def detect(self, image):
cuda.select_device(0)
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
ROOT_DIR = "/home/bernihoh/Bachelor/SMS/MaskRCNN/samples/SMSNetworks/face_feature_detection/"
MODEL_DIR = os.path.join(ROOT_DIR, "logsFaceFeatureDetection")
COCO_MODEL_PATH = "/home/bernihoh/Bachelor/SMS/MaskRCNN/samples/SMSNetworks/face_feature_detection/mask_rcnn_face_feature_detection_0029.h5"
config = InferenceConfig()
config.display()
# Create model object in inference mode.
model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config)
# Load weights trained on MS-COCO
model.load_weights(COCO_MODEL_PATH, by_name=True)
class_names = ["bg", "iris_l", "inner_eye_l", "outer_eye_l", "eye_brow_l", "cheek_l", "iris_r",
"inner_eye_r", "outer_eye_r", "eye_brow_r", "cheek_r", "nose_tip", "nose", "mouth",
"chin", "face", "head", "distortion"]
results = model.detect([image], verbose=1)
r = results[0]
session.close()
cuda.close()
return r
def train(self,
sess: tf.InteractiveSession,
train_x: np.ndarray,
train_y: np.ndarray,
valid_x: np.ndarray,
valid_y: np.ndarray,
epoch=20,
batch_size=128,
valid_batch_size=50,
step=200,
verbose=True):
print(get_now(), 'start training')
train_idx = sorted(range(len(train_x)),
key=lambda x: len(train_x[x]),
reverse=True)
valid_idx = sorted(range(len(valid_x)),
key=lambda x: len(valid_x[x]),
reverse=True)
sess.run(tf.global_variables_initializer())
best_aupr = best_auc = current = 0
for idx_epoch in range(epoch):
for i in range(0, len(train_idx), batch_size):
batch_idx = train_idx[i:i + batch_size]
train_loss = self.train_step(sess, train_x[batch_idx],
train_y[batch_idx])
current += 1
if current % step == 0:
print(get_now())
valid_loss, valid_res = 0, np.empty([len(valid_idx), 1],
dtype=int)
for j in range(0, len(valid_idx), valid_batch_size):
valid_batch_idx = valid_idx[j:j + valid_batch_size]
loss, output = self.valid_step(
sess, valid_x[valid_batch_idx],
valid_y[valid_batch_idx])
valid_res[valid_batch_idx] = output
valid_loss += loss * len(valid_batch_idx)
valid_loss /= len(valid_idx)
auc, aupr = get_auc(valid_y, valid_res), get_aupr(
valid_y, valid_res)
if aupr > best_aupr:
best_aupr = aupr
best_auc = auc
self.saver.save(sess, self.model_path)
if verbose:
print(get_now(), current, current * batch_size,
idx_epoch, i + batch_size, 'train loss:',
round(train_loss, 5), 'valid loss:',
round(valid_loss, 5), 'AUC:', round(auc,
5), 'AUPR:',
round(aupr, 5))
print(get_now(), 'Summary ', 'Best AUC:', best_auc, 'Best AUPR:',
best_aupr)
def get_gradient(network: keras.models.Sequential, X: numpy.array, session: tensorflow.InteractiveSession) -> numpy.array:
"""
Calculates the gradient for a given network/input
:param network: Network
:type network: keras.Sequential
:param X: Input to network.
:type X: numpy.array
:param session: Session object
:type session: tensorflow.InteractiveSession
:returns: Gradient as numpy.array
"""
x = tensorflow.placeholder(tensorflow.float32, X.shape)
prediction = network(x)
y_shape = tensorflow.shape(prediction)
classes = y_shape[1]
index = tensorflow.argmax(prediction, axis=1)
target = tensorflow.one_hot(index, classes, on_value=1.0, off_value=0.0)
logits, = prediction.op.inputs
loss = tensorflow.nn.softmax_cross_entropy_with_logits(labels=target, logits=logits)
gradient, = tensorflow.gradients(loss, x)
return session.run(gradient, feed_dict={x: X, keras.backend.learning_phase(): 0})
def matrix_to_text(token, x):
print('Converting to text vector...')
