def learning(train_set, dataset_path, lbl_train, neural_nets, nn_for_learn, indexes, debug=False):
if nn_for_learn[NET_12]:
if debug:
print("First network learning")
for i in range(0, indexes[NET_12]):
if debug:
print(i)
all_images, all_labels, coordinates = prepare_images.prepare(dataset_path + train_set[i].decode('utf8'),
lbl_train[i], debug=debug)
if debug:
print("Image prepared")
images = all_images[all_labels == 1]
labels = np.ones(images.shape[0])
neg_size = int(labels.shape[0] * NEGATIVE_MULTIPLIER)
neg_indexes = np.random.choice(np.arange(all_images.shape[0] * all_images.shape[1]),
neg_size, replace=False)
neg_indexes = np.unravel_index(neg_indexes, (all_images.shape[0], all_images.shape[1]))
neg_labels = all_labels[neg_indexes]
neg_images = all_images[neg_indexes]
images = np.concatenate((images, neg_images))
labels = np.concatenate((labels, neg_labels))
if debug:
print("images.shape, labels.shape")
print(images.shape, labels.shape)
neural_nets[NET_12].learning(dataset=convert48to12(images), labels=labels, debug_print=debug, n_epochs=5)
if nn_for_learn[NET_24]:
if debug:
print("Second network learning")
for i in range(indexes[NET_24]):
all_images, all_labels, coordinates = prepare_images.prepare(dataset_path + train_set[i].decode('utf8'),
lbl_train[i], debug=debug)
predicted_labels = neural_nets[NET_12].predict(all_images)
images = all_images[predicted_labels == 1]
labels = all_labels[predicted_labels == 1]
nn_for_learn[NET_24].learning(dataset=convert48to24(images), labels=labels, debug_print=debug, n_epochs=10)
if nn_for_learn[NET_48]:
if debug:
print("Second network learning")
for i in range(indexes[NET_48]):
all_images, all_labels, coordinates = prepare_images.prepare(dataset_path + train_set[i].decode('utf8'),
lbl_train[i], debug=debug)
predicted_labels = neural_nets[NET_12].predict(all_images)
images = all_images[predicted_labels == 1]
predicted_labels = neural_nets[NET_24].predict(images)
images = all_images[predicted_labels == 1]
labels = all_labels[predicted_labels == 1]
nn_for_learn[NET_48].learning(dataset=convert48to24(images), labels=labels, debug_print=debug, n_epochs=10)
def test_prepare(self):
dataset_path = "c:/_Hive/_diploma/LISA Traffic Sign Dataset/signDatabasePublicFramesOnly/vid0/frameAnnotations-vid_cmp2.avi_annotations/"
annotation_path = dataset_path + 'frameAnnotations.csv'
image_data = np.genfromtxt(annotation_path, delimiter=';', names=True, dtype=None)
files = dict()
for image in image_data:
filepath = image['Filename']
if filepath not in files:
img = Image(filepath)
img.add_sign(label=image['Annotation_tag'],
coordinates=(image['Upper_left_corner_X'], image['Upper_left_corner_Y'],
image['Lower_right_corner_X'], image['Lower_right_corner_Y']))
files[filepath] = img
else:
files[filepath].add_sign(label=image['Annotation_tag'],
