def main(parser_data):
device = torch.device(parser_data.device if torch.cuda.is_available() else "cpu")
print("Using {} device training.".format(device.type))
data_transform = {
"val": transforms.Compose([transforms.Resize(),
transforms.ToTensor(),
transforms.Normalization()])
}
# read class_indict
label_json_path = './pascal_voc_classes.json'
assert os.path.exists(label_json_path), "json file {} dose not exist.".format(label_json_path)
json_file = open(label_json_path, 'r')
class_dict = json.load(json_file)
json_file.close()
category_index = {v: k for k, v in class_dict.items()}
VOC_root = parser_data.data_path
# check voc root
if os.path.exists(os.path.join(VOC_root, "VOCdevkit")) is False:
raise FileNotFoundError("VOCdevkit dose not in path:'{}'.".format(VOC_root))
# 注意这里的collate_fn是自定义的，因为读取的数据包括image和targets，不能直接使用默认的方法合成batch
batch_size = parser_data.batch_size
nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0, 8]) # number of workers
print('Using %g dataloader workers' % nw)
# load validation data set
# VOCdevkit -> VOC2012 -> ImageSets -> Main -> val.txt
val_dataset = VOCDataSet(VOC_root, "2012", transforms=data_transform["val"], train_set="val.txt")
val_dataset_loader = torch.utils.data.DataLoader(val_dataset,
batch_size=batch_size,
shuffle=False,
num_workers=nw,
pin_memory=True,
collate_fn=val_dataset.collate_fn)
# create model num_classes equal background + 20 classes
backbone = Backbone()
model = SSD300(backbone=backbone, num_classes=parser_data.num_classes + 1)
# 载入你自己训练好的模型权重
weights_path = parser_data.weights
assert os.path.exists(weights_path), "not found {} file.".format(weights_path)
model.load_state_dict(torch.load(weights_path, map_location=device)['model'])
# print(model)
model.to(device)
# evaluate on the test dataset
coco = get_coco_api_from_dataset(val_dataset)
iou_types = ["bbox"]
coco_evaluator = CocoEvaluator(coco, iou_types)
cpu_device = torch.device("cpu")
model.eval()
with torch.no_grad():
for images, targets in tqdm(val_dataset_loader, desc="validation..."):
# 将图片传入指定设备device
images = torch.stack(images, dim=0).to(device)
# inference
results = model(images)
outputs = []
for index, (bboxes_out, labels_out, scores_out) in enumerate(results):
# 将box的相对坐标信息（0-1）转为绝对值坐标(xmin, ymin, xmax, ymax)
height_width = targets[index]["height_width"]
# 还原回原图尺度
bboxes_out[:, [0, 2]] = bboxes_out[:, [0, 2]] * height_width[1]
bboxes_out[:, [1, 3]] = bboxes_out[:, [1, 3]] * height_width[0]
info = {"boxes": bboxes_out.to(cpu_device),
"labels": labels_out.to(cpu_device),
"scores": scores_out.to(cpu_device)}
outputs.append(info)
res = {target["image_id"].item(): output for target, output in zip(targets, outputs)}
coco_evaluator.update(res)
coco_evaluator.synchronize_between_processes()
# accumulate predictions from all images
coco_evaluator.accumulate()
coco_evaluator.summarize()
coco_eval = coco_evaluator.coco_eval["bbox"]
# calculate COCO info for all classes
coco_stats, print_coco = summarize(coco_eval)
# calculate voc info for every classes(IoU=0.5)
voc_map_info_list = []
for i in range(len(category_index)):
stats, _ = summarize(coco_eval, catId=i)
voc_map_info_list.append(" {:15}: {}".format(category_index[i + 1], stats[1]))
print_voc = "\n".join(voc_map_info_list)
print(print_voc)
# 将验证结果保存至txt文件中
with open("record_mAP.txt", "w") as f:
record_lines = ["COCO results:",
print_coco,
"",
"mAP(IoU=0.5) for each category:",
print_voc]
f.write("\n".join(record_lines))
def main(parser_data):
device = torch.device(
