.info('Weight Decay : {}'.format(parser.weight_decay))
net_logger.info('Learning Rate : {}'.format(parser.lr))
# optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
optimizer = optim.Adam(retinanet.parameters(),
lr=parser.lr,
weight_decay=parser.weight_decay)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
# print('Num training images: {}'.format(len(dataset_train)))
net_logger.info('Num Training Images: {}'.format(len(dataset_train)))
for epoch_num in range(start_epoch, parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
# print(data['img'][0,:,:,:].shape)
# print(data['annot'])
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
if (iter_num % 10 == 0):
_log = 'Epoch: {} | Iter: {} | Class loss: {:1.5f} | BBox loss: {:1.5f} | Running loss: {:1.5f}'.format(
epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist))
net_logger.info(_log)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if (epoch_num + 1) % 1 == 0:
test(dataset_val, retinanet, epoch_num, parser, net_logger)
# if parser.dataset == 'coco':
# print('Evaluating dataset')
# coco_eval.evaluate_coco(dataset_val, retinanet)
# elif parser.dataset == 'csv' and parser.csv_val is not None:
# print('Evaluating dataset')
# mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
print('Learning Rate:', str(scheduler._last_lr))
torch.save(
retinanet.module,
os.path.join(
'saved', '{}_{}_{}.pt'.format(parser.dataset, network_name,
epoch_num)))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_classes', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--img_path', help='Path to file to save val images')
parser.add_argument('--model', help='Path to model (.pt) file.')
parser = parser.parse_args(args)
if parser.dataset == 'coco':
dataset_val = CocoDataset(parser.coco_path, set_name='train2017', transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
dataset_val = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes, transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=1, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=1, collate_fn=collater, batch_sampler=sampler_val)
retinanet = torch.load(parser.model)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.eval()
unnormalize = UnNormalizer()
def draw_caption(image, box, caption):
b = np.array(box).astype(int)
cv2.putText(image, caption, (b[0], b[1] - 10), cv2.FONT_HERSHEY_PLAIN, 1, (0, 0, 0), 2)
cv2.putText(image, caption, (b[0], b[1] - 10), cv2.FONT_HERSHEY_PLAIN, 1, (255, 255, 255), 1)
for idx, data in enumerate(dataloader_val):
with torch.no_grad():
st = time.time()
if torch.cuda.is_available():
scores, classification, transformed_anchors = retinanet(data['img'].cuda().float())
else:
scores, classification, transformed_anchors = retinanet(data['img'].float())
print('Elapsed time: {}'.format(time.time()-st))
idxs = np.where(scores.cpu()>0.5)
img = np.array(255 * unnormalize(data['img'][0, :, :, :])).copy()
img[img<0] = 0
img[img>255] = 255
img = np.transpose(img, (1, 2, 0))
img = cv2.cvtColor(img.astype(np.uint8), cv2.COLOR_BGR2RGB)
for j in range(idxs[0].shape[0]):
bbox = transformed_anchors[idxs[0][j], :]
x1 = int(bbox[0])
y1 = int(bbox[1])
x2 = int(bbox[2])
y2 = int(bbox[3])
label_name = dataset_val.labels[int(classification[idxs[0][j]])]
draw_caption(img, (x1, y1, x2, y2), label_name)
cv2.rectangle(img, (x1, y1), (x2, y2), color=(0, 0, 255), thickness=2)
print(label_name)
cv2.write(img_path + str(idx) + ".jpeg", img)
cv2.waitKey(0)
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.',
default='coco')
parser.add_argument(
'--coco_path',
help='Path to COCO directory',
default=
'/media/zhuzhu/ec114170-f406-444f-bee7-a3dc0a86cfa2/dataset/coco')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--use-gpu',
help='training on cpu or gpu',
action='store_false',
default=True)
parser.add_argument('--device-ids', help='GPU device ids', default=[0])
args = parser.parse_args()
# ------------------------------ Create the data loaders -----------------------------
if args.dataset == 'coco':
if args.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(args.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(args.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
sampler_train = AspectRatioBasedSampler(dataset_train,
batch_size=2,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_train)
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if args.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif args.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif args.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=False)
elif args.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif args.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
if args.use_gpu:
retinanet = nn.DataParallel(retinanet,
device_ids=args.device_ids).cuda()
# retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(args.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
nn.utils.clip_grad_norm_(retinanet.parameters(),
0.1) # 梯度的最大范数为0.1
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if args.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(retinanet.module,
'{}_retinanet_{}.pt'.format(args.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--model_path', help='Path to model', type=str)
parser = parser.parse_args(args)
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
# Create the model
retinanet = model.resnet50(num_classes=dataset_val.num_classes(),
pretrained=True)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet.load_state_dict(torch.load(parser.model_path))
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet.load_state_dict(torch.load(parser.model_path))
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = False
retinanet.eval()
retinanet.module.freeze_bn()
def draw_caption(image, box, caption):
b = np.array(box).astype(int)
cv2.putText(image, caption, (b[0], b[1] - 10), cv2.FONT_HERSHEY_PLAIN,
1, (0, 0, 0), 2)
cv2.putText(image, caption, (b[0], b[1] - 10), cv2.FONT_HERSHEY_PLAIN,
1, (255, 255, 255), 1)
for idx, data in enumerate(dataset_val):
with torch.no_grad():
st = time.time()
# run network
if torch.cuda.is_available():
scores, labels, boxes = retinanet(data['img'].permute(
2, 0, 1).cuda().float().unsqueeze(dim=0))
else:
scores, labels, boxes = retinanet(data['img'].permute(
2, 0, 1).float().unsqueeze(dim=0))
print('Elapsed time: {}'.format(time.time() - st))
idxs = np.where(scores.cpu() > 0.5)
tensor = data['img'] * np.array(
[[[0.229, 0.224, 0.225]]]) + np.array([[[0.485, 0.456, 0.406]]
])
img = tensor.mul(255).clamp(0, 255).byte().cpu().numpy()
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
for j in range(idxs[0].shape[0]):
bbox = boxes[idxs[0][j], :]
x1 = int(bbox[0])
y1 = int(bbox[1])
x2 = int(bbox[2])
y2 = int(bbox[3])
label_name = dataset_val.labels[int(labels[idxs[0][j]])]
draw_caption(img, (x1, y1, x2, y2), label_name)
cv2.rectangle(img, (x1, y1), (x2, y2),
color=(0, 0, 255),
thickness=2)
print(label_name)
cv2.imshow('img', img)
cv2.waitKey(0)
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument(
'--dataset', help='Dataset type, must be one of csv or coco.') #数据集类型
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50) #选择与训练模型
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
#决定图片数据集的顺序和batch_size,返回的是图片的分组
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=2,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
#多GPU运行
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
#collections:模块实现了特定目标的容器，以提供Python标准内建容器 dict、list、set、tuple 的替代选择
#collections.deque:返回双向队列对象，最长长度为500
loss_hist = collections.deque(maxlen=500)
# model.train() ：启用 BatchNormalization 和 Dropout
# model.eval() ：不启用 BatchNormalization 和 Dropout
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
#反向传播
loss.backward()
#梯度裁剪，梯度小于/大于阈值时，更新的梯度为阈值（此处为小于0.1）
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
#更新所有的参数，一旦梯度被如backward()之类的函数计算好后，我们就可以调用这个函数
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
#optimizer.step()通常用在每个mini-batch之中，而scheduler.step()通常用在epoch里面
#有用了optimizer.step()，模型才会更新，而scheduler.step()是对lr进行调整。
scheduler.step(np.mean(epoch_loss))
torch.save(retinanet.module,
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.',
default='csv')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)',
default='data/train_retinanet.csv')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)',
default='data/class_retinanet.csv')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)',
default='data/val_retinanet.csv')
parser.add_argument('--model_path',
default='coco_resnet_50_map_0_335_state_dict.pt',
help='Path to file containing pretrained retinanet')
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs_detection',
help='Number of epochs for detection',
type=int,
default=50)
parser.add_argument('--epochs_classification',
help='Number of epochs for classification',
type=int,
default=50)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=1,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if parser.model_path is not None:
print('loading ', parser.model_path)
if 'coco' in parser.model_path:
retinanet.load_state_dict(torch.load(parser.model_path),
strict=False)
else:
retinanet = torch.load(parser.model_path)
print('Pretrained model loaded!')