coordinates=(image['Upper_left_corner_X'], image['Upper_left_corner_Y'],
image['Lower_right_corner_X'], image['Lower_right_corner_Y']))
images = np.array(list(files.keys()))
images.sort()
lbl = np.array([files.get(key).get_coordinates() for key in images])
print(images[0].decode('utf8'))
imgs, lbls = prepare_images.prepare(dataset_path + images[0].decode('utf8'), lbl[0])
test_img = cv2.imread(dataset_path + images[0].decode('utf8'), cv2.IMREAD_UNCHANGED)
# noinspection PyAugmentAssignment
test_img = test_img / 255
for j in range(lbls.shape[0]):
# Rectangle = namedtuple('Rectangle', ['xmin', 'ymin', 'xmax', 'ymax'])
(x1, y1, x2, y2) = (coords[j].xmin, coords[j].ymin, coords[j].xmax, coords[j].ymax)
test_img_roi = np.array(
[test_img[y1:y2, x1:x2, 0], test_img[y1:y2, x1:x2, 1], test_img[y1:y2, x1:x2, 2]])
# if j > 325:
# prepare_images.show_roi([imgs[j], test_img_roi])
self.assertTrue(np.allclose(imgs[j], test_img_roi), msg="In imgs[{}]".format(j))
def test_prepare(self):
dataset_path = "c:/_Hive/_diploma/LISA Traffic Sign Dataset/signDatabasePublicFramesOnly/vid0/frameAnnotations-vid_cmp2.avi_annotations/"
annotation_path = dataset_path + 'frameAnnotations.csv'
image_data = np.genfromtxt(annotation_path, delimiter=';', names=True, dtype=None)
files = dict()
for image in image_data:
filepath = image['Filename']
if filepath not in files:
img = Image(filepath)
img.add_sign(label=image['Annotation_tag'],
coordinates=(image['Upper_left_corner_X'], image['Upper_left_corner_Y'],
image['Lower_right_corner_X'], image['Lower_right_corner_Y']))
files[filepath] = img
else:
files[filepath].add_sign(label=image['Annotation_tag'],
coordinates=(image['Upper_left_corner_X'], image['Upper_left_corner_Y'],
image['Lower_right_corner_X'], image['Lower_right_corner_Y']))
images = np.array(list(files.keys()))
images.sort()
lbl = np.array([files.get(key).get_coordinates() for key in images])
print(images[0].decode('utf8'))
imgs, lbls, coords = prepare_images.prepare(dataset_path + images[0].decode('utf8'), lbl[0])
test_img = cv2.imread(dataset_path + images[0].decode('utf8'), cv2.IMREAD_UNCHANGED)
# noinspection PyAugmentAssignment
test_img = test_img / 255
for j in range(lbls.shape[0]):
# Rectangle = namedtuple('Rectangle', ['xmin', 'ymin', 'xmax', 'ymax'])
(x1, y1, x2, y2) = (coords[j].xmin, coords[j].ymin, coords[j].xmax, coords[j].ymax)
test_img_roi = np.array(
[test_img[y1:y2, x1:x2, 0], test_img[y1:y2, x1:x2, 1], test_img[y1:y2, x1:x2, 2]])
# if j > 325:
# prepare_images.show_roi([imgs[j], test_img_roi])
self.assertTrue(np.allclose(imgs[j], test_img_roi), msg="In imgs[{}]".format(j))
def testing_results(neural_nets, nn_params, nn_for_test, test_set_without_negatives, numbers_of_test_imgs=10):
first_net, second_net = neural_nets[:2]
first_net.load_params(nn_params[0])
if nn_for_test[1]:
second_net.load_params(nn_params[1])
print("Testing...")