reverse_word_dict = dict(map(reversed, token.word_index.items()))
InteractiveSession()
seqs_to_words = lambda y: list(
map(reverse_word_dict.get,
argmax(y, axis=-1).eval()))
return seqs_to_words(x)
def create_face_net_dict_files_for_train_val(picture_path, pic_file):
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
pic = skimage.io.imread(picture_path + pic_file)
ffn = FFN(pic)
ffn_results = ffn.detect()
session.close()
if ffn_results == "redo":
print("redo")
return "Failure"
else:
output = open(
"/home/bernihoh/Bachelor/SMS/modification/netcompare/services/face_net/face_net_pkl_val/"
+ pic_file + ".pkl", "wb")
pkl.dump(ffn_results["face_score_input"], output)
output.close()
return "Success"
def define_video_config():
"""
Определяются параметры использования видеокарты
"""
# Определяются параметры использования видеокарты
config_proto = ConfigProto()
# config_proto.gpu_options.per_process_gpu_memory_fraction = 0.8
config_proto.gpu_options.allow_growth = True
session = InteractiveSession(config=config_proto)
def prepare_environment():
np.random.seed(1)
random.seed(1)
from tensorflow import ConfigProto
from tensorflow import InteractiveSession
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
def __init__(self, image_size=[28, 28], char_number=10, channel=1):
self.image_size = image_size
self.char_number = char_number
self.channel = channel
# self.learning_rate = learning_rate
# self.eval_every = eval_every
# self.epochs = epochs
# self.evaluation_size = evaluation_size
# self.batch_size = batch_size
# self.optimizer = optimizer
self.inputs = placeholder(float32,
shape=[None, image_size[0] * image_size[1]],
name="inputs")
self.model = reshape(self.inputs,
[-1, image_size[0], image_size[1], channel])
self.labels = placeholder(float32, shape=[None, 10], name="labels")
self.sess = InteractiveSession()
def sequences_to_text(token, x):
print('Converting to text...')
reverse_word_dict = dict(map(reversed, token.word_index.items()))
InteractiveSession()
from_categorical = lambda y: argmax(y, axis=-1).eval()
seqs_to_words = lambda y: list(
map(reverse_word_dict.get, from_categorical(y)))
words_to_sentence = lambda y: ' '.join(filter(None, y))
word_matrix = list(map(seqs_to_words, x))
sentence_list = list(map(words_to_sentence, word_matrix))
return '\n'.join(sentence_list)
def predict_step(self, sess: tf.InteractiveSession, data_x: np.ndarray):
batch_num, res = len(data_x), []
for i in range(batch_num):
feed_dict = {
self.data_smi: data_x[i][0],
self.data_adj: data_x[i][1],
self.data_seq: data_x[i][2],
self.training: False
}
output = sess.run(self.output, feed_dict=feed_dict)
res.append(output[0])
return res
def detect(self):
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
self.MODEL_DIR = self.model_dir()
self.NETWORK_PATH = self.network_path()
self.InferenceConfig = self.inference_config()
self.InferenceConfig.display()
self.model = modellib.MaskRCNN(mode="inference",
model_dir=self.MODEL_DIR,
config=self.InferenceConfig)
self.model.load_weights(self.NETWORK_PATH, by_name=True)
results = self.model.detect([self.image], verbose=1)
"""r looks like this: r['rois'], r['masks'], r['class_ids'], r['scores']"""
r = results[0]
class_names = self.class_names()
visualize.display_instances(self.image, r['rois'], r['masks'],
r['class_ids'], class_names, r['scores'])
session.close()
ret = self.further_image_manipulation(r)
return ret
def train_step(self, sess: tf.InteractiveSession, train_x: np.ndarray,
train_y: np.ndarray):
batch_loss, batch_num = 0, len(train_x)
for i in range(batch_num):
batch_x, batch_y = train_x[i], train_y[i]
feed_dict = {
self.data_smi: batch_x[0],
self.data_adj: batch_x[1],
self.data_seq: batch_x[2],
self.data_y: np.array(batch_y).reshape([-1, 1]),
self.training: True
}
_, loss = sess.run([self.optimizer, self.loss],
