def evaluate(model, dataset_name, list_path, iou_thres, conf_thres, nms_thres,
img_size, batch_size):
""" Evaluate with the given model on the dataset specified by the given dataset_path to list of samples """
model.eval()
# Get dataloader
dataset = ListDataset(dataset_name,
list_path,
img_size=img_size,
augment=False,
multiscale=False)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=batch_size,
shuffle=False,
num_workers=1,
collate_fn=dataset.collate_fn)
Tensor = torch.cuda.FloatTensor if torch.cuda.is_available(
) else torch.FloatTensor
labels = []
sample_metrics = [] # List of tuples (TP, confs, pred)
for batch_i, (img_paths, imgs, targets) in enumerate(
tqdm.tqdm(dataloader, desc="Validation round")):
print(img_paths)
imgs = Variable(imgs.type(Tensor), requires_grad=False)
labels += targets[:, 1].tolist() # extract labels
targets[:, 2:] = xywh2xyxy(targets[:, 2:])
targets[:, 2:] *= img_size # rescale target
# model forward
with torch.no_grad():
outputs = model(imgs)
outputs = non_max_suppression(outputs,
conf_thres=conf_thres,
nms_thres=nms_thres)
# evaluation stats
print(f'outputs: {outputs}')
sample_metrics += get_batch_statistics(outputs,
targets,
iou_threshold=iou_thres)
# concatenate sample statistics
true_positives, pred_scores, pred_labels = [
np.concatenate(x, 0) for x in list(zip(*sample_metrics))
]
precision, recall, AP, f1, ap_class = ap_per_class(true_positives,
pred_scores,
pred_labels, labels)
return precision, recall, AP, f1, ap_class
def evaluate(model, path, iou_thres, conf_thres, nms_thres, img_size,
batch_size):
model.eval()
# Get dataloader
dataset = ListDataset(label_file=path,
img_size=img_size,
augment=False,
multiscale=False,
normalized_labels=False)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=batch_size,
shuffle=False,
num_workers=1,
collate_fn=dataset.collate_fn)
Tensor = torch.cuda.FloatTensor if torch.cuda.is_available(
) else torch.FloatTensor
labels = []
sample_metrics = [] # List of tuples (TP, confs, pred)
for batch_i, (_, imgs, targets) in enumerate(
tqdm.tqdm(dataloader, desc="Detecting objects")):
# Extract labels
labels += targets[:, 1].tolist()
# Rescale target
targets[:, 2:] = xywh2xyxy(targets[:, 2:])
targets[:, 2:] *= img_size
imgs = Variable(imgs.type(Tensor), requires_grad=False)
with torch.no_grad():
outputs = model(imgs)
outputs = non_max_suppression(outputs,
conf_thres=conf_thres,
nms_thres=nms_thres)
sample_metrics += get_batch_statistics(outputs,
targets,
iou_threshold=iou_thres)
# Concatenate sample statistics
true_positives, pred_scores, pred_labels = [
np.concatenate(x, 0) for x in list(zip(*sample_metrics))
]
precision, recall, AP, f1, ap_class = ap_per_class(true_positives,
pred_scores,
pred_labels, labels)
return precision, recall, AP, f1, ap_class
if not fn.endswith('.jpg'):
continue
# input image
img_path = os.path.join(opt.image_folder, fn)
# single image prediction -> plot bbox and save
# detections_list is a list containting detections corresponded with samples
detections_list, detections_rescaled, inference_time = detect_img(
img_path, model, device, opt.img_size, classes, opt.conf_thres,
opt.nms_thres, output_dir, opt.save_plot)
# label/targets
label_path = img_path.replace('images',
'labels').replace('.jpg', '.txt')
_, targets = ListDataset(
opt.dataset_name,
'data/custom/train_valid/phantom_20/valid.txt').preprocess(
img_path, label_path)
labels += targets[:, 1].tolist()
targets[:, 2:] = xywh2xyxy(targets[:, 2:])
targets[:, 2:] *= opt.img_size # rescale target
# prediction speed
inference_time_total += inference_time
# validation score computation
sample_metrics += get_batch_statistics(detections_list,
targets,
iou_threshold=opt.iou_thres)
if detections_list[0] is not None:
ious = sample_metrics[-1][-1]
count += 1
class_names = load_classes(data_config["names"])
# Initiate model
model = Darknet(opt.model_def).to(device)
model.apply(weights_init_normal)
# If specified we start from checkpoint
if opt.pretrained_weights:
if opt.pretrained_weights.endswith(".pth"):
model.load_state_dict(torch.load(opt.pretrained_weights))
else:
model.load_darknet_weights(opt.pretrained_weights)
# Get dataloader
dataset = ListDataset(train_path,
augment=True,
multiscale=opt.multiscale_training)
dataloader = torch.utils.data.DataLoader(
dataset,
batch_size=opt.batch_size,
shuffle=True,
num_workers=opt.n_cpu,
pin_memory=True,
collate_fn=dataset.collate_fn,
)
optimizer = torch.optim.Adam(model.parameters())
metrics = [
"grid_size",
"loss",
class_names = load_classes(data_config["names"])
# Initiate model
model = Darknet(opt.model_def).to(device)
model.apply(weights_init_normal)
# If specified we start from checkpoint
if opt.pretrained_weights:
if opt.pretrained_weights.endswith(".pth"):
model.load_state_dict(torch.load(opt.pretrained_weights))
else:
model.load_darknet_weights(opt.pretrained_weights)
# Get dataloader
dataset = ListDataset(label_file=train_path,
augment=True,
multiscale=opt.multiscale_training,
normalized_labels=False)
dataloader = torch.utils.data.DataLoader(
dataset,
batch_size=opt.batch_size,
shuffle=True,
num_workers=opt.n_cpu,
pin_memory=True,
collate_fn=dataset.collate_fn,
)
optimizer = torch.optim.Adam(model.parameters())
metrics = [
"grid_size",
"loss",
img, boxes, labels = random_crop(img, boxes, labels)
img, boxes = resize(img,
boxes,
size=(img_size, img_size),
random_interpolation=True)
img, boxes = random_flip(img, boxes)
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
trainset = ListDataset(root='/usr/local/share/data/SynthText/',
list_file='data_train.txt',
transform=transform_train)
def transform_test(img, boxes, labels):
img, boxes = resize(img, boxes, size=(img_size, img_size))
img = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
])(img)
boxes, labels = box_coder.encode(boxes, labels)
return img, boxes, labels
testset = ListDataset(root='/usr/local/share/data/SynthText/',
list_file='data_val.txt',