test_img = np.array(list(test_set_without_negatives.keys())[:numbers_of_test_imgs])
lbl_test = np.array([test_set_without_negatives.get(key).get_coordinates() for key in test_img])
tp_all = np.zeros(numbers_of_test_imgs)
tn_all = np.zeros(numbers_of_test_imgs)
fp_all = np.zeros(numbers_of_test_imgs)
fn_all = np.zeros(numbers_of_test_imgs)
f1_score_all = np.zeros(numbers_of_test_imgs)
tn_percent_all = np.zeros(numbers_of_test_imgs)
for i in range(test_img.shape[0]):
imgs, lbls, coords = prepare_images.prepare(DATASET_PATH + test_img[i].decode('utf8'), lbl_test[i])
y_pred = np.zeros_like(lbls)
for j in range(imgs.shape[0]):
# TODO добавить nms в цепочку
y_pred[j] = first_net.predict(nn.convert48to24(imgs[j]))
if nn_for_test[1]:
tmp = imgs[y_pred == 1]
y_pred[y_pred == 1] = second_net.predict(tmp)
tmp = lbls - y_pred
tp = np.sum((y_pred == 1) & (lbls == 1))
tn = np.sum((y_pred == 0) & (lbls == 0))
fp = np.sum(tmp == -1)
fn = np.sum(tmp == 1)
f1_score = 2 * tp / (2 * tp + fn + fp)
tp_all[i] = tp
tn_all[i] = tn
fp_all[i] = fp
fn_all[i] = fn
f1_score_all[i] = f1_score
print(" f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score, tp, tn, fp, fn))
tn_percent = tn / (tn + fp) * 100
tn_percent_all[i] = tn_percent
print("True negative percent from all negatives = {}".format(tn_percent))
tmp = np.arange(lbls.shape[0] * lbls.shape[1]).reshape(lbls.shape)
tmp = tmp[y_pred == 1]
rects = [coords.get(key, None) for key in tmp]
prepare_images.save_img_with_rectangles(DATASET_PATH, test_img[i].decode('utf8'), rects)
# prepare_images.show_rectangles(dataset_path + test_img[i].decode('utf8'), rects, show_type='opencv')
# rects = prepare_images.nms(rects, 0.3)
# prepare_images.show_rectangles(dataset_path + test_img[i].decode('utf8'), rects, show_type='opencv')
print("f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score_all.mean(), tp_all.mean(), tn_all.mean(), fp_all.mean(), fn_all.mean()))
print("True negative percent from all negatives = {}".format(tn_percent_all.mean()))
return tp_all, tn_all, fp_all, fn_all, f1_score_all, tn_percent_all
def test_load_params(batch_size=45, random_state=123, init=False):
train_sets = test_init()
train_set_without_negatives = train_sets['train_set']
net = nn.Network(batch_size=batch_size, random_state=random_state)
test_img = np.array(list(train_set_without_negatives.keys()))
test_img.sort()
lbl_test = np.array([train_set_without_negatives.get(key).get_coordinates() for key in test_img])
for i in range(test_img.shape[0]):
imgs, lbls = prepare_images.prepare(DATASET_PATH + test_img[i].decode('utf8'), lbl_test[i])
y_pred = net.predict_values(imgs)
tmp = lbls - y_pred
tp = np.sum((y_pred == 1) & (lbls == 1))
tn = np.sum((y_pred == 0) & (lbls == 0))
fp = np.sum(tmp == -1)
fn = np.sum(tmp == 1)
f1_score = 2 * tp / (2 * tp + fn + fp)
print(" f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score, tp, tn, fp, fn))
def learning_localization_networks(train_set,
dataset_path,
lbl_train,
neural_nets,
debug=False):
if debug:
print("First network learning")
for i in range(0, neural_nets['net_12']['indexes']):
if debug:
print(i)
all_images, all_labels = prepare_images.prepare(
dataset_path + train_set[i].decode('utf8'),
lbl_train[i],
debug=debug)
if debug:
print("Image prepared")
images = all_images[all_labels == 1]
labels = np.ones(images.shape[0])
neg_size = int(labels.shape[0] * NEGATIVE_MULTIPLIER)
neg_indexes = np.random.choice(np.arange(all_images.shape[0] *
all_images.shape[1]),
neg_size,
replace=False)
neg_indexes = np.unravel_index(
neg_indexes, (all_images.shape[0], all_images.shape[1]))
neg_labels = all_labels[neg_indexes]
neg_images = all_images[neg_indexes]
images = np.concatenate((images, neg_images))
labels = np.concatenate((labels, neg_labels))
if debug:
print("images.shape, labels.shape")
print(images.shape, labels.shape)
neural_nets['net_12']['neural_net'].learning(
dataset=convert48to12(images),
labels=labels,
debug_print=debug,
n_epochs=5)
def test_load_params(batch_size=45, random_state=123, init=False):
train_sets = test_init()
train_set_without_negatives = train_sets['train_set']
net = nn.Network(batch_size=batch_size, random_state=random_state)
test_img = np.array(list(train_set_without_negatives.keys()))
test_img.sort()
lbl_test = np.array([
train_set_without_negatives.get(key).get_coordinates()
for key in test_img
])
for i in range(test_img.shape[0]):
imgs, lbls = prepare_images.prepare(
DATASET_PATH + test_img[i].decode('utf8'), lbl_test[i])
y_pred = net.predict_values(imgs)
tmp = lbls - y_pred
tp = np.sum((y_pred == 1) & (lbls == 1))
tn = np.sum((y_pred == 0) & (lbls == 0))
fp = np.sum(tmp == -1)
fn = np.sum(tmp == 1)
f1_score = 2 * tp / (2 * tp + fn + fp)
print(
" f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score, tp, tn, fp, fn))
def testing_results(neural_nets,
nn_params,
test_set_without_negatives,
numbers_of_test_imgs=10):
# neural_nets[0].load_params(nn_params[0])
print("Testing...")
