def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
# Read data and apply label preserving distortions
image_batch, label_batch = facenet.read_and_augument_data(
image_list, label_list, args.image_size, args.batch_size,
args.max_nrof_epochs, args.random_crop, args.random_flip,
args.nrof_preprocess_threads)
print('Total number of classes: %d' % len(train_set))
print('Total number of examples: %d' % len(image_list))
# Node for input images
image_batch.set_shape((None, args.image_size, args.image_size, 3))
image_batch = tf.identity(image_batch, name='input')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Build the inference graph
prelogits, _ = network.inference(image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
weight_decay=args.weight_decay)
with tf.variable_scope('Logits'):
n = int(prelogits.get_shape()[1])
m = len(train_set)
w = tf.get_variable(
'w',
shape=[n, m],
dtype=tf.float32,
initializer=tf.truncated_normal_initializer(stddev=0.1),
regularizer=slim.l2_regularizer(args.weight_decay),
trainable=True)
b = tf.get_variable('b', [m], initializer=None, trainable=True)
logits = tf.matmul(prelogits, w) + b
# Add DeCov regularization loss
if args.decov_loss_factor > 0.0:
logits_decov_loss = facenet.decov_loss(
logits) * args.decov_loss_factor
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
logits_decov_loss)
# Add center loss
update_centers = tf.no_op('update_centers')
if args.center_loss_factor > 0.0:
prelogits_center_loss, update_centers = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.scalar_summary('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
logits, label_batch, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.all_variables(), args.log_histograms)
# Create a saver
save_variables = list(set(tf.all_variables()) - set([w]) - set([b]))
saver = tf.train.Saver(save_variables, max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.initialize_all_variables())
sess.run(tf.initialize_local_variables())
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if pretrained_model:
saver.restore(sess, pretrained_model)
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, phase_train_placeholder,
learning_rate_placeholder, global_step, total_loss,
train_op, summary_op, summary_writer,
regularization_losses, args.learning_rate_schedule_file,
update_centers)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
start_time = time.time()
_, _, accuracy, val, val_std, far = lfw.validate(
sess,
lfw_paths,
actual_issame,
args.seed,
args.batch_size,
image_batch,
phase_train_placeholder,
embeddings,
nrof_folds=args.lfw_nrof_folds)
print('Accuracy: %1.3f+-%1.3f' %
(np.mean(accuracy), np.std(accuracy)))
print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' %
(val, val_std, far))
lfw_time = time.time() - start_time
# Add validation loss and accuracy to summary
summary = tf.Summary()
#pylint: disable=maybe-no-member
summary.value.add(tag='lfw/accuracy',
simple_value=np.mean(accuracy))
summary.value.add(tag='lfw/val_rate', simple_value=val)
summary.value.add(tag='time/lfw', simple_value=lfw_time)
summary_writer.add_summary(summary, step)
with open(os.path.join(log_dir, 'lfw_result.txt'),
'at') as f:
f.write('%d\t%.5f\t%.5f\n' %
(step, np.mean(accuracy), val))
return model_dir
def main(args):
network = importlib.import_module(
args.model_def) # --model_def models.inception_resnet_v1
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir),
subdir) # 日志的地址
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir),
subdir) # 训练模型存储地址
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
# 读取训练图片的路径
if args.filter_filename:
train_set = filter_dataset(train_set,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
# train set中是元组，一个标签，对应一个或多个路径
# 所以这里返回的nrof_classes是样本类别个数
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
# global_step对应的是全局批的个数，根据这个参数可以更新学习率
global_step = tf.Variable(0, trainable=False)
# global_step经常在滑动平均，学习速率变化的时候需要用到,系统会自动更新这个参数的值，从1开始。
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
# range_size的大小和样本个数一样
range_size = array_ops.shape(labels)[0]
# QueueRunner：保存的是队列中的入列操作，保存在一个list当中，其中每个enqueue运行在一个线程当中
# range_input_producer：返回的是一个队列，队列中有打乱的整数，范围是从0到range size
# range_size大小和总的样本个数一样
# 并将一个QueueRunner添加到当前图的QUEUE_RUNNER集合中
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