parser_data.device if torch.cuda.is_available() else "cpu")
print("Using {} device training.".format(device.type))
if not os.path.exists("save_weights"):
os.mkdir("save_weights")
results_file = "results{}.txt".format(
datetime.datetime.now().strftime("%Y%m%d-%H%M%S"))
data_transform = {
"train":
transforms.Compose([
transforms.SSDCropping(),
transforms.Resize(),
transforms.ColorJitter(),
transforms.ToTensor(),
transforms.RandomHorizontalFlip(),
transforms.Normalization(),
transforms.AssignGTtoDefaultBox()
]),
"val":
transforms.Compose([
transforms.Resize(),
transforms.ToTensor(),
transforms.Normalization()
])
}
VOC_root = parser_data.data_path
# check voc root
if os.path.exists(os.path.join(VOC_root, "VOCdevkit")) is False:
raise FileNotFoundError(
"VOCdevkit dose not in path:'{}'.".format(VOC_root))
# VOCdevkit -> VOC2012 -> ImageSets -> Main -> train.txt
train_dataset = VOCDataSet(VOC_root,
"2012",
data_transform['train'],
train_set='train.txt')
# 注意训练时，batch_size必须大于1
batch_size = parser_data.batch_size
assert batch_size > 1, "batch size must be greater than 1"
# 防止最后一个batch_size=1，如果最后一个batch_size=1就舍去
drop_last = True if len(train_dataset) % batch_size == 1 else False
nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0,
8]) # number of workers
print('Using %g dataloader workers' % nw)
train_data_loader = torch.utils.data.DataLoader(
train_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=nw,
collate_fn=train_dataset.collate_fn,
drop_last=drop_last)
# VOCdevkit -> VOC2012 -> ImageSets -> Main -> val.txt
val_dataset = VOCDataSet(VOC_root,
"2012",
data_transform['val'],
train_set='val.txt')
val_data_loader = torch.utils.data.DataLoader(
val_dataset,
batch_size=batch_size,
shuffle=False,
num_workers=nw,
collate_fn=train_dataset.collate_fn)
model = create_model(num_classes=args.num_classes + 1)
model.to(device)
# define optimizer
params = [p for p in model.parameters() if p.requires_grad]
optimizer = torch.optim.SGD(params,
lr=0.0005,
momentum=0.9,
weight_decay=0.0005)
# learning rate scheduler
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
step_size=5,
gamma=0.3)
# 如果指定了上次训练保存的权重文件地址，则接着上次结果接着训练
if parser_data.resume != "":
checkpoint = torch.load(parser_data.resume, map_location='cpu')
model.load_state_dict(checkpoint['model'])
optimizer.load_state_dict(checkpoint['optimizer'])
lr_scheduler.load_state_dict(checkpoint['lr_scheduler'])
parser_data.start_epoch = checkpoint['epoch'] + 1
print("the training process from epoch{}...".format(
parser_data.start_epoch))
train_loss = []
learning_rate = []
val_map = []
# 提前加载验证集数据，以免每次验证时都要重新加载一次数据，节省时间
val_data = get_coco_api_from_dataset(val_data_loader.dataset)
for epoch in range(parser_data.start_epoch, parser_data.epochs):
mean_loss, lr = utils.train_one_epoch(model=model,
optimizer=optimizer,
data_loader=train_data_loader,
device=device,
epoch=epoch,
print_freq=50)
train_loss.append(mean_loss.item())
learning_rate.append(lr)
# update learning rate
lr_scheduler.step()
coco_info = utils.evaluate(model=model,
data_loader=val_data_loader,
device=device,
data_set=val_data)
# write into txt
with open(results_file, "a") as f:
# 写入的数据包括coco指标还有loss和learning rate
result_info = [
str(round(i, 4)) for i in coco_info + [mean_loss.item()]
] + [str(round(lr, 6))]
txt = "epoch:{} {}".format(epoch, ' '.join(result_info))
f.write(txt + "\n")
val_map.append(coco_info[1]) # pascal mAP
# save weights
save_files = {
'model': model.state_dict(),
'optimizer': optimizer.state_dict(),