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
#Here training the detection
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=4,
verbose=True)
loss_hist = collections.deque(maxlen=500)
loss_style_classif = nn.CrossEntropyLoss()
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
mAP_list = []
mAPbest = 0
for epoch_num in range(parser.epochs_detection):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
[classification_loss, regression_loss], style = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
[classification_loss, regression_loss
], style = retinanet([data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
if torch.cuda.is_available():
style_loss = loss_style_classif(
style,
torch.tensor(data['style']).cuda())
else:
style_loss = loss_style_classif(
style, torch.tensor(data['style']))
loss = classification_loss + regression_loss + style_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.4f} | Regression loss: {:1.4f} | Style loss: {:1.4f} | Running loss: {:1.4f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), float(style_loss),
np.mean(loss_hist)))
del classification_loss
del regression_loss
del style_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAPclasses, mAP, accu = csv_eval.evaluate(dataset_val, retinanet)
mAP_list.append(mAP)
print('mAP_list', mAP_list)
if mAP > mAPbest:
print('Saving best checkpoint')
torch.save(retinanet, 'model_best.pt')
mAPbest = mAP
scheduler.step(np.mean(epoch_loss))
torch.save(retinanet.module,
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
# Here we aggregate all the data to don't have to appy the Retinanet during training.
retinanet.load_state_dict(torch.load('model_best.pt').state_dict())
List_feature = []
List_target = []
retinanet.training = False
retinanet.eval()
retinanet.module.style_inference = True
retinanet.module.freeze_bn()
epoch_loss = []
with torch.no_grad():
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
_, _, feature_vec = retinanet(data['img'].cuda().float())
else:
_, _, feature_vec = retinanet(data['img'].float())
List_feature.append(torch.squeeze(feature_vec).cpu())
List_target.append(data['style'][0])
except Exception as e:
print(e)
continue
print('END of preparation of the data for classification of style')
# Here begins Style training. Need to set to style_train. They are using the same loader, as it was expected to train both at the same time.
batch_size_classification = 64
dataloader_train_style = torch.utils.data.DataLoader(
StyleDataset(List_feature, List_target),
batch_size=batch_size_classification)
retinanet.load_state_dict(torch.load('model_best.pt').state_dict())
# Here training the detection
retinanet.module.style_inference = False
retinanet.module.style_train(True)
retinanet.training = True
retinanet.train()
optimizer = optim.Adam(
retinanet.module.styleClassificationModel.parameters(),
lr=5e-3,
weight_decay=1e-3)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
mode='max',
patience=4,
verbose=True)
loss_hist = collections.deque(maxlen=500)
loss_style_classif = nn.CrossEntropyLoss()
retinanet.train()
retinanet.module.freeze_bn()
criterion = nn.CrossEntropyLoss()
accu_list = []
accubest = 0
for epoch_num in range(parser.epochs_classification):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
total = 0
correct = 0
for iter_num, data in enumerate(dataloader_train_style):
try:
optimizer.zero_grad()
inputs, targets = data
if torch.cuda.is_available():
inputs, targets = inputs.cuda(), targets.cuda()
outputs = retinanet.module.styleClassificationModel(
inputs, 0, 0, 0, True)
loss = criterion(outputs, targets)
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
total += targets.size(0)
_, predicted = torch.max(outputs.data, 1)
correct += predicted.eq(targets.data).cpu().sum()
print(
'| Epoch [%3d/%3d] Iter[%3d/%3d]\t\tLoss: %.4f Acc@1: %.3f%%'
%
(epoch_num, parser.epochs_classification, iter_num + 1,
(len(dataloader_train_style) // batch_size_classification)
+ 1, loss.item(), 100. * correct / total))
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAPclasses, mAP, accu = csv_eval.evaluate(dataset_val, retinanet)
accu_list.append(accu)
print('mAP_list', mAP_list, 'accu_list', accu_list)
if accu > accubest:
print('Saving best checkpoint')
torch.save(retinanet.module, 'model_best_classif.pt')
accubest = accu
scheduler.step(accu)
torch.save(retinanet.module,
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet.module, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--model', help='Path to model (.pt) file.')
parser.add_argument('--finetune',
help='if load trained retina model',
type=bool,
default=False)
parser.add_argument('--gpu', help='', type=bool, default=False)
parser.add_argument('--batch_size', help='', type=int, default=2)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
#sampler = AspectRatioBasedSampler(dataset_train, batch_size=2, drop_last=False)
sampler = AspectRatioBasedSampler(dataset_train,
parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
'''
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
'''
use_gpu = parser.gpu
#import pdb
#pdb.set_trace()
#读coco预训练模型
retinanet = model.resnet50(num_classes=80, pretrained=True)
retinanet.load_state_dict(torch.load(parser.model))
for param in retinanet.parameters():
param.requires_grad = False
retinanet.regressionModel = model.RegressionModel(256)
retinanet.classificationModel = model.ClassificationModel(
256, num_classes=dataset_train.num_classes())
prior = 0.01
retinanet.classificationModel.output.weight.data.fill_(0)
retinanet.classificationModel.output.bias.data.fill_(-math.log(
(1.0 - prior) / prior))
retinanet.regressionModel.output.weight.data.fill_(0)
retinanet.regressionModel.output.bias.data.fill_(0)
# for m in retinanet.classificationModel.modules():
# if isinstance(m, nn.Conv2d):
# n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
# m.weight.data.normal_(0, math.sqrt(2. / n))
# elif isinstance(m, nn.BatchNorm2d):
# m.weight.data.fill_(1)
# m.bias.data.zero_()
# for m in retinanet.regressionModel.modules():
# if isinstance(m, nn.Conv2d):
# n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
# m.weight.data.normal_(0, math.sqrt(2. / n))
# elif isinstance(m, nn.BatchNorm2d):
# m.weight.data.fill_(1)
# m.bias.data.zero_()
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if use_gpu and torch.cuda.is_available():
#retinanet.load_state_dict(torch.load(parser.model))
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
#retinanet.load_state_dict(torch.load(parser.model))
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(
[{
'params': retinanet.module.regressionModel.parameters()
}, {
'params': retinanet.module.classificationModel.parameters()
}], 1e-6)
#optimizer = optim.Adam(retinanet.parameters(), lr=1e-6)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
#import pdb
#pdb.set_trace()
optimizer.zero_grad()
if use_gpu and torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot'].cuda()])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
if epoch_num % 5 == 0:
torch.save(
retinanet.module,
'{}_freezinetune_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument("--load_model_path",
type=str,
default=None,
help="Path to model (.pt) file.")
parser.add_argument('--dataset_type',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument('--backbone',
help='Backbone choice: [ResNet, ResNeXt]',
type=str,
default='ResNet')
parser.add_argument(
'--depth',
help='ResNet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument("--batch_size",
type=int,
default=2,
help="size of the batches")
parser.add_argument("--lr",
type=float,
default=1e-5,
help="adam: learning rate")
parser = parser.parse_args(args)
results_dir = "results"
save_images_dir = os.path.join(results_dir, "images")
save_models_dir = os.path.join(results_dir, "saved_models")
os.makedirs(results_dir, exist_ok=True)
os.makedirs(save_images_dir, exist_ok=True)
os.makedirs(save_models_dir, exist_ok=True)
# Get today datetime
today = datetime.date.today()
today = "%d%02d%02d" % (today.year, today.month, today.day)
# Get current timme
now = time.strftime("%H%M%S")
# Backbone name
backbone_name = parser.backbone + str(parser.depth)
# DataSet name
dataset_path = ''
# Create the data loaders
if parser.dataset_type == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
# dataset_train = CocoDataset(parser.coco_path, set_name='train2017',
# transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
# dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
# transform=transforms.Compose([Normalizer(), Resizer()]))
dataset_train = CocoDataset(
parser.coco_path,
set_name='train',
# transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
transform=transforms.Compose(
[Normalizer(), AugmenterWithImgaug(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
dataset_path = parser.coco_path
elif parser.dataset_type == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
dataset_path = parser.csv_train
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Retrain the model
if parser.load_model_path is not None:
# Load pretrained models
print("\nLoading model from: [%s]" % parser.load_model_path)
retinanet = torch.load(parser.load_model_path)
print("\nStart retrain...")
# Create the model
else:
print("\nStart train...")
if parser.backbone == 'ResNet':
if parser.depth == 18:
retinanet = model.resnet18(
num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(
num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(
num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(
num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(
num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152'
)
elif parser.backbone == 'ResNeXt':
if parser.depth == 50:
retinanet = model.resnext50_32x4d(
num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnext101_32x8d(
num_classes=dataset_train.num_classes(), pretrained=True)
pass
else:
raise ValueError(
"Unsupported model depth, must be one of 50, 101")
else:
raise ValueError("Choice a backbone, [ResNet, ResNeXt]")
# Get dataset name
dataset_name = os.path.split(dataset_path)[-1]
# Checkpoint name
save_ckpt_name = r"%s_%s-%s-RetinaNet-backbone(%s)-ep(%d)-bs(%d)-lr(%s)" \
% (today, now, dataset_name, backbone_name, parser.epochs, parser.batch_size, parser.lr)
os.makedirs(os.path.join(save_images_dir, "%s" % save_ckpt_name),
exist_ok=True)
os.makedirs(os.path.join(save_models_dir, "%s" % save_ckpt_name),
exist_ok=True)
tb_log_path = os.path.join("tf_log", save_ckpt_name)
tb_writer = SummaryWriter(os.path.join(results_dir, tb_log_path))
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=parser.lr)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
val_loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
epoch_prev_time = time.time()
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
total_classification_loss = 0.0
total_regression_loss = 0.0
total_running_loss = 0.0
total_val_classification_loss = 0.0
total_val_regression_loss = 0.0
total_val_running_loss = 0.0
batch_prev_time = time.time()
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
# sum the loss for tensorboard at this batch
total_regression_loss += regression_loss
total_classification_loss += classification_loss
total_running_loss += loss.item()
# log = 'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
# epoch_num, iter_num, float(classification_loss), float(regression_loss), np.mean(loss_hist))
# Determine approximate time left
data_done = iter_num
data_left = len(dataloader_train) - data_done
batch_time_left = datetime.timedelta(
seconds=data_left * (time.time() - batch_prev_time))
batch_time_left = chop_microseconds(batch_time_left)
batches_done = epoch_num * len(dataloader_train) + iter_num
batches_left = parser.epochs * len(
dataloader_train) - batches_done
total_time_left = datetime.timedelta(
seconds=batches_left * (time.time() - epoch_prev_time))
total_time_left = chop_microseconds(total_time_left)
batch_prev_time = time.time()
epoch_prev_time = time.time()
# Print training step log
prefix_log = '[Epoch: {}/{}] | [Batch: {}/{}]'.format(
epoch_num + 1, parser.epochs, iter_num + 1,
len(dataloader_train))
suffix_log = '[Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}] ETA: {} / {}'.format(
float(classification_loss), float(regression_loss),
np.mean(loss_hist), batch_time_left, total_time_left)
printProgressBar(iteration=iter_num + 1,
total=len(dataloader_train),
prefix=prefix_log,
suffix=suffix_log)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
# Validation
with torch.no_grad():