# 返回的是一个出列操作，每次出列一个epoch中需要用到的样本个数
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
# tf.placeholder：用于得到传递进来的真实的训练样本：
# 不必指定初始值，可在运行时，通过 Session.run的函数的feed_dict参数指定；
# 学习率
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
# 批大小
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
# 用于判断是训练还是测试
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# 图像路径
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
# 图像标签
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
# 上面的队列是一个样本索引的出列队列，只用来出列
# 用来每次出列一个epoch中需要用到的样本
# 这里是第二个队列，这个队列用来入列，队列的大小为10万，每个元素的大小为shapes=[(1,), (1,)],
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
# 这时一个入列的操作，这个操作将在session run的时候用到
# 每次入列的是image_paths_placeholder, labels_placeholder对
# 注意，这里只有2个队列，一个用来出列打乱的元素序号
# 一个根据对应的需要读取指定的文件
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
# 4个线程
# 在不同的线程中入列不同的tensor需要入列的样本在images_and_labels中
# 创建的线程个数为len(images_and_labels)
# 在这里应该是有4个入列线程，因为images_and_labels只append4次
# 线程i入列张量images_and_labels[i]
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
# batch_join的作用是创建样本批，用于批处理
# capacity控制着用于增长队列的预取的个数
# batch_size用于出列的一个批的大小
# enqueue_many表示一次出列多个数据
# shapes：样本的shape，默认根据images_and_labels[i]推断出来
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
# 创建网络图:除了全连接层和损失层
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss --center_loss_factor 1e-2
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
# 将center加入到名字为tf.GraphKeys.REGULARIZATION_LOSSES的集合当中来
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
# 将指数衰减应用到学习率上
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
# 将softmax和交叉熵一起做,得到最后的损失函数,提高效率
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
# tf.reduce_mean(x) ==> 2.5 #如果不指定第二个参数，那么就在所有的元素中取平均值
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
# 根据REGULARIZATION_LOSSES返回一个收集器中所收集的值的列表
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# 我们选择L2-正则化来实现这一点，L2正则化将网络中所有权重的平方和加到损失函数。如果模型使用大权重，则对应重罚分，并且如果模型使用小权重，则小罚分。
# 这就是为什么我们在定义权重时使用了regularizer参数，并为它分配了一个l2_regularizer。这告诉了TensorFlow要跟踪
# l2_regularizer这个变量的L2正则化项（并通过参数reg_constant对它们进行加权）。
# 所有正则化项被添加到一个损失函数可以访问的集合——tf.GraphKeys.REGULARIZATION_LOSSES。
# 将所有正则化损失的总和与先前计算的triplet_loss相加，以得到我们的模型的总损失。
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
# 确定优化方法并求根据损失函数求梯度,在这里面,每更新一次参数,global_step会加1
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
set_A_vars = [
v for v in tf.trainable_variables()
if v.name.startswith('InceptionResnetV1')
] # 加载官方预训练模型
saver_set_A = tf.train.Saver(set_A_vars,
max_to_keep=200) # 加载官方预训练模型，一共加三行代码
# Create a saver
# 创建一个saver用于保存或从内存中恢复一个模型参数
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=200)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
# 能够在gpu上分配的最大内存
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver_set_A.restore(sess, pretrained_model) # 加载官方预训练模型
# saver.restore(sess, pretrained_model)#可以加载以前训练的softmax模型
# Training and validation loop，max_nrof_epochs = 80，epoch_size = 默认default=1000
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
# step可以看做是全局的批处理个数
step = sess.run(global_step, feed_dict=None)
# epoch_size是一个epoch中批的个数
# 这个epoch是全局的批处理个数除以一个epoch中批的个数得到epoch
epoch = step // args.epoch_size # 双斜杠为取整的5//3 = 1
# Train for one epoch
train(args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already,存储模型
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, embeddings, label_batch,
lfw_paths, actual_issame, args.lfw_batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer)
return model_dir
def main(argv=None): # pylint: disable=unused-argument
if FLAGS.model_name:
subdir = FLAGS.model_name
preload_model = True
else:
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
preload_model = False
log_dir = os.path.join(os.path.expanduser(FLAGS.logs_base_dir), subdir)
model_dir = os.path.join(os.path.expanduser(FLAGS.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.mkdir(model_dir)
np.random.seed(seed=FLAGS.seed)
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, validation_set = facenet.split_dataset(dataset, FLAGS.train_set_fraction, FLAGS.split_mode)
print('Model directory: %s' % model_dir)
with tf.Graph().as_default():
tf.set_random_seed(FLAGS.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(FLAGS.batch_size, FLAGS.image_size, FLAGS.image_size, 3),
