def generateAnchors(train_annotation_folder, train_image_folder,
train_cache_file, model_labels):
print("Generating anchor boxes for training images and annotation...")
num_anchors = 9
train_imgs, train_labels = parse_voc_annotation(train_annotation_folder,
train_image_folder,
train_cache_file,
model_labels)
# run k_mean to find the anchors
annotation_dims = []
for image in train_imgs:
for obj in image['object']:
relative_w = (float(obj['xmax']) -
float(obj['xmin'])) / image['width']
relatice_h = (float(obj["ymax"]) -
float(obj['ymin'])) / image['height']
annotation_dims.append(tuple(map(float, (relative_w, relatice_h))))
annotation_dims = np.array(annotation_dims)
centroids = run_kmeans(annotation_dims, num_anchors)
# write anchors to file
print('Average IOU for', num_anchors, 'anchors:',
'%0.2f' % avg_IOU(annotation_dims, centroids))
anchors = centroids.copy()
widths = anchors[:, 0]
sorted_indices = np.argsort(widths)
anchor_array = []
reverse_anchor_array = []
out_string = ""
r = "anchors: ["
for i in sorted_indices:
anchor_array.append(int(anchors[i, 0] * 416))
anchor_array.append(int(anchors[i, 1] * 416))
out_string += str(int(anchors[i, 0] * 416)) + ',' + str(
int(anchors[i, 1] * 416)) + ', '
reverse_count = len(anchor_array) - 1
while (reverse_count > -1):
reverse_anchor_array.append(anchor_array[reverse_count])
reverse_count -= 1
print("Anchor Boxes generated.")
return anchor_array, reverse_anchor_array
def _main_(args):
config_path = args.conf
with open(config_path) as config_buffer:
config = json.loads(config_buffer.read())
###############################
# Create the validation generator
###############################
valid_ints, labels = parse_voc_annotation(
config['valid']['valid_annot_folder'],
config['valid']['valid_image_folder'], config['valid']['cache_name'],
config['model']['labels'])
labels = labels.keys() if len(
config['model']['labels']) == 0 else config['model']['labels']
labels = sorted(labels)
valid_generator = BatchGenerator(
instances=valid_ints,
anchors=config['model']['anchors'],
labels=labels,
downsample=
32, # ratio between network input's size and network output's size, 32 for YOLOv3
max_box_per_image=0,
batch_size=config['train']['batch_size'],
min_net_size=config['model']['min_input_size'],
max_net_size=config['model']['max_input_size'],
shuffle=True,
jitter=0.0,
norm=normalize)
###############################
# Load the model and do evaluation
###############################
os.environ['CUDA_VISIBLE_DEVICES'] = config['train']['gpus']
infer_model = load_model(config['train']['saved_weights_name'])
# compute mAP for all the classes
average_precisions = evaluate(infer_model, valid_generator)
# print the score
for label, average_precision in average_precisions.items():
print(labels[label] + ': {:.4f}'.format(average_precision))
print('mAP: {:.4f}'.format(
sum(average_precisions.values()) / len(average_precisions)))