val_batch_prev_time = time.time()
for iter_num, data in enumerate(dataloader_val):
try:
val_classification_loss, val_regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
val_classification_loss = val_classification_loss.mean()
val_regression_loss = val_regression_loss.mean()
val_loss = val_classification_loss + val_regression_loss
if bool(val_loss == 0):
continue
val_loss_hist.append(float(val_loss))
# sum the loss for tensorboard at this batch
total_val_regression_loss += val_regression_loss
total_val_classification_loss += val_classification_loss
total_val_running_loss += val_loss.item()
# Determine approximate time left
data_done = iter_num
data_left = len(dataloader_val) - data_done
val_batch_time_left = datetime.timedelta(
seconds=data_left *
(time.time() - val_batch_prev_time))
val_batch_time_left = chop_microseconds(
val_batch_time_left)
batches_done = epoch_num * len(dataloader_val) + (
epoch_num + 1) * len(dataloader_train) + iter_num
batches_left = parser.epochs * (len(
dataloader_train) + len(dataloader_val)) - batches_done
total_time_left = datetime.timedelta(
seconds=batches_left * (time.time() - epoch_prev_time))
total_time_left = chop_microseconds(total_time_left)
val_batch_prev_time = time.time()
epoch_prev_time = time.time()
# Print training step log
prefix_log = 'Validation: [Epoch: {}/{}] | [Batch: {}/{}]'.format(
epoch_num + 1, parser.epochs, iter_num + 1,
len(dataloader_val))
suffix_log = '[Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}] ETA: {} / {}'.format(
float(val_classification_loss),
float(val_regression_loss), np.mean(val_loss_hist),
val_batch_time_left, total_time_left)
printProgressBar(iteration=iter_num + 1,
total=len(dataloader_val),
prefix=prefix_log,
suffix=suffix_log)
del val_classification_loss
del val_regression_loss
except Exception as e:
print(e)
continue
# Evaluate AP
if parser.dataset_type == 'coco':
print('Evaluating dataset')
# coco_eval.evaluate_coco(dataset_val, retinanet)
coco_eval.evaluate_coco_and_save_image(
dataset_val, retinanet,
os.path.join(save_images_dir, save_ckpt_name), epoch_num + 1)
elif parser.dataset_type == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
# calculate loss average
average_classification_loss = total_classification_loss / len(
dataloader_train)
average_regression_loss = total_regression_loss / len(dataloader_train)
average_running_loss = total_running_loss / len(dataloader_train)
# TensorBoard
tb_writer.add_scalar(tag='Classification Loss',
scalar_value=average_classification_loss,
global_step=epoch_num + 1)
tb_writer.add_scalar(tag='Regression Loss',
scalar_value=average_regression_loss,
global_step=epoch_num + 1)
tb_writer.add_scalar(tag='Total Loss',
scalar_value=average_running_loss,
global_step=epoch_num + 1)
# Save model
print("\nSave model to [%s] at %d epoch\n" %
(save_ckpt_name, epoch_num + 1))
checkpoint_path = os.path.join(
save_models_dir, "%s/RetinaNet_backbone(%s)_%d.pt" %
(save_ckpt_name, backbone_name, epoch_num + 1))
torch.save(retinanet.module, checkpoint_path)
# torch.save(retinanet.module, '{}_retinanet_{}.pt'.format(parser.dataset_type, epoch_num + 1))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', default='csv', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_train', default='dataset/pascal_train.csv', help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes', default='dataset/classes.csv', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', default='dataset/pascal_val.csv', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--depth', help='Resnet depth, must be one of 18, 34, 50, 101, 152', type=int, default=50)
parser.add_argument('--epochs', help='Number of epochs', type=int, default=100)
parser.add_argument('--weights_folder', help='path to save weight', type=str, required=True)
parser = parser.parse_args(args)
if not os.path.exists(parser.weights_folder):
os.makedirs(parser.weights_folder)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path, set_name='train2017',
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError('Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train, batch_size=5, drop_last=False)
dataloader_train = DataLoader(dataset_train, num_workers=4, collate_fn=collater, batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=8, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=4, collate_fn=collater, batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience=3, verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
# import ipdb; ipdb.set_trace()
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
total_loss = 0
total_regression_loss = 0
total_classification_loss = 0
with tqdm(dataloader_train, unit="batch") as tepoch:
for data in tepoch:
# for iter_num, data in tepoch:#enumerate(dataloader_train):
tepoch.set_description(f"Epoch {epoch_num}")
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet([data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
total_loss = total_loss + loss
total_regression_loss = total_regression_loss + regression_loss
total_classification_loss = total_classification_loss + classification_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
# print(
# 'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
# epoch_num, iter_num, float(classification_loss), float(regression_loss), np.mean(loss_hist)))
tepoch.set_postfix(cls_loss="{:1.5f}".format(classification_loss), reg_loss="{:1.5f}".format(regression_loss))
time.sleep(0.1)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
tb.add_scalar('Training loss', total_loss, epoch_num)
tb.add_scalar('Training regression loss', total_regression_loss, epoch_num)
tb.add_scalar('Training accuracy loss', total_classification_loss, epoch_num)
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(retinanet.module, '{}/{}_retinanet_{}.pt'.format(parser.weights_folder,parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, '{}/model_final.pt'.format(parser.weights_folder))
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--finetune',
help='if load trained retina model',
type=bool,
default=False)
parser.add_argument('--gpu', help='', type=bool, default=False)
parser.add_argument('--batch_size', help='', type=int, default=2)
parser.add_argument('--c',
help='continue with formal model',
type=bool,
default=False)
parser.add_argument('--model', help='model path')
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
#sampler = AspectRatioBasedSampler(dataset_train, batch_size=2, drop_last=False)
sampler = AspectRatioBasedSampler(dataset_train,
parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=16,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=8,
collate_fn=collater,
batch_sampler=sampler_val)
epochpassed = 0
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
if parser.c:
retinanet = torch.load(parser.model)
#import pdb
#pdb.set_trace()
epochpassed = int(parser.model.split('.')[1].split('_')[-1])
use_gpu = parser.gpu
#torch.cuda.set_device(5)
#import pdb
#pdb.set_trace()
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if use_gpu and torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5) #original:1e-5
#optimizer =optim.SGD(retinanet.parameters(), lr=0.01,weight_decay=0.0001, momentum=0.9)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
writer = SummaryWriter()
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
epoch_classification_loss = []
epoch_regression_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
#import pdb
#pdb.set_trace()
optimizer.zero_grad()
if use_gpu and torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot'].cuda()])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
epoch_classification_loss.append(float(classification_loss))
epoch_regression_loss.append(float(regression_loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Epoch loss: {:1.5f}\r'
.format(epoch_num + epochpassed, iter_num,
float(classification_loss), float(regression_loss),
np.mean(loss_hist)),
end='')
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
print(
'Epoch: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Epoch loss: {:1.5f}'
.format(epoch_num + epochpassed,
np.mean(epoch_classification_loss),
np.mean(epoch_regression_loss), np.mean(epoch_loss)))
writer.add_scalar('lossrecord/regressionloss',
np.mean(epoch_regression_loss),
epoch_num + epochpassed)
writer.add_scalar('lossrecord/classificationloss',
np.mean(epoch_regression_loss),
epoch_num + epochpassed)
writer.add_scalar('lossrecord/epochloss', np.mean(epoch_loss),
epoch_num + epochpassed)
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
if epoch_num % 10 == 0:
torch.save(
retinanet.module,
'./models/{}_retinanet{}_highResolution4fold_{}.pt'.format(
parser.dataset, parser.depth, epoch_num + epochpassed))
#retinanet.eval()
torch.save(
retinanet.module,
'./models/{}_retinanet{}_highResolution4fold_{}.pt'.format(
parser.dataset, parser.depth, parser.epochs + epochpassed))
writer.close()
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.',
default='csv')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--batch_size', help='Batch size', type=int, default=2)
parser.add_argument('--num_workers',
help='Number of workers',
type=int,
default=4)
parser.add_argument('--models_out',
help='The directory to save models',
type=str)
parser = parser.parse_args(args)
if not os.path.exists(parser.models_out):
os.makedirs(parser.models_out)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=parser.num_workers,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=parser.num_workers,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
writer = SummaryWriter(log_dir="tensor_log/" + parser.models_out)
global_steps = 0
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
running_loss = np.mean(loss_hist)
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), running_loss))
global_steps += 1
writer.add_scalar("Loss/Classification",
float(classification_loss), global_steps)
writer.add_scalar("Loss/Regression", float(regression_loss),
global_steps)
writer.add_scalar("Loss/Running", running_loss, global_steps)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
#for k, v in mAP.items():
# writer.add_scalar("Accuracy/map_{}".format(k), v, epoch_num)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module,
os.path.join(
parser.models_out,
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num)))
retinanet.eval()
torch.save(retinanet, os.path.join(parser.models_out, 'model_final.pt'))
def main(args=None):
parser = argparse.ArgumentParser(
description="Simple training script for training a RetinaNet network.")
parser.add_argument("--dataset",
help="Dataset type, must be one of csv or coco.")
parser.add_argument("--coco_path", help="Path to COCO directory")
parser.add_argument(
"--csv_train",
help="Path to file containing training annotations (see readme)")
parser.add_argument("--csv_classes",
help="Path to file containing class list (see readme)")
parser.add_argument(
"--csv_val",
help=
"Path to file containing validation annotations (optional, see readme)",
)
parser.add_argument(
"--depth",
help="Resnet depth, must be one of 18, 34, 50, 101, 152",
type=int,
default=50,
)
parser.add_argument("--batch_size", help="Batch size", type=int, default=2)
parser.add_argument("--epochs",
help="Number of epochs",
type=int,
default=100)
parser.add_argument("--workers",
help="Number of workers of dataleader",
type=int,
default=4)
parser = parser.parse_args(args)
writer = SummaryWriter("logs")
# Create the data loaders
if parser.dataset == "coco":
if parser.coco_path is None:
raise ValueError("Must provide --coco_path when training on COCO,")
dataset_train = CocoDataset(
parser.coco_path,
set_name="train2017",
transform=transforms.Compose(
[Normalizer(), Augmenter(),
Resizer()]),
)
dataset_val = CocoDataset(
parser.coco_path,
set_name="val2017",
transform=transforms.Compose([Normalizer(),
Resizer()]),
)
elif parser.dataset == "csv":
if parser.csv_train is None:
raise ValueError("Must provide --csv_train when training on COCO,")
if parser.csv_classes is None:
raise ValueError(
"Must provide --csv_classes when training on COCO,")
dataset_train = CSVDataset(
train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(), Augmenter(),
Resizer()]),
)
if parser.csv_val is None:
dataset_val = None
print("No validation annotations provided.")
else:
dataset_val = CSVDataset(
train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(),
Resizer()]),
)
else:
raise ValueError(
"Dataset type not understood (must be csv or coco), exiting.")