name='input')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Build the inference graph
embeddings = facenet.inference_nn4_max_pool_96(images_placeholder, phase_train=phase_train_placeholder)
# Split example embeddings into anchor, positive and negative
anchor, positive, negative = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(anchor, positive, negative)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, _ = facenet.train(loss, global_step)
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=0)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
with sess.as_default():
if preload_model:
ckpt = tf.train.get_checkpoint_state(model_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
else:
raise ValueError('Checkpoint not found')
# Training and validation loop
for epoch in range(FLAGS.max_nrof_epochs):
# Train for one epoch
step = train(sess, train_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, train_op, summary_op, summary_writer)
# Validate epoch
validate(sess, validation_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, train_op, summary_op, summary_writer)
# Save the model checkpoint after each epoch
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
graphdef_dir = os.path.join(model_dir, 'graphdef')
graphdef_filename = 'graph_def.pb'
if (not os.path.exists(os.path.join(graphdef_dir, graphdef_filename))):
print('Saving graph definition')
tf.train.write_graph(sess.graph_def, graphdef_dir, graphdef_filename, False)
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
#train_set = facenet.get_dataset(args.data_dir)
train_set = facenet.get_huge_dataset(args.data_dir, args.trainset_start,
args.trainset_end)
nrof_classes = len(train_set)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
if args.filter_filename:
print('Filtering...')
train_set = filter_dataset(train_set, args.filter_filename,
args.filter_percentile,
args.filter_min_nrof_images_per_class,
args.trainset_start)
nrof_classes = len(train_set)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
meta_file, ckpt_file = facenet.get_model_filenames(
os.path.expanduser(args.pretrained_model))
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unpack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_png(file_contents)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
#image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, centers, _, centers_cts_batch_reshape = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
#prelogits_center_loss, _ = facenet.center_loss_similarity(prelogits, label_batch, args.center_loss_alfa, nrof_classes) ####Similarity cosine distance, center loss
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
logits, label_batch, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
#### Training accuracy of softmax: check the underfitting or overfiting #############################
correct_prediction = tf.equal(tf.argmax(logits, 1), label_batch)
train_acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
########################################################################################################
########## edit mzh #####################
# Create list with variables to restore
restore_vars = []
update_gradient_vars = []
if args.pretrained_model:
update_gradient_vars = tf.global_variables()
for var in tf.global_variables():
if not ('Embeddings/' in var.op.name or 'Centralisation/'
in var.op.name or 'centers' in var.op.name
or 'centers_cts' in var.op.name):
restore_vars.append(var)
#else:
#update_gradient_vars.append(var)
else:
restore_vars = tf.global_variables()
update_gradient_vars = tf.global_variables()
########## edit mzh #####################
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
restore_saver = tf.train.Saver(restore_vars) ## mzh
#saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
saver = tf.train.Saver(tf.global_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
#saver.restore(sess, pretrained_model)
restore_saver.restore(
sess,
os.path.join(os.path.expanduser(args.pretrained_model),
ckpt_file))
# Training and validation loop
print('Running training')
epoch = 0
acc = 0
val = 0
far = 0
best_acc = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file, prelogits_center_loss,