'lr_scheduler': lr_scheduler.state_dict(),
'epoch': epoch
}
torch.save(save_files, "./save_weights/ssd300-{}.pth".format(epoch))
# plot loss and lr curve
if len(train_loss) != 0 and len(learning_rate) != 0:
from plot_curve import plot_loss_and_lr
plot_loss_and_lr(train_loss, learning_rate)
# plot mAP curve
if len(val_map) != 0:
from plot_curve import plot_map
plot_map(val_map)
def main():
# get devices
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
# create model
# 目标检测数 + 背景
num_classes = 20 + 1
model = create_model(num_classes=num_classes)
# load train weights
train_weights = "./save_weights/ssd300-14.pth"
train_weights_dict = torch.load(train_weights, map_location=device)['model']
model.load_state_dict(train_weights_dict)
model.to(device)
# read class_indict
json_path = "./pascal_voc_classes.json"
assert os.path.exists(json_path), "file '{}' dose not exist.".format(json_path)
json_file = open(json_path, 'r')
class_dict = json.load(json_file)
category_index = {v: k for k, v in class_dict.items()}
# load image
original_img = Image.open("./test.jpg")
# from pil image to tensor, do not normalize image
data_transform = transforms.Compose([transforms.Resize(),
transforms.ToTensor(),
transforms.Normalization()])
img, _ = data_transform(original_img)
# expand batch dimension
img = torch.unsqueeze(img, dim=0)
model.eval()
with torch.no_grad():
# initial model
init_img = torch.zeros((1, 3, 300, 300), device=device)
model(init_img)
time_start = time_synchronized()
predictions = model(img.to(device))[0] # bboxes_out, labels_out, scores_out
time_end = time_synchronized()
print("inference+NMS time: {}".format(time_end - time_start))
predict_boxes = predictions[0].to("cpu").numpy()
predict_boxes[:, [0, 2]] = predict_boxes[:, [0, 2]] * original_img.size[0]
predict_boxes[:, [1, 3]] = predict_boxes[:, [1, 3]] * original_img.size[1]
predict_classes = predictions[1].to("cpu").numpy()
predict_scores = predictions[2].to("cpu").numpy()
if len(predict_boxes) == 0:
print("没有检测到任何目标!")
draw_box(original_img,
predict_boxes,
predict_classes,
predict_scores,
category_index,
thresh=0.5,
line_thickness=5)
plt.imshow(original_img)
plt.show()
def main(args):
init_distributed_mode(args)
print(args)
device = torch.device(args.device)
results_file = "results{}.txt".format(
datetime.datetime.now().strftime("%Y%m%d-%H%M%S"))
# Data loading code
print("Loading data")
data_transform = {
"train":
transforms.Compose([
transforms.SSDCropping(),
transforms.Resize(),
transforms.ColorJitter(),
transforms.ToTensor(),
transforms.RandomHorizontalFlip(),
transforms.Normalization(),
transforms.AssignGTtoDefaultBox()
]),
"val":
transforms.Compose([
transforms.Resize(),
transforms.ToTensor(),
transforms.Normalization()
])
}
VOC_root = args.data_path
# check voc root
if os.path.exists(os.path.join(VOC_root, "VOCdevkit")) is False:
raise FileNotFoundError(
"VOCdevkit dose not in path:'{}'.".format(VOC_root))
# load train data set
train_data_set = VOC2012DataSet(VOC_root,
data_transform["train"],
train_set='train.txt')
# load validation data set
val_data_set = VOC2012DataSet(VOC_root,
data_transform["val"],
train_set='val.txt')
print("Creating data loaders")
if args.distributed:
train_sampler = torch.utils.data.distributed.DistributedSampler(
train_data_set)
test_sampler = torch.utils.data.distributed.DistributedSampler(
val_data_set)
else:
train_sampler = torch.utils.data.RandomSampler(train_data_set)
test_sampler = torch.utils.data.SequentialSampler(val_data_set)
if args.aspect_ratio_group_factor >= 0:
# 统计所有图像比例在bins区间中的位置索引
group_ids = create_aspect_ratio_groups(
train_data_set, k=args.aspect_ratio_group_factor)
train_batch_sampler = GroupedBatchSampler(train_sampler, group_ids,
args.batch_size)
else:
train_batch_sampler = torch.utils.data.BatchSampler(train_sampler,
args.batch_size,
drop_last=True)
data_loader = torch.utils.data.DataLoader(
train_data_set,
batch_sampler=train_batch_sampler,
num_workers=args.workers,
collate_fn=train_data_set.collate_fn)
data_loader_test = torch.utils.data.DataLoader(
val_data_set,
batch_size=1,
sampler=test_sampler,
num_workers=args.workers,
collate_fn=train_data_set.collate_fn)
print("Creating model")
model = create_model(num_classes=args.num_classes + 1, device=device)
model_without_ddp = model
if args.distributed:
model = torch.nn.parallel.DistributedDataParallel(
model, device_ids=[args.gpu])
model_without_ddp = model.module
params = [p for p in model.parameters() if p.requires_grad]
optimizer = torch.optim.SGD(params,
lr=args.lr,
momentum=args.momentum,
weight_decay=args.weight_decay)
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
step_size=args.lr_step_size,
gamma=args.lr_gamma)
# lr_scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=args.lr_steps, gamma=args.lr_gamma)
# 如果传入resume参数，即上次训练的权重地址，则接着上次的参数训练
if args.resume:
# If map_location is missing, torch.load will first load the module to CPU
# and then copy each parameter to where it was saved,
# which would result in all processes on the same machine using the same set of devices.
checkpoint = torch.load(
args.resume, map_location='cpu') # 读取之前保存的权重文件(包括优化器以及学习率策略)
model_without_ddp.load_state_dict(checkpoint['model'])
optimizer.load_state_dict(checkpoint['optimizer'])
lr_scheduler.load_state_dict(checkpoint['lr_scheduler'])
args.start_epoch = checkpoint['epoch'] + 1
if args.test_only:
utils.evaluate(model, data_loader_test, device=device)
return
train_loss = []
learning_rate = []
val_map = []
print("Start training")
start_time = time.time()
for epoch in range(args.start_epoch, args.epochs):
if args.distributed:
train_sampler.set_epoch(epoch)
mean_loss, lr = utils.train_one_epoch(model,
optimizer,
data_loader,
device,
epoch,
args.print_freq,
warmup=True)
# only first process to save training info
if args.rank in [-1, 0]:
train_loss.append(mean_loss.item())
learning_rate.append(lr)
# update learning rate
lr_scheduler.step()
# evaluate after every epoch
coco_info = utils.evaluate(model, data_loader_test, device=device)
if args.rank in [-1, 0]:
# write into txt
with open(results_file, "a") as f:
# 写入的数据包括coco指标还有loss和learning rate
result_info = [
str(round(i, 4))
for i in coco_info + [mean_loss.item(), lr]
]
txt = "epoch:{} {}".format(epoch, ' '.join(result_info))
f.write(txt + "\n")
val_map.append(coco_info[1]) # pascal mAP
if args.output_dir:
# 只在主节点上执行保存权重操作
save_on_master(
{
'model': model_without_ddp.state_dict(),
'optimizer': optimizer.state_dict(),
'lr_scheduler': lr_scheduler.state_dict(),
'args': args,
'epoch': epoch
}, os.path.join(args.output_dir, 'model_{}.pth'.format(epoch)))
total_time = time.time() - start_time
total_time_str = str(datetime.timedelta(seconds=int(total_time)))
print('Training time {}'.format(total_time_str))
if args.rank in [-1, 0]:
# plot loss and lr curve
if len(train_loss) != 0 and len(learning_rate) != 0:
from plot_curve import plot_loss_and_lr
plot_loss_and_lr(train_loss, learning_rate)
# plot mAP curve
if len(val_map) != 0:
from plot_curve import plot_map
plot_map(val_map)