feed_dict=feed_dict)
batch_loss += loss / batch_num
return batch_loss
def valid_step(self, sess: tf.InteractiveSession, valid_x: np.ndarray,
valid_y: np.ndarray):
batch_loss, batch_num, res = 0, len(valid_x), []
for i in range(batch_num):
batch_x, batch_y = valid_x[i], valid_y[i]
feed_dict = {
self.data_smi: batch_x[0],
self.data_adj: batch_x[1],
self.data_seq: batch_x[2],
self.data_y: np.array(batch_y).reshape([-1, 1]),
self.training: False
}
loss, output = sess.run([self.loss, self.output],
feed_dict=feed_dict)
batch_loss += loss / batch_num
res.append(output[0])
return batch_loss, res
def detect(self, image_path, background_color):
cuda.select_device(0)
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
ROOT_DIR = os.path.abspath("/")
MODEL_DIR = os.path.join(ROOT_DIR, "logs")
sys.path.append(ROOT_DIR) # To find local version of the library
sys.path.append(os.path.join(ROOT_DIR, "samples/coco/")) # To find local version
COCO_MODEL_PATH = "/home/bernihoh/Bachelor/MaskRCNN/mask_rcnn_coco.h5"
if not os.path.exists(COCO_MODEL_PATH):
utils.download_trained_weights(COCO_MODEL_PATH)
config = InferenceConfig()
config.display()
model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config)
model.load_weights(COCO_MODEL_PATH, by_name=True)
COCO_DIR = "/home/bernihoh/Bachelor/MaskRCNN/samples/coco" # geändert: Zeile eingefügt
dataset = coco.CocoDataset()
dataset.load_coco(COCO_DIR, "train")
dataset.prepare()
class_names = ['BG', 'person', 'bicycle', 'car', 'motorcycle', 'airplane',
'bus', 'train', 'truck', 'boat', 'traffic light',
'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird',
'cat', 'dog', 'horse', 'sheep', 'cow', 'elephant', 'bear',
'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie',
'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball',
'kite', 'baseball bat', 'baseball glove', 'skateboard',
'surfboard', 'tennis racket', 'bottle', 'wine glass', 'cup',
'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple',
'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza',
'donut', 'cake', 'chair', 'couch', 'potted plant', 'bed',
'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
'keyboard', 'cell phone', 'microwave', 'oven', 'toaster',
'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors',
'teddy bear', 'hair drier', 'toothbrush']
image = skimage.io.imread(image_path)
print(dataset.class_names)
plt.imshow(image)
# Run detection
results = model.detect([image], verbose=1)
# Visualize results
r = results[0]
visualize.display_instances(image, r['rois'], r['masks'], r['class_ids'], class_names, r['scores'])
used_class = r["class_ids"]
print(used_class)
mask = r["masks"]
mask = mask.astype(np.ubyte)
# maskimg = mask[:, :, 1] ^ mask[:, :, 1]
maskimg = np.zeros((image.shape[0], image.shape[1]))
maskimg = maskimg.astype(np.ubyte)
background_mask = np.full((image.shape[0], image.shape[1]), background_color, dtype=np.uint8)
skimage.io.imshow(background_mask)
plt.show()
for i in range(mask.shape[2]):
# skimage.io.imshow(mask[:, :, i])
# plt.show()
# maskimg = maskimg | mask[:, :, i]
a = used_class[i] - 1
if used_class[i] - 1 < 0:
a = 0
background_mask = background_mask - mask[:, :, i] * (a + 127)
maskimg = np.maximum(maskimg, mask[:, :, i] * a)
skimage.io.imshow(background_mask)
plt.show()
maskimg = np.maximum(maskimg, background_mask)
# maskimg[maskimg == 0] = 124
# maskimg = skimage.exposure.rescale_intensity(maskimg)
skimage.io.imshow(maskimg)
plt.show()
# skimage.io.imsave("/home/bernihoh/Bachelor/MaskRCNN/ownimages/mask138-1.jpg", maskimg)
session.close()
cuda.close()
return maskimg
from dataset.dataset_factory import DataSetFactory, CocoSetFactory
# Добавляется в поиск библиотека MaskRCNN
add_mask_rcnn(sys.argv[0])
from mask_rcnn.mrcnn.model import MaskRCNN