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(
dataset_train,
num_workers=parser.workers,
collate_fn=collater,
batch_sampler=sampler,
)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=parser.workers,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
"Unsupported model depth, must be one of 18, 34, 50, 101, 152")
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=10,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print("Num training images: {}".format(len(dataset_train)))
global_step = 0
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
global_step = iter_num + epoch_num * len(dataloader_train)
try:
optimizer.zero_grad()
classification_loss, regression_loss = retinanet(
[data["img"].cuda().float(), data["annot"]])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
if iter_num % 10 == 0:
print(
"Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}"
.format(
epoch_num,
iter_num,
float(classification_loss),
float(regression_loss),
np.mean(loss_hist),
))
writer.add_scalars(
"training",
{
"loss": loss,
"loss_cls": classification_loss,
"loss_reg": regression_loss,
},
global_step,
)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == "coco":
print("Evaluating dataset")
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == "csv" and parser.csv_val is not None:
print("Evaluating dataset")
mAP = csv_eval.evaluate(dataset_val, retinanet)
valid_mAP = [x[0] for x in mAP.values() if x[1] > 0]
mmAP = sum(valid_mAP) / len(mAP)
writer.add_scalars("validation", {"mmAP": mmAP}, global_step)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module,
"checkpoints/{}_retinanet_{}.pt".format(parser.dataset, epoch_num),
)
retinanet.eval()
torch.save(retinanet, "checkpoints/odel_final.pt")
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152, 5032, 10132',
type=int,
default=10148)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=200)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=2,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 5032:
retinanet = model.resnext50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 10132:
retinanet = model.resnext101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 10148:
retinanet = model_SE.SEresnext101(
num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(),
lr=1e-5) #change_weight_decay
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
total_classification_loss = 0.0
total_regression_loss = 0.0
epoch_number = 0
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
#############################
# total_classification_loss += classification_loss
# total_regression_loss += regression_loss
# epoch_number = epoch_num
fp = open(output_path + "clas_reg_loss.txt", "a")
fp.write(
str(epoch_num) + ',' + str(float(classification_loss)) +
',' + str(float(regression_loss)) + ',' +
str(np.mean(loss_hist)) + '\n')
# writer.add_scalar('Classification_loss', float(classification_loss), epoch_num)
# writer.add_scalar('Regression_loss', float(regression_loss), epoch_num)
# writer.flush()
#############################
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
#############################
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module, output_path +
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, output_path + 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_train', help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--iou',default='05')
parser.add_argument('--depth', help='Resnet depth, must be one of 18, 34, 50, 101, 152', type=int, default=50)
parser.add_argument('--epochs', help='Number of epochs', type=int, default=100)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path, set_name='train2017',
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError('Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
val_dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train, batch_size=8, drop_last=False)
dataloader_train = DataLoader(dataset_train, num_workers=3, collate_fn=collater, batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=8, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=3, collate_fn=collater, batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=5e-5)
lr_scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience=3, verbose=True)
multistep_scheduler = optim.lr_scheduler.MultiStepLR(optimizer, milestones=[5,8,11,20], gamma=0.2)
loss_hist = collections.deque(maxlen=500)
val_loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
val_epoch_loss=[]
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet([data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Train: Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f} | Epoch loss: {:1.5f} '.format(
epoch_num, iter_num, float(classification_loss), float(regression_loss), np.mean(loss_hist),epoch_loss[-1]))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
for iter_num, data in enumerate(dataloader_val):
try:
#optimizer.zero_grad()
#retinanet.eval()
with torch.no_grad():
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet((data['img'].cuda().float(), data['annot']))
else:
classification_loss, regression_loss = retinanet((data['img'].float(), data['annot']))
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
#loss.backward()
#torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
#optimizer.step()
val_loss_hist.append(float(loss))
val_epoch_loss.append(float(loss))
print(
'Val: Epoch: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f} | Epoch loss: {:1.5f} '.format(
epoch_num, float(classification_loss), float(regression_loss), np.mean(val_loss_hist),val_epoch_loss[-1]))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
#mAP_train = csv_eval.evaluate(val_dataset_train,retinanet,iou_threshold=float(parser.iou)/10)
mAP_val = csv_eval.evaluate(dataset_val, retinanet,iou_threshold=float(parser.iou)/10)
#writer.add_scalar('train_mAP_Questions',mAP_train[0][0],epoch_num)
writer.add_scalar('val_mAP_Questions', mAP_val[0][0], epoch_num)
writer.add_scalar('val_loss',np.mean(val_epoch_loss),epoch_num)
writer.add_scalar('train_loss',np.mean(epoch_loss),epoch_num)
lr_scheduler.step(np.mean(epoch_loss))
#one_scheduler.step()
multistep_scheduler.step()
torch.save(retinanet.module, '{}_retinanet_{}.pt'.format(parser.iou, epoch_num))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument('--model_save_path',
help='Path to save model',
type=str)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=8,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
# add draw tensorboard code
writer = SummaryWriter(log_dir='./logs/416*416/', flush_secs=60)
# if Cuda:
# graph_inputs = torch.from_numpy(np.random.rand(1, 3, input_shape[0], input_shape[1])).type(
# torch.FloatTensor).cuda()
# else:
# graph_inputs = torch.from_numpy(np.random.rand(1, 3, input_shape[0], input_shape[1])).type(torch.FloatTensor)
# writer.add_graph(model, (graph_inputs,))
# add gap save model count variable
n = 0
for epoch_num in range(parser.epochs):
n += 1
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
### begin calculate train loss
for iter_num, data in enumerate(dataloader_train):
# try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
# except Exception as e:
# print(e)
# continue
### begin calculate valid loss
for iter_num, data in enumerate(dataloader_val):
# try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss_hist.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Valid-Classification loss: {:1.5f} | Valid-Regression loss: {:1.5f} | Running Valid loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
print('Epoch: {} | mAP: {:.3f}'.format(epoch_num, float(mAP)))
scheduler.step(np.mean(epoch_loss))
if n % 10 == 0:
torch.save(
retinanet.module, parser.model_save_path +
'/' + '{}_retinanet_{}_{:.3f}.pt'.format(
parser.dataset, epoch_num, mAP))
retinanet.eval()
torch.save(retinanet, parser.model_save_path + '/' + 'model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(
description="Simple training script for training a RetinaNet network."
)
parser.add_argument("--dataset", help="Dataset type, must be one of csv or coco.")
parser.add_argument("--model", default=None, help="Path to trained model")
parser.add_argument("--coco_path", help="Path to COCO directory")
parser.add_argument(
"--csv_train", help="Path to file containing training annotations (see readme)"
)
parser.add_argument(
"--csv_classes", help="Path to file containing class list (see readme)"
)
parser.add_argument(
"--csv_val",
help="Path to file containing validation annotations (optional, see readme)",
)
parser.add_argument(
"--depth",
help="Resnet depth, must be one of 18, 34, 50, 101, 152",
type=int,
default=50,
)
parser.add_argument("--epochs", help="Number of epochs", type=int, default=100)
parser.add_argument(
"--result_dir",
default="results",
help="Path to store training results",
type=str,
)
parser.add_argument(
"--batch_num", default=8, help="Number of samples in a batch", type=int
)
parser = parser.parse_args(args)
print(parser)
# parameters
BATCH_SIZE = parser.batch_num
IMAGE_MIN_SIDE = 1440
IMAGE_MAX_SIDE = 2560
# Create the data loaders
if parser.dataset == "coco":
if parser.coco_path is None:
raise ValueError("Must provide --coco_path when training on COCO,")
# TODO: parameterize arguments for Resizer, and other transform functions
# resizer: min_side=608, max_side=1024
dataset_train = CocoDataset(
parser.coco_path,
# set_name="train2017",
set_name="train_images_full",
transform=transforms.Compose(
[Normalizer(), Augmenter(), Resizer(passthrough=True),]
),
)
dataset_val = CocoDataset(
parser.coco_path,
# set_name="val2017",
set_name="val_images_full",
transform=transforms.Compose([Normalizer(), Resizer(passthrough=True),]),
)
elif parser.dataset == "csv":
if parser.csv_train is None:
raise ValueError("Must provide --csv_train when training on COCO,")
if parser.csv_classes is None:
raise ValueError("Must provide --csv_classes when training on COCO,")
dataset_train = CSVDataset(
train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]),
)
if parser.csv_val is None:
dataset_val = None
print("No validation annotations provided.")
else:
dataset_val = CSVDataset(
train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]),
)
else:
raise ValueError("Dataset type not understood (must be csv or coco), exiting.")