cross_entropy_mean, acc, val, far,
centers_cts_batch_reshape, train_acc)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
acc, val, far = evaluate(
sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, embeddings, label_batch,
lfw_paths, actual_issame, args.lfw_batch_size,
args.lfw_nrof_folds, log_dir, step, summary_writer,
args.evaluate_mode)
## saving the best_model
if acc > best_acc:
best_acc = acc
best_model_dir = os.path.join(model_dir, 'best_model')
if not os.path.isdir(
best_model_dir
): # Create the log directory if it doesn't exist
os.makedirs(best_model_dir)
if os.listdir(best_model_dir):
for file in glob.glob(
os.path.join(best_model_dir, '*.*')):
os.remove(file)
for file in glob.glob(os.path.join(model_dir, '*.*')):
shutil.copy(file, best_model_dir)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
image_size = (args.image_size, args.image_size)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
stat_file_name = os.path.join(log_dir, 'stat.h5')
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir, 'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
dataset = facenet.get_dataset(args.data_dir)
if args.filter_filename:
dataset = filter_dataset(dataset, os.path.expanduser(args.filter_filename),
args.filter_percentile, args.filter_min_nrof_images_per_class)
if args.validation_set_split_ratio>0.0:
train_set, val_set = facenet.split_dataset(dataset, args.validation_set_split_ratio, args.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list)>0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size, num_epochs=None,
shuffle=True, seed=None, capacity=32)
index_dequeue_op = index_queue.dequeue_many(args.batch_size*args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,1), name='image_paths')
labels_placeholder = tf.placeholder(tf.int32, shape=(None,1), name='labels')
control_placeholder = tf.placeholder(tf.int32, shape=(None,1), name='control')
nrof_preprocess_threads = 4
input_queue = data_flow_ops.FIFOQueue(capacity=2000000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1,), (1,), (1,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder, control_placeholder], name='enqueue_op')
image_batch, label_batch = facenet.create_input_pipeline(input_queue, image_size, nrof_preprocess_threads, batch_size_placeholder)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Number of examples in training set: %d' % len(image_list))
print('Number of classes in validation set: %d' % len(val_set))
print('Number of examples in validation set: %d' % len(val_image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=slim.initializers.xavier_initializer(),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Norm for the prelogits
eps = 1e-4
prelogits_norm = tf.reduce_mean(tf.norm(tf.abs(prelogits)+eps, ord=args.prelogits_norm_p, axis=1))
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_norm * args.prelogits_norm_loss_factor)
# Add center loss
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch, logits=logits, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
correct_prediction = tf.cast(tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)), tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
nrof_steps = args.max_nrof_epochs*args.epoch_size
nrof_val_samples = int(math.ceil(args.max_nrof_epochs / args.validate_every_n_epochs)) # Validate every validate_every_n_epochs as well as in the last epoch
stat = {
'loss': np.zeros((nrof_steps,), np.float32),
'center_loss': np.zeros((nrof_steps,), np.float32),
'reg_loss': np.zeros((nrof_steps,), np.float32),
'xent_loss': np.zeros((nrof_steps,), np.float32),
'prelogits_norm': np.zeros((nrof_steps,), np.float32),
'accuracy': np.zeros((nrof_steps,), np.float32),
'val_loss': np.zeros((nrof_val_samples,), np.float32),
'val_xent_loss': np.zeros((nrof_val_samples,), np.float32),
'val_accuracy': np.zeros((nrof_val_samples,), np.float32),
'lfw_accuracy': np.zeros((args.max_nrof_epochs,), np.float32),
'lfw_valrate': np.zeros((args.max_nrof_epochs,), np.float32),
'learning_rate': np.zeros((args.max_nrof_epochs,), np.float32),
'time_train': np.zeros((args.max_nrof_epochs,), np.float32),
'time_validate': np.zeros((args.max_nrof_epochs,), np.float32),
'time_evaluate': np.zeros((args.max_nrof_epochs,), np.float32),