# import imgaug
# START_MODEL_PATH = 'start_mask/mask_rcnn_start.h5'
START_MODEL_PATH = 'mask_rcnn_detect_cfg_0005.h5'
COCO_PATH = '/home/ids/coco'
CLASS_IDS = [1, 2]
# Определяются параметры использования видеокарты
config_proto = ConfigProto()
# config_proto.gpu_options.per_process_gpu_memory_fraction = 0.8
config_proto.gpu_options.allow_growth = True
session = InteractiveSession(config=config_proto)
# Подготавливаем тренировочный и тестовый набор данных
# train_set = DataSetFactory.new_instance('data/img')
# test_set = DataSetFactory.new_instance('data/val')
train_set = CocoSetFactory.new_instance(COCO_PATH, 'train', 2017, CLASS_IDS)
test_set = CocoSetFactory.new_instance(COCO_PATH, 'val', 2017, CLASS_IDS)
# Определяется конфигурация
config = DetectConfig()
config.display()
# Определяется модель
model = MaskRCNN(mode='training', model_dir='./', config=config)
# Произвести загрузку стартовой модели
model.load_weights(START_MODEL_PATH,
by_name=True,
def main(argv=None):
# EXAMPLES
argparser = argparse.ArgumentParser(description='A visual question answerer.')
# optional arguments
argparser.add_argument('--load', type=str,
help='Filename of existing model, default None')
argparser.add_argument('--test', type=str,
help='Filename of test data, default qa.894.raw.test.txt')
argparser.add_argument('--q', type=str,
help='Pose one question for the visual question answerer')
argparser.add_argument('--e', type=int,
help='Number of epochs, default 1')
argparser.add_argument('--ld1', type=int,
help='Latent dimension 1, default 512')
argparser.add_argument('--b', type=int,
help='Batch size, default 32')
argparser.add_argument('--drop', type=float,
help='Dropout percentage, default 0.5')
argparser.add_argument('--aeweight', type=float,
help='Weight of the autoencoder loss function compared to the answer loss function, default 1.0')
argparser.add_argument('--wups', action="store_true",
help='Compute the WUPS Score')
argparser.add_argument('--textonly', action="store_true",
help='Ignore the images')
argparser.add_argument('--visualonly', action="store_true",
help='Without autoencoder')
argparser.add_argument('--improve', action="store_true",
help='Further train the loaded model')
argparser.add_argument('--checkpoint', action="store_true",
help='Save at every epoch')
args = argparser.parse_args(argv)
# Hyper Parameters
ld1 = 512
epochs = 1
batch = 32
drop = 0.5
ae_weight = 1.0
test = "qa.894.raw.test.txt"
if args.ld1:
ld1 = args.ld1
if args.e:
epochs = args.e
if args.b:
batch = args.b
if args.drop:
drop = args.drop
if args.aeweight:
ae_weight = args.aeweight
if args.test:
test = args.test
print('--e Number of epochs: ' + str(epochs))
print('--ld1 Latent dimension 1: ' + str(ld1))
print('--b Batch size: ' + str(batch))
print('--drop Dropout percentage: ' + str(drop))
print('--aeweight Autoencoder Loss Weight: ' + str(ae_weight))
print('')
InteractiveSession()
test_x, test_t, qa_answer, qa_question, train_token = model(epochs,drop,ae_weight,ld1,batch,
args.load,args.q,test,args.textonly,args.visualonly,args.improve,args.checkpoint)
nbtest = len(test_t)
if args.wups:
postp.print_wups_acc(test_t,qa_answer,train_token)
if not args.visualonly:
postp.print_ae_acc(test_x,qa_question,train_token)
postp.print_compare(test_x,test_t,qa_answer,qa_question,nbtest,train_token)
import sys
import pickle
from optparse import OptionParser
import time
from keras_frcnn import config
from keras import backend as K
from keras.layers import Input
from keras.models import Model
from keras_frcnn import roi_helpers
from keras.applications.mobilenet import preprocess_input
from tensorflow import ConfigProto
from tensorflow import InteractiveSession
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
sys.setrecursionlimit(40000)
parser = OptionParser()
parser.add_option("-p", "--path", dest="test_path", help="Path to test data.")