sampler = AspectRatioBasedSampler(
dataset_train, batch_size=BATCH_SIZE, drop_last=False
)
dataloader_train = DataLoader(
dataset_train, num_workers=16, collate_fn=collater, batch_sampler=sampler
)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(
dataset_val, batch_size=BATCH_SIZE, drop_last=False
)
dataloader_val = DataLoader(
dataset_val, num_workers=16, collate_fn=collater, batch_sampler=sampler_val
)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(
num_classes=dataset_train.num_classes(), pretrained=True
)
elif parser.depth == 34:
retinanet = model.resnet34(
num_classes=dataset_train.num_classes(), pretrained=True
)
elif parser.depth == 50:
retinanet = model.resnet50(
num_classes=dataset_train.num_classes(), pretrained=True
)
elif parser.depth == 101:
retinanet = model.resnet101(
num_classes=dataset_train.num_classes(), pretrained=True
)
elif parser.depth == 152:
retinanet = model.resnet152(
num_classes=dataset_train.num_classes(), pretrained=True
)
else:
raise ValueError("Unsupported model depth, must be one of 18, 34, 50, 101, 152")
if parser.model:
retinanet = torch.load(parser.model)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-4)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(
optimizer, patience=3, verbose=True
)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print("Num training images: {}".format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
p_bar = tqdm(dataloader_train)
for iter_num, data in enumerate(p_bar):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data["img"].cuda().float(), data["annot"]]
)
else:
classification_loss, regression_loss = retinanet(
[data["img"].float(), data["annot"]]
)
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
mean_loss = np.mean(loss_hist)
p_bar.set_description(
f"Epoch: {epoch_num} | Iteration: {iter_num} | "
f"Class loss: {float(classification_loss.item()):.5f} | "
f"Regr loss: {float(regression_loss.item()):.5f} | "
f"Running loss: {mean_loss:.5f}"
)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == "coco":
print("Evaluating dataset")
coco_eval.evaluate_coco(
dataset_val, retinanet, result_dir=parser.result_dir
)
elif parser.dataset == "csv" and parser.csv_val is not None:
print("Evaluating dataset")
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
# TODO: Fix string formating mix (adopt homogeneous format)
torch.save(
retinanet.module,
f"{parser.result_dir}/"
+ "{}_retinanet_{}.pt".format(parser.dataset, epoch_num),
)
retinanet.eval()
torch.save(retinanet, "model_final.pt")
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument('--exp_name',
help='Path to folder for saving the model and log',
type=str)
parser.add_argument('--output_folder',
help='Path to folder for saving all the experiments',
type=str)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100) # 100
parser.add_argument('--batch_size', help='Batch size', type=int, default=2)
parser.add_argument('--lr',
help='Number of epochs',
type=float,
default=1e-5)
parser.add_argument('--caption',
help='Any thing in particular about the experiment',
type=str)
parser.add_argument('--server',
help='seerver name',
type=str,
default='ultron')
parser.add_argument('--detector',
help='detection algo',
type=str,
default='RetinaNet')
parser.add_argument('--arch', help='model architecture', type=str)
parser.add_argument('--pretrain', default=False, action='store_true')
parser.add_argument('--freeze_batchnorm',
default=False,
action='store_true')
parser = parser.parse_args(args)
output_folder_path = os.path.join(parser.output_folder, parser.exp_name)
if not os.path.exists(output_folder_path):
os.makedirs(output_folder_path)
PARAMS = {
'dataset': parser.dataset,
'exp_name': parser.exp_name,
'depth': parser.depth,
'epochs': parser.epochs,
'batch_size': parser.batch_size,
'lr': parser.lr,
'caption': parser.caption,
'server': parser.server,
'arch': parser.arch,
'pretrain': parser.pretrain,
'freeze_batchorm': parser.freeze_batchnorm
}
exp = neptune.create_experiment(
name=parser.exp_name,
params=PARAMS,
tags=[parser.arch, parser.detector, parser.dataset, parser.server])
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=parser.batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=3,
collate_fn=collater,
batch_sampler=sampler_val)
# Create the model
if parser.depth == 18 and parser.arch == 'Resnet':
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.depth == 10 and parser.arch == 'Resnet':
retinanet = model.resnet10(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.depth == 18 and parser.arch == 'BiRealNet18':
checkpoint_path = None
if parser.pretrain:
checkpoint_path = '/media/Rozhok/Bi-Real-net/pytorch_implementation/BiReal18_34/models/imagenet_baseline/checkpoint.pth.tar'
retinanet = birealnet18(checkpoint_path,
num_classes=dataset_train.num_classes())
elif parser.depth == 34 and parser.arch == 'Resnet':
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.depth == 50 and parser.arch == 'Resnet':
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.depth == 101 and parser.arch == 'Resnet':
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrain)
elif parser.arch == 'ofa':
print("Model is ResNet50D.")
bn_momentum = 0.1
bn_eps = 1e-5
retinanet = ResNet50D(
n_classes=dataset_train.num_classes(),
bn_param=(bn_momentum, bn_eps),
dropout_rate=0,
width_mult=1.0,
depth_param=3,
expand_ratio=0.35,
)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
print(retinanet)
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=parser.lr)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
if parser.freeze_batchnorm:
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
exp.log_metric('Current lr', float(optimizer.param_groups[0]['lr']))
exp.log_metric('Current epoch', int(epoch_num))
retinanet.train()
if parser.freeze_batchnorm:
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
exp.log_metric('Training: Classification loss',
float(classification_loss))
exp.log_metric('Training: Regression loss',
float(regression_loss))
exp.log_metric('Training: Totalloss', float(loss))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val,
retinanet,
output_folder_path,
exp=exp)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module,
os.path.join(
output_folder_path,
'{}_retinanet_{}.pt'.format(parser.dataset, epoch_num)))
retinanet.eval()
torch.save(retinanet, os.path.join(output_folder_path, 'model_final.pt'))
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
# parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--HW2_path', help='Path to HW2 directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'HW2':
if parser.HW2_path is None:
raise ValueError('Must provide --HW2_path when training on HW2,')
dataset_train = HW2Dataset(parser.HW2_path,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
#dataset_val = HW2Dataset(parser.HW2_path,
# transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
# sampler = AspectRatioBasedSampler(dataset_train, batch_size=2, drop_last=False)
dataloader_train = DataLoader(dataset_train,
batch_size=8,
num_workers=3,
collate_fn=collater)
# if dataset_val is not None:
# sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=1, drop_last=False)
# dataloader_val = DataLoader(dataset_val, num_workers=3, collate_fn=collater, batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
#retinanet.load_state_dict(torch.load('coco_resnet_50_map_0_335_state_dict.pt'))
#retinanet_state = retinanet.state_dict()
#loaded = torch.load('coco_resnet_50_map_0_335_state_dict.pt')
#pretrained = {k:v for k, v in loaded.items() if k in retinanet_state}
#retinanet_state.update(pretrained)
#retinanet.load_state_dict(retinanet_state)
retinanet = torch.load('saved_models_3/HW2_retinanet_0.pt')
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-4)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(pre_epoch, parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
# if parser.dataset == 'coco':
# print('Evaluating dataset')
# coco_eval.evaluate_coco(dataset_val, retinanet)
# elif parser.dataset == 'csv' and parser.csv_val is not None:
# print('Evaluating dataset')
# mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module,
'saved_models_3/{}_retinanet_{}.pt'.format(parser.dataset,
epoch_num))
# retinanet.eval()
torch.save(retinanet, 'saved_models_3/model_final.pt')
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.', default='show')
parser.add_argument('--coco_path', help='Path to COCO directory', default='/mnt/marathon')
parser.add_argument('--image_size', help='image size', type=int, nargs=2, default=IMAGE_SIZE)
parser.add_argument('--limit', help='limit', type=int, nargs=2, default=(0, 0))
parser.add_argument('--batch_size', help='batch size', type=int, default=BATCH_SIZE)
parser.add_argument('--num_works', help='num works', type=int, default=NUM_WORKERS)
parser.add_argument('--num_classes', help='num classes', type=int, default=3)
parser.add_argument('--merge_val', help='merge_val', type=int, default=MERGE_VAL)
parser.add_argument('--do_aug', help='do_aug', type=int, default=DO_AUG)
parser.add_argument('--lr_choice', default=LR_CHOICE, choices=['lr_scheduler', 'lr_map', 'lr_fn'], type=str)
parser.add_argument('--lr', help='lr', type=float, default=LR)
parser.add_argument("--lr_map", dest="lr_map", action=StoreDictKeyPair, default=LR_MAP)
parser.add_argument("--lr_fn", dest="lr_fn", action=StoreDictKeyPair, default=LR_FN)
parser.add_argument('--depth', help='Resnet depth, must be one of 18, 34, 50, 101, 152', type=int, default=DEPTH)
parser.add_argument('--epochs', help='Number of epochs', type=int, default=EPOCHS)
parser = parser.parse_args(args)
print('dataset:', parser.dataset)
print('depth:', parser.depth)
print('epochs:', parser.epochs)
print('image_size:', parser.image_size)
print('batch_size:', parser.batch_size)
print('num_works:', parser.num_works)
print('merge_val:', parser.merge_val)
print('do_aug:', parser.do_aug)
print('lr_choice:', parser.lr_choice)
print('lr:', parser.lr)
print('lr_map:', parser.lr_map)
print('lr_fn:', parser.lr_fn)
print('num_classes:', parser.num_classes)
print('limit:', parser.limit)
# Create the data loaders
# dataset_train, _ = torch.utils.data.random_split(dataset_train, [NUM_COCO_DATASET_TRAIN, len(dataset_train) - NUM_COCO_DATASET_TRAIN])
# dataset_val, _ = torch.utils.data.random_split(dataset_val, [NUM_COCO_DATASET_VAL, len(dataset_val) - NUM_COCO_DATASET_VAL])
transform_train = None
transform_vail = None
collate_fn = None
if parser.do_aug:
transform_train = get_augumentation('train', parser.image_size[0], parser.image_size[1])
transform_vail = get_augumentation('test', parser.image_size[0], parser.image_size[1])
collate_fn = detection_collate