'prelogits_hist': np.zeros((args.max_nrof_epochs, 1000), np.float32),
}
for epoch in range(1,args.max_nrof_epochs+1):
step = sess.run(global_step, feed_dict=None)
# Train for one epoch
t = time.time()
cont = train(args, sess, epoch, image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder, phase_train_placeholder, batch_size_placeholder, control_placeholder, global_step,
total_loss, train_op, summary_op, summary_writer, regularization_losses, args.learning_rate_schedule_file,
stat, cross_entropy_mean, accuracy, learning_rate,
prelogits, prelogits_center_loss, args.random_rotate, args.random_crop, args.random_flip, prelogits_norm, args.prelogits_hist_max, args.use_fixed_image_standardization)
stat['time_train'][epoch-1] = time.time() - t
if not cont:
break
t = time.time()
if len(val_image_list)>0 and ((epoch-1) % args.validate_every_n_epochs == args.validate_every_n_epochs-1 or epoch==args.max_nrof_epochs):
validate(args, sess, epoch, val_image_list, val_label_list, enqueue_op, image_paths_placeholder, labels_placeholder, control_placeholder,
phase_train_placeholder, batch_size_placeholder,
stat, total_loss, regularization_losses, cross_entropy_mean, accuracy, args.validate_every_n_epochs, args.use_fixed_image_standardization)
stat['time_validate'][epoch-1] = time.time() - t
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, epoch)
# Evaluate on LFW
t = time.time()
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder, control_placeholder,
embeddings, label_batch, lfw_paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, log_dir, step, summary_writer, stat, epoch,
args.lfw_distance_metric, args.lfw_subtract_mean, args.lfw_use_flipped_images, args.use_fixed_image_standardization)
stat['time_evaluate'][epoch-1] = time.time() - t
print('Saving statistics')
with h5py.File(stat_file_name, 'w') as f:
for key, value in stat.iteritems():
f.create_dataset(key, data=value)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
image_size = (args.image_size, args.image_size)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
try:
os.makedirs(log_dir)
except OSError as exc:
print(exc)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
try:
os.makedirs(model_dir)
except OSError as exc:
print(exc)
stat_file_name = os.path.join(log_dir, 'stat.h5')
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
dataset = facenet.get_dataset(args.data_dir)
if args.filter_filename:
dataset = filter_dataset(dataset,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
if args.validation_set_split_ratio > 0.0:
train_set, val_set = facenet.split_dataset(
dataset, args.validation_set_split_ratio,
args.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(
val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='labels')
control_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='control')
nrof_preprocess_threads = 4
input_queue = data_flow_ops.FIFOQueue(
capacity=2000000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1, ), (1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder, control_placeholder],
name='enqueue_op')
image_batch, label_batch = facenet.create_input_pipeline(
input_queue, image_size, nrof_preprocess_threads,
batch_size_placeholder)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Number of examples in training set: %d' % len(image_list))
print('Number of classes in validation set: %d' % len(val_set))
print('Number of examples in validation set: %d' % len(val_image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=slim.initializers.xavier_initializer(),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Norm for the prelogits
eps = 1e-4
prelogits_norm = tf.reduce_mean(
tf.norm(tf.abs(prelogits) + eps, ord=args.prelogits_norm_p,
axis=1))
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_norm * args.prelogits_norm_loss_factor)
# Add center loss
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch,
args.center_loss_alfa,
nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
correct_prediction = tf.cast(
tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)),
tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
config = tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False)
config.gpu_options.visible_device_list = str(hvd.local_rank())
sess = tf.Session(config=config)
