def predict_image_class(query_folder, target_features, cache_dict, class_minimum):
minimum = (float('inf'), 0)
for query_image_path in os.listdir(os.path.join(query_path, query_folder)):
t = time.time()
query = os.path.join(query_path, query_folder, query_image_path)
t1 = time.time() - t
cache_dict, query_features = memory_cache(cache_dict, model.model, query, os.path.join(cache_path, query_folder, query_image_path + '.pth'), transform)
t2 = time.time() - t
y = LSHash.euclidean_dist(target_features.cpu().numpy()[0], query_features.cpu().numpy()[0])
t3 = time.time() - t
print(t1, t2, t3)
if y < minimum[0]:
minimum = (y, query_folder)
class_minimum[query_folder] = minimum
minimum = (float('inf'), 0)
with torch.no_grad():
target_image_ori = Image.fromarray(draw[b[1]:b[3],
b[0]:b[2]])
target_image = transform(target_image_ori)
x = torch.zeros((1, 3, 224, 224))
x[0] = target_image
target_features = model.model._forward_impl(x.cuda())
for query_folder in os.listdir(query_path):
for query_image_path in os.listdir(
os.path.join(query_path, query_folder)):
query = os.path.join(query_path, query_folder,
query_image_path)
cache_dict, query_features = memory_cache(
cache_dict, model.model, query,
os.path.join(cache_path, query_folder,
query_image_path + '.pth'),
transform)
y = LSHash.euclidean_dist(
target_features.cpu().numpy()[0],
query_features.cpu().numpy()[0])
if y < minimum[0]:
minimum = (y, query_folder)
if minimum[0] > 1:
minimum = (minimum[0], 'obj')
# print(minimum)
obj = ET.SubElement(root, 'object')
ET.SubElement(obj, 'name').text = minimum[1]
bndbx = ET.SubElement(obj, 'bndbox')
ET.SubElement(bndbx, 'xmin').text = str(b[0])
ET.SubElement(bndbx, 'ymin').text = str(b[1])