else:
transform_train = transforms.Compose([
# Normalizer(),
# Augmenter(),
Resizer(*parser.image_size)])
transform_vail = transforms.Compose([
# Normalizer(),
Resizer(*parser.image_size)])
collate_fn = collater
if parser.dataset == 'h5':
dataset_train = H5CoCoDataset('{}/train_small.hdf5'.format(parser.coco_path), 'train_small')
dataset_val = H5CoCoDataset('{}/test.hdf5'.format(parser.coco_path), 'test')
else:
dataset_train = CocoDataset(parser.coco_path, set_name='train_small', do_aug=parser.do_aug,
transform=transform_train, limit_len=parser.limit[0])
dataset_val = CocoDataset(parser.coco_path, set_name='test', do_aug=parser.do_aug,
transform=transform_vail, limit_len=parser.limit[1])
# 混合val
if parser.merge_val:
dataset_train += dataset_val
print('training images: {}'.format(len(dataset_train)))
print('val images: {}'.format(len(dataset_val)))
steps_pre_epoch = len(dataset_train) // parser.batch_size
print('steps_pre_epoch:', steps_pre_epoch)
sampler = AspectRatioBasedSampler(dataset_train, batch_size=parser.batch_size, drop_last=False)
dataloader_train = DataLoader(dataset_train, batch_size=1, num_workers=parser.num_works, shuffle=False,
collate_fn=collate_fn, batch_sampler=sampler)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=parser.num_classes, pretrained=PRETRAINED)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=parser.num_classes, pretrained=PRETRAINED)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=parser.num_classes, pretrained=PRETRAINED)
elif parser.depth == 101250:
retinanet = model.resnet101with50weight(num_classes=parser.num_classes, pretrained=PRETRAINED)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=parser.num_classes, pretrained=PRETRAINED)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=parser.num_classes, pretrained=PRETRAINED)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
retinanet = retinanet.cuda()
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
if parser.lr_choice == 'lr_map':
lr_now = lr_change_map(1, 0, parser.lr_map)
elif parser.lr_choice == 'lr_fn':
lr_now = float(parser.lr_fn['LR_START'])
elif parser.lr_choice == 'lr_scheduler':
lr_now = parser.lr
# optimizer = optim.Adam(retinanet.parameters(), lr=lr_now)
optimizer = optim.AdamW(retinanet.parameters(), lr=lr_now)
# optimizer = optim.SGD(retinanet.parameters(), lr=lr_now, momentum=0.9, weight_decay=5e-4)
# optimizer = optim.SGD(retinanet.parameters(), lr=lr_now)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience=PATIENCE, factor=FACTOR, verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
iteration_loss_path = 'iteration_loss.csv'
if os.path.isfile(iteration_loss_path):
os.remove(iteration_loss_path)
epoch_loss_path = 'epoch_loss.csv'
if os.path.isfile(epoch_loss_path):
os.remove(epoch_loss_path)
eval_train_path = 'eval_train_result.csv'
if os.path.isfile(eval_train_path):
os.remove(eval_train_path)
eval_val_path = 'eval_val_result.csv'
if os.path.isfile(eval_val_path):
os.remove(eval_val_path)
USE_KAGGLE = True if os.environ.get('KAGGLE_KERNEL_RUN_TYPE', False) else False
if USE_KAGGLE:
iteration_loss_path = '/kaggle/working/' + iteration_loss_path
epoch_loss_path = '/kaggle/working/' + epoch_loss_path
eval_val_path = '/kaggle/working/' + eval_val_path
eval_train_path = '/kaggle/working/' + eval_train_path
with open(epoch_loss_path, 'a+') as epoch_loss_file, \
open(iteration_loss_path, 'a+') as iteration_loss_file, \
open(eval_train_path, 'a+') as eval_train_file, \
open(eval_val_path, 'a+') as eval_val_file:
epoch_loss_file.write('epoch_num,mean_epoch_loss\n')
iteration_loss_file.write('epoch_num,iteration,classification_loss,regression_loss,iteration_loss\n')
eval_train_file.write('epoch_num,map50\n')
eval_val_file.write('epoch_num,map50\n')
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
optimizer.zero_grad()
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
iteration_loss = np.mean(loss_hist)
print('\rEpoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
epoch_num+1, iter_num+1, float(classification_loss), float(regression_loss), iteration_loss), end=' ' * 50)
iteration_loss_file.write('{},{},{:1.5f},{:1.5f},{:1.5f}\n'.format(epoch_num+1,
epoch_num * steps_pre_epoch + (iter_num+1), float(classification_loss), float(regression_loss),
iteration_loss))
iteration_loss_file.flush()
del classification_loss
del regression_loss
mean_epoch_loss = np.mean(epoch_loss)
epoch_loss_file.write('{},{:1.5f}\n'.format(epoch_num+1, mean_epoch_loss))
epoch_loss_file.flush()
if parser.lr_choice == 'lr_map':
lr_now = lr_change_map(epoch_num+1, lr_now, parser.lr_map)
adjust_learning_rate(optimizer, lr_now)
elif parser.lr_choice == 'lr_fn':
lr_now = lrfn(epoch_num+1, parser.lr_fn)
adjust_learning_rate(optimizer, lr_now)
elif parser.lr_choice == 'lr_scheduler':
scheduler.step(mean_epoch_loss)
# if parser.dataset != 'show':
# print('Evaluating dataset_train')
# coco_eval.evaluate_coco(dataset_train, retinanet, parser.dataset, parser.do_aug, eval_train_file, epoch_num)
print('Evaluating dataset_val')
coco_eval.evaluate_coco(dataset_val, retinanet, parser.dataset, parser.do_aug, eval_val_file, epoch_num)
return parser
def main(args=None):
parser = argparse.ArgumentParser(description = 'Simple training script for training a RetinaNet network.')
parser.add_argument('--s', help = 'training session', type = int)
parser.add_argument('--bs', help = 'batch size', type = int, default = 4)
parser.add_argument('--lr', help = 'learning rate', type = float, default = 0.001)
parser.add_argument('--save_int', help = 'interval for saving model', type = int)
parser.add_argument('--dataset', help = 'Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help = 'Path to COCO directory')
parser.add_argument('--csv_train', help = 'Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes', help = 'Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help = 'Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--depth', help = 'Resnet depth, must be one of 18, 34, 50, 101, 152', type = int, default = 50)
parser.add_argument('--epochs', help = 'Number of epochs', type = int, default = 100)
parser.add_argument('--use_tb', help = 'whether to use tensorboard', action = 'store_true')
parser.add_argument('--use_aug', help = 'whether to use data augmentation', action = 'store_true')
parser = parser.parse_args(args)
session = parser.s
session_dir = 'session_{:02d}'.format(session)
assert os.path.isdir('models'), '[ERROR] models folder not exist'
assert os.path.isdir('logs'), '[ERROR] logs folder not exist'
model_dir = os.path.join('models', session_dir)
logs_dir = os.path.join('logs', session_dir)
if not os.path.isdir(model_dir):
os.mkdir(model_dir)
if not os.path.isdir(logs_dir):
os.mkdir(logs_dir)
# set up tensorboard logger
tb_writer = None
if parser.use_tb:
tb_writer = SummaryWriter('logs')
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
if parser.use_aug:
#transform = transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
dataset_train = CocoDataset(parser.coco_path, set_name='train2017', transform = transforms.Compose([Normalizer(), Augmenter(), ToTensor()]))
else:
dataset_train = CocoDataset(parser.coco_path, set_name='train2017', transform = transforms.Compose([Normalizer(), ToTensor()]))
dataset_val = CocoDataset(parser.coco_path, set_name='val2017', transform = transforms.Compose([Normalizer(), ToTensor()]))
#transform = transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError('Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Augmenter(), ToTensor()]))
#transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Augmenter(), ToTensor()]))
#transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train, batch_size = parser.bs, drop_last = False)
dataloader_train = DataLoader(dataset_train, num_workers = 0, collate_fn = collater, batch_sampler = sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size = parser.bs, drop_last = False)
dataloader_val = DataLoader(dataset_val, num_workers = 0, collate_fn = collater, batch_sampler = sampler_val)
print('# classes: {}'.format(dataset_train.num_classes))
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes = dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes = dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes = dataset_train.num_classes(), pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes = dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes = dataset_train.num_classes(), pretrained=True)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
retinanet = retinanet.cuda()
# disable multi-GPU train
retinanet = torch.nn.DataParallel(retinanet).cuda()
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr = parser.lr)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience = 3, verbose = True)
loss_hist = collections.deque(maxlen = 500)
retinanet.train()
#retinanet.module.freeze_bn() if DataParallel activated
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
retinanet.train()
# retinanet.module.freeze_bn() if DataParallel activated
retinanet.module.freeze_bn()
epoch_loss = []
iter_per_epoch = len(dataloader_train)
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
assert data['img'][0].shape[0] == 3, '[ERROR] data first dim should be 3! ({})'.format(data['img'][0].shape)
# data['img']: (B, C, H, W)
# data['annot']: [x1, y1, x2, y2, class_id]
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
# epoch starts from 0
if (iter_num + 1) % 1 == 0:
print(
'Epoch: {} | Iteration: {} | Total loss: {:1.5f} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
epoch_num, iter_num, float(loss), float(classification_loss), float(regression_loss), np.mean(loss_hist)
)
)
# update tensorboard
if tb_writer is not None:
crt_iter = (epoch_num) * iter_per_epoch + (iter_num + 1)
tb_dict = {
'total_loss': float(loss),
'classification_loss': float(classification_loss),
'regression_loss': float(regression_loss)
}
tb_writer.add_scalars('session_{:02d}/loss'.format(session), tb_dict, crt_iter)
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
if (epoch_num + 1) % parser.save_int == 0:
# retinanet (before DataParallel): <class 'retinanet.model.ResNet'>, no self.module
# retinanet (after DataParallel): <class 'torch.nn.parallel.data_parallel.DataParallel>, self.module available
# retinanet.module (after DataParallel): <class 'retinanet.model.ResNet'>
torch.save(retinanet.module.state_dict(), os.path.join(model_dir, 'retinanet_s{:02d}_e{:03d}.pth'.format(session, epoch_num)))
if parser.use_tb:
tb_writer.close()
retinanet.eval()
torch.save(retinanet.module.state_dict(), os.path.join(model_dir, 'retinanet_s{:02d}_e{:03d}.pth'.format(session, epoch_num)))
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=100)