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
# Horovod: broadcast initial variable states from rank 0 to all other processes.
# This is necessary to ensure consistent initialization of all workers when
# training is started with random weights or restored from a checkpoint.
bcast = hvd.broadcast_global_variables(0)
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
nrof_steps = args.max_nrof_epochs * args.epoch_size
nrof_val_samples = int(
math.ceil(args.max_nrof_epochs / args.validate_every_n_epochs)
) # Validate every validate_every_n_epochs as well as in the last epoch
stat = {
'loss':
np.zeros((nrof_steps, ), np.float32),
'center_loss':
np.zeros((nrof_steps, ), np.float32),
'reg_loss':
np.zeros((nrof_steps, ), np.float32),
'xent_loss':
np.zeros((nrof_steps, ), np.float32),
'prelogits_norm':
np.zeros((nrof_steps, ), np.float32),
'accuracy':
np.zeros((nrof_steps, ), np.float32),
'val_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_xent_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_accuracy':
np.zeros((nrof_val_samples, ), np.float32),
'lfw_accuracy':
np.zeros((args.max_nrof_epochs, ), np.float32),
'lfw_valrate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'learning_rate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_train':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_validate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_evaluate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'prelogits_hist':
np.zeros((args.max_nrof_epochs, 1000), np.float32),
}
for epoch in range(1, args.max_nrof_epochs + 1):
step = sess.run(global_step, feed_dict=None)
# Train for one epoch
t = time.time()
cont = train(
args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
control_placeholder, global_step, total_loss, train_op,
summary_op, summary_writer, regularization_losses,
args.learning_rate_schedule_file, stat, cross_entropy_mean,
accuracy, learning_rate, prelogits, prelogits_center_loss,
args.random_rotate, args.random_crop, args.random_flip,
prelogits_norm, args.prelogits_hist_max,
args.use_fixed_image_standardization)
stat['time_train'][epoch - 1] = time.time() - t
if not cont:
break
t = time.time()
if len(val_image_list) > 0 and (
(epoch - 1) % args.validate_every_n_epochs
== args.validate_every_n_epochs - 1
or epoch == args.max_nrof_epochs):
validate(args, sess, epoch, val_image_list, val_label_list,
enqueue_op, image_paths_placeholder,
labels_placeholder, control_placeholder,
phase_train_placeholder, batch_size_placeholder,
stat, total_loss, regularization_losses,
cross_entropy_mean, accuracy,
args.validate_every_n_epochs,
args.use_fixed_image_standardization)
stat['time_validate'][epoch - 1] = time.time() - t
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, epoch)
# Evaluate on LFW
t = time.time()
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, control_placeholder,
embeddings, label_batch, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_nrof_folds, log_dir,
step, summary_writer, stat, epoch,
args.lfw_distance_metric, args.lfw_subtract_mean,
args.lfw_use_flipped_images,
args.use_fixed_image_standardization)
stat['time_evaluate'][epoch - 1] = time.time() - t
print('Saving statistics')
with h5py.File(stat_file_name, 'w') as f:
for key, value in stat.items():
f.create_dataset(key, data=value)
return model_dir
def train():
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, test_set = facenet.split_dataset(dataset, 0.9)
"""Train CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(FLAGS.batch_size, 96, 96, 3), name='Input')
# Build a Graph that computes the logits predictions from the inference model
embeddings = facenet.inference_no_batch_norm_deeper(images_placeholder, tf.constant(True))
#embeddings = facenet.inference(images_placeholder, tf.constant(False))
# Split example embeddings into anchor, positive and negative
#a, p, n = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss_modified(embeddings)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, grads = facenet.train(loss, global_step)
# Create a saver
saver = tf.train.Saver(tf.all_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
check_num = tf.add_check_numerics_ops()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, graph_def=sess.graph_def)
epoch = 1
with sess.as_default():
while epoch<FLAGS.max_nrof_epochs:
batch_number = 0
while batch_number<FLAGS.epoch_size:
print('Loading new data')
image_data, num_per_class = facenet.load_data(train_set)
print('Selecting suitable triplets for training')
start_time = time.time()
emb_list = []
# Run a forward pass for the sampled images
nrof_examples_per_epoch = FLAGS.people_per_batch*FLAGS.images_per_person
nrof_batches_per_epoch = int(np.floor(nrof_examples_per_epoch/FLAGS.batch_size))
#for i in xrange(nrof_batches_per_epoch):
#feed_dict = facenet.get_batch(images_placeholder, image_data, i)
#emb_list += sess.run([embeddings], feed_dict=feed_dict)
#emb_array = np.vstack(emb_list) # Stack the embeddings to a nrof_examples_per_epoch x 128 matrix
## Select triplets based on the embeddings
#apn, nrof_random_negs, nrof_triplets = facenet.select_triplets(emb_array, num_per_class, image_data)
#duration = time.time() - start_time
#print('(nrof_random_negs, nrof_triplets) = (%d, %d): time=%.3f seconds' % (nrof_random_negs, nrof_triplets, duration))
count = 0
# while count<nrof_triplets*3 and batch_number<FLAGS.epoch_size:
while batch_number<FLAGS.epoch_size:
start_time = time.time()
# feed_dict = facenet.get_batch(images_placeholder, apn, batch_number)
feed_dict = facenet.get_batch(images_placeholder, image_data, batch_number)
grad_tensors, grad_vars = zip(*grads)
grads_eval = sess.run(grad_tensors, feed_dict=feed_dict)
for gt, gv in zip(grads_eval, grad_vars):
print('%40s: %6d %6f %6f' % (gv.op.name, np.sum(np.isnan(gt)), np.max(gt), np.min(gt)))
duration = time.time() - start_time
print('Epoch: [%d][%d/%d]\tTime %.3f\ttripErr %2.3f' % (epoch, batch_number, FLAGS.epoch_size, duration, err))