parser.add_option("-n", "--num_rois", type="int", dest="num_rois",
help="Number of ROIs per iteration. Higher means more memory use.", default=32)
parser.add_option("--config_filename", dest="config_filename", help=
"Location to read the metadata related to the training (generated when training).",
default="config.pickle")
parser.add_option("--network", dest="network", help="Base network to use. Supports vgg or resnet50.", default='resnet50')
parser.add_option("--write", dest="write", help="to write out the image with detections or not.", action='store_true')
parser.add_option("--load", dest="load", help="specify model path.", default=None)
(options, args) = parser.parse_args()
class CNNModel:
def __init__(self, image_size=[28, 28], char_number=10, channel=1):
self.image_size = image_size
self.char_number = char_number
self.channel = channel
# self.learning_rate = learning_rate
# self.eval_every = eval_every
# self.epochs = epochs
# self.evaluation_size = evaluation_size
# self.batch_size = batch_size
# self.optimizer = optimizer
self.inputs = placeholder(float32,
shape=[None, image_size[0] * image_size[1]],
name="inputs")
self.model = reshape(self.inputs,
[-1, image_size[0], image_size[1], channel])
self.labels = placeholder(float32, shape=[None, 10], name="labels")
self.sess = InteractiveSession()
def addLayer(self, layer):
self.model = layer.implement(self.model)
def addOutputLayer(self, layer):
self.loss = layer.implement(self.model, self.labels)
def train(self,
dataset,
eval_every=5,
epochs=500,
evaluation_size=500,
batch_size=100,
optimizer=train.MomentumOptimizer(0.005, 0.9)):
train_step = optimizer.minimize(self.loss)
prediction = argmax(self.model, 1, name="prediction")
result = equal(argmax(self.labels, 1), prediction, name="result")
self.accuracy = reduce_mean(cast(result, float32), name="accuracy")
train_loss = []
train_accuracy = []
test_accuracy = []
global_variables_initializer().run()
for i in range(epochs):
# Lay ra batch_size hinh anh tu tap train
train_batch = dataset.train.next_batch(batch_size)
train_dict = {
self.inputs: train_batch[0],
self.labels: train_batch[1]
}
if i % eval_every == 0:
# Cu sau eval_every buoc lap thi test mot lan
test_batch = dataset.test.next_batch(evaluation_size)
temp_train_loss = self.loss.eval(feed_dict=train_dict)
temp_train_accuracy = self.accuracy.eval(feed_dict=train_dict)
temp_test_accuracy = self.accuracy.eval(
feed_dict={
self.inputs: test_batch[0],
self.labels: test_batch[1]
})
print(
'Epoch # %d, Train Loss: %g. Train Accuracy (Test Accuracy): %g (%g)'
% (i, temp_train_loss, temp_train_accuracy,
temp_test_accuracy))
# Luu cac gia tri de ve bieu do
train_loss.append(temp_train_loss)
train_accuracy.append(temp_train_accuracy)
test_accuracy.append(temp_test_accuracy)
# Chay thuat toan toi uu ham mat mat
self.sess.run(train_step, feed_dict=train_dict)
# Show plots
loss_per_epoch(epochs, eval_every, train_loss)
train_test_accuracy(epochs, eval_every, train_accuracy, test_accuracy)
def test(self, dataset):
"""
test model with entire mnist_test
"""
print('test accuracy %g' % self.accuracy.eval(
feed_dict={
self.inputs: dataset.test.images,
self.labels: dataset.test.labels,
}))
def save(self, model_path):
"""
save model to folder in model_path
"""
saver = train.Saver()
saver.save(self.sess, model_path)