parser.add_argument('--local_rank', help='Local rank', type=int, default=0)
parser.add_argument('--distributed', action='store_true')
parser.add_argument('--pretrained', action='store_true')
parser = parser.parse_args(args)
torch.cuda.set_device(parser.local_rank)
DISTRIBUTED = parser.distributed and config.DISTRIBUTED
if DISTRIBUTED:
distributed.init_process_group(backend="nccl")
device = torch.device(f'cuda:{parser.local_rank}')
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
if DISTRIBUTED:
sampler = DistributedSampler(dataset_train)
dataloader_train = DataLoader(dataset_train,
num_workers=4,
batch_size=batch_size,
collate_fn=collater,
sampler=sampler,
pin_memory=True,
drop_last=True)
if dataset_val is not None:
sampler_val = DistributedSampler(dataset_val)
dataloader_val = DataLoader(dataset_val,
batch_size=1,
num_workers=4,
collate_fn=collater,
sampler=sampler_val,
pin_memory=True,
drop_last=True)
else:
sampler = AspectRatioBasedSampler(dataset_train,
batch_size=batch_size,
drop_last=False)
dataloader_train = DataLoader(dataset_train,
num_workers=4,
collate_fn=collater,
batch_sampler=sampler,
pin_memory=True)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val,
batch_size=1,
drop_last=False)
dataloader_val = DataLoader(dataset_val,
num_workers=4,
collate_fn=collater,
batch_sampler=sampler_val,
pin_memory=True)
# Create the model
if parser.depth == 18:
retinanet = model.retinanet18(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrained)
elif parser.depth == 34:
retinanet = model.retinanet34(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrained)
elif parser.depth == 50:
retinanet = model.retinanet50(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrained)
elif parser.depth == 101:
retinanet = model.retinanet101(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrained)
elif parser.depth == 152:
retinanet = model.retinanet152(num_classes=dataset_train.num_classes(),
pretrained=parser.pretrained)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
if use_cuda:
retinanet = retinanet.cuda()
if RESTORE:
retinanet.load_state_dict(torch.load(RESTORE))
if DISTRIBUTED:
retinanet = torch.nn.parallel.DistributedDataParallel(
retinanet, device_ids=[parser.local_rank])
print("Let's use", parser.local_rank, "GPU!")
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
if DISTRIBUTED:
retinanet.module.freeze_bn()
else:
retinanet.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
for epoch_num in range(parser.epochs):
save_to_disk = parser.local_rank == 0
retinanet.train()
if DISTRIBUTED:
retinanet.module.freeze_bn()
else:
retinanet.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if use_cuda:
classification_loss, regression_loss = retinanet(
[data['img'].cuda().float(), data['annot'].cuda()])
else:
classification_loss, regression_loss = retinanet(
[data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
if save_to_disk:
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, iter_num,
float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if save_to_disk:
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
if DISTRIBUTED:
torch.save(
retinanet.module.state_dict(),
'{}/{}_retinanet_{}.pt'.format(checkpoints_dir,
parser.dataset, epoch_num))
else:
torch.save(
retinanet.state_dict(),
'{}/{}_retinanet_{}.pt'.format(checkpoints_dir,
parser.dataset, epoch_num))
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_train', help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--depth', help='Resnet depth, must be one of 18, 34, 50, 101, 152', type=int, default=50)
parser.add_argument('--epochs', help='Number of epochs', type=int, default=100)
parser.add_argument('--dcn_layers', type =str, help = 'comma seperated str where laters to be used, 0..3',default = None)
parser.add_argument('--use_depth', action='store_true', help='if specified, use depth for deformconv')
parser = parser.parse_args(args)
use_dcn = [False, False, False, False]
if parser.dcn_layers is not None:
_t = parser.dcn_layers.split(',')
for __t in _t:
use_dcn[int(__t)] = True
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path, set_name='train2017',
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError('Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Augmenter(), Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train, batch_size=128, drop_last=False)
dataloader_train = DataLoader(dataset_train, num_workers=3, collate_fn=collater, batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=1, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=3, collate_fn=collater, batch_sampler=sampler_val)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(), pretrained=True, use_dcn = use_dcn, use_depth = parser.use_depth)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(), pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(), pretrained=True)
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
writer = SummaryWriter()
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience=3, verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
#mAP = csv_eval.evaluate(dataset_val, retinanet)
global_step = 0
for epoch_num in range(parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
global_step += 1
if torch.cuda.is_available():
if parser.use_depth and 'depth' in data:
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']],depth = data['depth'].cuda())
else:
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
else:
if parser.use_depth and 'depth' in data:
classification_loss, regression_loss = retinanet([data['img'].float(), data['annot']],depth=data['depth'])
else:
classification_loss, regression_loss = retinanet([data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
writer.add_scalar('CLS Loss',classification_loss,global_step)
writer.add_scalar('REG Loss',regression_loss,global_step)
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
epoch_num, iter_num, float(classification_loss), float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(retinanet.module, '{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
writer.close()
def main(args=None):
parser = argparse.ArgumentParser(description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset', help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument('--csv_train', help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes', help='Path to file containing class list (see readme)')
parser.add_argument('--csv_val', help='Path to file containing validation annotations (optional, see readme)')
parser.add_argument('--depth', help='Resnet depth, must be one of 18, 34, 50, 101, 152', type=int, default=50)
parser.add_argument('--config', help='Config file path that contains scale and ratio values', type=str)
parser.add_argument('--epochs', help='Number of epochs', type=int, default=50)
parser.add_argument('--init-lr', help='Initial learning rate for training process', type=float, default=1e-3)
parser.add_argument('--batch-size', help='Number of input images per step', type=int, default=1)
parser.add_argument('--num-workers', help='Number of worker used in dataloader', type=int, default=1)
# For resuming training from saved checkpoint
parser.add_argument('--resume', help='Whether to resume training from checkpoint', action='store_true')
parser.add_argument('--saved-ckpt', help='Resume training from this checkpoint', type=str)
parser.add_argument('--multi-gpus', help='Allow to use multi gpus for training task', action='store_true')
parser.add_argument('--snapshots', help='Location to save training snapshots', type=str, default="snapshots")
parser.add_argument('--log-dir', help='Location to save training logs', type=str, default="logs")
parser.add_argument('--expr-augs', help='Allow to use use experiment augmentation methods', action='store_true')
parser.add_argument('--aug-methods', help='(Experiment) Augmentation methods to use, separate by comma symbol', type=str, default="rotate,hflip,brightness,contrast")
parser.add_argument('--aug-prob', help='Probability of applying (experiment) augmentation in range [0.,1.]', type=float, default=0.5)
parser = parser.parse_args(args)
train_transforms = [Normalizer(), Resizer(), Augmenter()]
# Define transform methods
if parser.expr_augs:
aug_map = get_aug_map(p=parser.aug_prob)
aug_methods = parser.aug_methods.split(",")
for aug in aug_methods:
if aug in aug_map.keys():
train_transforms.append(aug_map[aug])
else:
print(f"{aug} is not available.")
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path, set_name='train2017',
transform=transforms.Compose(train_transforms))
dataset_val = CocoDataset(parser.coco_path, set_name='val2017',
transform=transforms.Compose([Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError('Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train, class_list=parser.csv_classes,
transform=transforms.Compose(train_transforms))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val, class_list=parser.csv_classes,
transform=transforms.Compose([Normalizer(), Resizer()]))
else:
raise ValueError('Dataset type not understood (must be csv or coco), exiting.')
sampler = AspectRatioBasedSampler(dataset_train, batch_size=parser.batch_size, drop_last=False)
dataloader_train = DataLoader(dataset_train, num_workers=parser.num_workers, collate_fn=collater, batch_sampler=sampler)
if dataset_val is not None:
sampler_val = AspectRatioBasedSampler(dataset_val, batch_size=parser.batch_size, drop_last=False)
dataloader_val = DataLoader(dataset_val, num_workers=parser.num_workers, collate_fn=collater, batch_sampler=sampler_val)
config = dict({"scales": None,
"ratios": None})
if parser.config:
config = load_config(parser.config, config)
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(), pretrained=True, ratios=config["ratios"], scales=config["scales"])
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(), pretrained=True, ratios=config["ratios"], scales=config["scales"])
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(), pretrained=True, ratios=config["ratios"], scales=config["scales"])
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(), pretrained=True, ratios=config["ratios"], scales=config["scales"])
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(), pretrained=True, ratios=config["ratios"], scales=config["scales"])
else:
raise ValueError('Unsupported model depth, must be one of 18, 34, 50, 101, 152')
optimizer = optim.Adam(retinanet.parameters(), lr=parser.init_lr)
if parser.resume:
if not parser.saved_ckpt:
print("No saved checkpoint provided for resuming training. Exiting now...")
return
if not os.path.exists(parser.saved_ckpt):
print("Invalid saved checkpoint path. Exiting now...")