batch_number+=1
count+=FLAGS.batch_size
epoch+=1
# Save the model checkpoint periodically.
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=epoch*FLAGS.epoch_size+batch_number)
def main(args):
TEMPORAL_DIM = int(args.image_w / 5)
SPATIAL_DIM = int(args.image_h / 5)
seed_batch_idx = args.seed_batch_idx
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
action_list, label_list = get_image_paths_and_labels(
dataPath=args.dataPath,
s_ID=None,
is_training=True,
x_type=args.x_type)
assert len(action_list) > 0, 'The dataset should not be empty'
## create input pipeline
queue_input_data = tf.placeholder(
tf.float32, shape=[None, args.image_h, args.image_w, 3])
queue_input_label = tf.placeholder(tf.int64, shape=[None, 1])
queue = tf.FIFOQueue(capacity=20000,
dtypes=[tf.float32, tf.int64],
shapes=[[args.image_h, args.image_w, 3], [1]])
enqueue_op = queue.enqueue_many([queue_input_data, queue_input_label])
dequeue_data, dequeue_lab = queue.dequeue()
dequeue_image = tf.image.per_image_standardization(dequeue_data)
# Create a queue that produces indices into the action_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
#index_dequeue_op = index_queue.dequeue_many(args.batch_size*args.epoch_size, 'index_dequeue')
index_dequeue_op = index_queue.dequeue_many(args.batch_size * 100,
'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
#image_batch, label_batch = tf.train.batch([dequeue_image, dequeue_lab], batch_size=args.batch_size, capacity=100)
image_batch, label_batch = tf.train.batch(
[dequeue_image, dequeue_lab],
batch_size=batch_size_placeholder,
capacity=1000)
label_batch = tf.squeeze(label_batch)
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
tf.summary.image("input_image", image_batch, max_outputs=4)
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(action_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate batch accuracy
correct_prediction = tf.equal(tf.argmax(logits, axis=1), label_batch)
batch_accuracy = tf.reduce_mean(tf.cast(correct_prediction,
tf.float32),
name='batch_accuracy')
tf.summary.scalar('batch_accuracy', batch_accuracy)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one 0batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=10)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(coord=coord, sess=sess)
enqueue_thread = threading.Thread(target=enqueue,
args=[
sess, index_dequeue_op,
enqueue_op, action_list,
label_list, queue_input_data,
queue_input_label, TEMPORAL_DIM,
SPATIAL_DIM, seed_batch_idx
])
enqueue_thread.isDaemon()
enqueue_thread.start()
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(
args,
sess,
epoch, #image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder,
phase_train_placeholder,
batch_size_placeholder,
global_step,
total_loss,
train_op,
summary_op,
summary_writer,
regularization_losses,
args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, embeddings, label_batch,
lfw_paths, actual_issame, args.lfw_batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer)
sess.run(queue.close(cancel_pending_enqueues=True))
coord.request_stop()
coord.join(threads)
sess.close()
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
np.random.seed(seed=args.seed)
random.seed(args.seed)
# load training dataset
train_set = facenet.get_dataset(args.data_dir)
if args.filter_min_nrof_images_per_class > 0:
train_set = clean_dataset(train_set,
args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
# load lfw dataset for validation
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
with tf.device('/cpu:0'):
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
phase_train_placeholder = tf.placeholder(tf.bool,
name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=1000000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder],
name='enqueue_op')
nrof_preprocess_threads = args.nrof_preprocess_threads
images_and_labels = []
for _ in range(nrof_preprocess_threads):
images_and_labels = distorted_inputs(images_and_labels,
input_queue, args)
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=args.batch_size,
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
print(image_batch)
# perfetching queue not working properly yet
batch_queue = slim.prefetch_queue.prefetch_queue(
[image_batch, label_batch], dynamic_pad=True, capacity=4)