test_img = np.array(
list(test_set_without_negatives.keys())[:numbers_of_test_imgs])
lbl_test = np.array([
test_set_without_negatives.get(key).get_coordinates()
for key in test_img
])
tp_all = np.zeros(numbers_of_test_imgs)
tn_all = np.zeros(numbers_of_test_imgs)
fp_all = np.zeros(numbers_of_test_imgs)
fn_all = np.zeros(numbers_of_test_imgs)
f1_score_all = np.zeros(numbers_of_test_imgs)
tn_percent_all = np.zeros(numbers_of_test_imgs)
for i in range(test_img.shape[0]):
imgs, lbls = prepare_images.prepare(
DATASET_PATH + test_img[i].decode('utf8'), lbl_test[i])
y_pred = np.zeros_like(lbls)
for j in range(imgs.shape[0]):
# TODO добавить nms в цепочку
y_pred[j] = neural_nets[0].predict(nn.convert48to12(imgs[j]))
# if nn_for_test[1]:
# tmp = imgs[y_pred == 1]
# y_pred[y_pred == 1] = neural_nets[1].predict(nn.convert48to24(tmp))
# if nn_for_test[2]:
# tmp = imgs[y_pred == 1]
# y_pred[y_pred == 1] = neural_nets[2].predict(tmp)
tmp = lbls - y_pred
tp = np.sum((y_pred == 1) & (lbls == 1))
tn = np.sum((y_pred == 0) & (lbls == 0))
fp = np.sum(tmp == -1)
fn = np.sum(tmp == 1)
f1_score = 2 * tp / (2 * tp + fn + fp)
tp_all[i] = tp
tn_all[i] = tn
fp_all[i] = fp
fn_all[i] = fn
f1_score_all[i] = f1_score
print(
" f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score, tp, tn, fp, fn))
tn_percent = tn / (tn + fp) * 100
tn_percent_all[i] = tn_percent
print(
"True negative percent from all negatives = {}".format(tn_percent))
tmp = np.arange(lbls.shape[0] * lbls.shape[1]).reshape(lbls.shape)
tmp = tmp[y_pred == 1]
# rects = [coords.get(key, None) for key in tmp]
# prepare_images.save_img_with_rectangles(DATASET_PATH, test_img[i].decode('utf8'), rects)
# prepare_images.show_rectangles(dataset_path + test_img[i].decode('utf8'), rects, show_type='opencv')
# rects = prepare_images.nms(rects, 0.3)
# prepare_images.show_rectangles(dataset_path + test_img[i].decode('utf8'), rects, show_type='opencv')
print(
"f1 score = {}, true positive = {}, true negative = {}, false positive = {}, false negative = {}"
.format(f1_score_all.mean(), tp_all.mean(), tn_all.mean(),
fp_all.mean(), fn_all.mean()))
print("True negative percent from all negatives = {}".format(
tn_percent_all.mean()))
return tp_all, tn_all, fp_all, fn_all, f1_score_all, tn_percent_all