return
# Restore last state
retinanet, optimizer, start_epoch = load_ckpt(parser.saved_ckpt, retinanet, optimizer)
if parser.epochs <= start_epoch:
print("Number of epochs must be higher than number of trained epochs of saved checkpoint.")
return
use_gpu = True
if use_gpu:
print("Using GPU for training process")
if torch.cuda.is_available():
if parser.multi_gpus:
print("Using multi-gpus for training process")
retinanet = torch.nn.DataParallel(retinanet.cuda(), device_ids=[0,1])
else:
retinanet = torch.nn.DataParallel(retinanet.cuda())
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, patience=1, verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
# Tensorboard writer
writer = SummaryWriter(parser.log_dir)
# Save snapshots dir
if not os.path.exists(parser.snapshots):
os.makedirs(parser.snapshots)
best_mAP = 0
start_epoch = 0 if not parser.resume else start_epoch
for epoch_num in range(start_epoch, parser.epochs):
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
epoch_csf_loss = []
epoch_reg_loss = []
for iter_num, data in enumerate(dataloader_train):
try:
optimizer.zero_grad()
if torch.cuda.is_available():
with torch.cuda.device(0):
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
else:
classification_loss, regression_loss = retinanet([data['img'].float(), data['annot']])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
epoch_csf_loss.append(float(classification_loss))
epoch_reg_loss.append(float(regression_loss))
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
print(
'\rEpoch: {}/{} | Iteration: {}/{} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'.format(
(epoch_num + 1), parser.epochs, (iter_num + 1), len(dataloader_train), float(classification_loss), float(regression_loss), np.mean(loss_hist)), end='')
del classification_loss
del regression_loss
except Exception as e:
print(e)
continue
# writer.add_scalar("Loss/train", loss, epoch_num)
_epoch_loss = np.mean(epoch_loss)
_epoch_csf_loss = np.mean(epoch_reg_loss)
_epoch_reg_loss = np.mean(epoch_reg_loss)
if parser.dataset == 'coco':
print('Evaluating dataset')
coco_eval.evaluate_coco(dataset_val, retinanet)
scheduler.step(_epoch_loss)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('\nEvaluating dataset')
APs = csv_eval.evaluate(dataset_val, retinanet)
mAP = round(mean(APs[ap][0] for ap in APs.keys()), 5)
print("mAP: %f" %mAP)
writer.add_scalar("validate/mAP", mAP, epoch_num)
# Handle lr_scheduler wuth mAP value
scheduler.step(mAP)
lr = get_lr(optimizer)
writer.add_scalar("train/classification-loss", _epoch_csf_loss, epoch_num)
writer.add_scalar("train/regression-loss", _epoch_reg_loss, epoch_num)
writer.add_scalar("train/loss", _epoch_loss, epoch_num)
writer.add_scalar("train/learning-rate", lr, epoch_num)
# Save model file, optimizer and epoch number
checkpoint = {
'epoch': epoch_num,
'state_dict': retinanet.state_dict(),
'optimizer': optimizer.state_dict(),
}
# torch.save(retinanet.module, os.path.join(parser.snapshots, '{}_retinanet_{}.pt'.format(parser.dataset, epoch_num)))
# Check whether this epoch's model achieves highest mAP value
is_best = False
if best_mAP < mAP:
best_mAP = mAP
is_best = True
save_ckpt(checkpoint, is_best, parser.snapshots, '{}_retinanet_{}.pt'.format(parser.dataset, epoch_num + 1))
print('\n')
retinanet.eval()
torch.save(retinanet, 'model_final.pt')
writer.flush()
def main(args=None):
parser = argparse.ArgumentParser(
description='Simple training script for training a RetinaNet network.')
parser.add_argument('--dataset',
help='Dataset type, must be one of csv or coco.')
parser.add_argument('--coco_path', help='Path to COCO directory')
parser.add_argument(
'--csv_train',
help='Path to file containing training annotations (see readme)')
parser.add_argument('--csv_classes',
help='Path to file containing class list (see readme)')
parser.add_argument(
'--csv_val',
help=
'Path to file containing validation annotations (optional, see readme)'
)
parser.add_argument(
'--depth',
help='Resnet depth, must be one of 18, 34, 50, 101, 152',
type=int,
default=50)
parser.add_argument('--epochs',
help='Number of epochs',
type=int,
default=25)
parser = parser.parse_args(args)
# Create the data loaders
if parser.dataset == 'coco':
if parser.coco_path is None:
raise ValueError('Must provide --coco_path when training on COCO,')
dataset_train = CocoDataset(parser.coco_path,
set_name='train2017',
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
dataset_val = CocoDataset(parser.coco_path,
set_name='val2017',
transform=transforms.Compose(
[Normalizer(), Resizer()]))
elif parser.dataset == 'csv':
if parser.csv_train is None:
raise ValueError('Must provide --csv_train when training on COCO,')
if parser.csv_classes is None:
raise ValueError(
'Must provide --csv_classes when training on COCO,')
dataset_train = CSVDataset(train_file=parser.csv_train,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Augmenter(),
Resizer()]))
if parser.csv_val is None:
dataset_val = None
print('No validation annotations provided.')
else:
dataset_val = CSVDataset(train_file=parser.csv_val,
class_list=parser.csv_classes,
transform=transforms.Compose(
[Normalizer(),
Resizer()]))
else:
raise ValueError(
'Dataset type not understood (must be csv or coco), exiting.')
# create samplers for both training and validation
# using muti CPU cores to accelerate data loading
sampler_train1 = torch.utils.data.SequentialSampler(dataset_train)
sampler_train2 = torch.utils.data.BatchSampler(sampler_train1,
batch_size=1,
drop_last=True)
dataloader_train = DataLoader(dataset_train,
num_workers=10,
collate_fn=collater,
batch_sampler=sampler_train2)
sampler_val1 = torch.utils.data.SequentialSampler(dataset_val)
sampler_val2 = torch.utils.data.BatchSampler(sampler_val1,
batch_size=1,
drop_last=True)
dataloader_val = DataLoader(dataset_val,
num_workers=10,
collate_fn=collater,
batch_sampler=sampler_val2)
# Create the model
if parser.depth == 18:
retinanet = model.resnet18(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 34:
retinanet = model.resnet34(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 50:
retinanet = model.resnet50(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 101:
retinanet = model.resnet101(num_classes=dataset_train.num_classes(),
pretrained=True)
elif parser.depth == 152:
retinanet = model.resnet152(num_classes=dataset_train.num_classes(),
pretrained=True)
else:
raise ValueError(
'Unsupported model depth, must be one of 18, 34, 50, 101, 152')
use_gpu = True
if use_gpu:
if torch.cuda.is_available():
retinanet = retinanet.cuda()
if torch.cuda.is_available():
retinanet = torch.nn.DataParallel(retinanet).cuda()
else:
retinanet = torch.nn.DataParallel(retinanet)
retinanet.training = True
# ADAM optimizer
optimizer = optim.Adam(retinanet.parameters(), lr=1e-5)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
verbose=True)
loss_hist = collections.deque(maxlen=500)
retinanet.train()
retinanet.module.freeze_bn()
print('Num training images: {}'.format(len(dataset_train)))
# using tensorboardX to show training process
writer = SummaryWriter('log')
iter_sum = 0
time_sum = 0
frame_num = 8
for epoch_num in range(parser.epochs):
# only work for frame_num > 8
frame_list = collections.deque(maxlen=frame_num)
anno_list = collections.deque(maxlen=frame_num)
retinanet.train()
retinanet.module.freeze_bn()
epoch_loss = []
for index, data in enumerate(dataloader_train):
try:
frame_list.append(data['img'])
anno_list.append(data['annot'])
# if frame_num != 32:
if index < 31:
continue
if index >= 697 and index <= 697 + 32:
continue
# real_frame is the frame we used for fish detection
# It's the last frame in the batch group
real_frame = frame_list[-1]
# the annotation for real_frame
annot = anno_list[-1]
# drop useless frames
data['img'] = torch.cat(list(frame_list), dim=0)
optimizer.zero_grad()
classification_loss, regression_loss = retinanet([
data['img'].cuda().float(),
real_frame.cuda().float(),
annot.cuda().float()
])
classification_loss = classification_loss.mean()
regression_loss = regression_loss.mean()
loss = classification_loss + regression_loss
if bool(loss == 0):
continue
loss.backward()
torch.nn.utils.clip_grad_norm_(retinanet.parameters(), 0.1)
optimizer.step()
loss_hist.append(float(loss))
epoch_loss.append(float(loss))
writer.add_scalar('loss_hist', np.mean(loss_hist), iter_sum)
writer.add_scalar('classification_loss',
float(classification_loss), iter_sum)
writer.add_scalar('regression_loss', float(regression_loss),
iter_sum)
writer.add_scalar('loss', float(loss), iter_sum)
print(
'Epoch: {} | Iteration: {} | Classification loss: {:1.5f} | Regression loss: {:1.5f} | Running loss: {:1.5f}'
.format(epoch_num, index, float(classification_loss),
float(regression_loss), np.mean(loss_hist)))
del classification_loss
del regression_loss
iter_sum = iter_sum + 1
except Exception as e:
print(e)
continue
if parser.dataset == 'coco':
print('Evaluating dataset')
# evaluate coco
coco_eval.evaluate_coco(dataset_val, dataloader_val, retinanet,
frame_num)
elif parser.dataset == 'csv' and parser.csv_val is not None:
print('Evaluating dataset')
mAP = csv_eval.evaluate(dataset_val, retinanet)
scheduler.step(np.mean(epoch_loss))
torch.save(
retinanet.module,
'checkpoint/{}_retinanet_{}.pt'.format(parser.dataset, epoch_num))
retinanet.eval()
torch.save(retinanet, 'save/model_final.pt')
writer.close()