image_batch, label_batch = batch_queue.dequeue()
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
#///////////////////////////////////////////////////////////////////////////////////
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay,
reuse=None)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
print(embeddings)
nrof_classes = len(train_set)
weights = tf.get_variable(
'softmax_weights',
shape=(args.embedding_size, nrof_classes),
dtype=tf.float32,
initializer=tf.variance_scaling_initializer(),
regularizer=slim.l2_regularizer(args.weight_decay),
trainable=True)
weights = tf.nn.l2_normalize(weights, 0, name='norm_weights')
if args.keep_probability < 1.0:
scaled_prelogits = slim.dropout(
scaled_prelogits,
args.keep_probability,
is_training=phase_train_placeholder,
scope='Dropout')
logits = facenet.combined_loss(
embeddings,
label_batch,
nrof_classes,
weights,
scale_factor=args.l2_constrained_scale_factor,
m1=args.m1,
m2=args.m2)
# Add norm loss
if args.norm_loss_factor > 0.0:
norm_loss = args.norm_loss_factor * tf.reduce_mean(
tf.pow(
tf.norm(prelogits, axis=1) -
args.l2_constrained_scale_factor, 2))
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, norm_loss)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.trainable_variables(), args.num_gpus,
args.log_histograms)
#///////////////////////////////////////////////////////////////////////////////////
# Create a saver
if args.finetune:
print("finetune model")
all_vars = tf.trainable_variables()
vars_to_restore = [
v for v in all_vars if not v.name.startswith('Logits')
]
else:
vars_to_restore = tf.trainable_variables()
saver = tf.train.Saver(vars_to_restore, max_to_keep=40)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# create corresponding model and log directories
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir),
subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(
args, os.path.join(model_dir, 'arguments.txt'))
#=============================================================================================================
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
allow_soft_placement=True,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
save_graph_def(sess, model_dir)
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, global_step, total_loss,
train_op, summary_op, summary_writer,
regularization_losses, args.learning_rate_schedule_file,
learning_rate, log_dir)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_nrof_folds, log_dir,
step, summary_writer)
return model_dir
def train_facenet(config_):
# display the model parameters
config_.display()
# Get the image_size
image_size = (config_.im_height, config_.im_width)
preprocess(config_)
# Create dirs to save the logs and checkpoints
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(config_.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(config_.models_base_dir),
subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# random seed for reproducability
np.random.seed(seed=config_.seed)
random.seed(config_.seed)
# get the training dataset, the dir should have structure
# person1
# person1_image_1
# person1_image_2
# .
# .
# person 2
# person2_image_1
# .
# .
dataset = facenet.get_dataset(config_.data_dir)
# get train and test dataset
if config_.validation_set_split_ratio > 0.0:
train_set, val_set = facenet.split_dataset(
dataset, config_.validation_set_split_ratio,
config_.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if config_.pretrained_model:
pretrained_model = os.path.expanduser(config_.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
# get the image pairs to test the quality of embedding
if config_.lfw_dir:
print('LFW directory: %s' % config_.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(config_.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(config_.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(config_.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(
val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
config_.batch_size * config_.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='labels')
control_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='control')
nrof_preprocess_threads = 4
input_queue = data_flow_ops.FIFOQueue(
capacity=2000000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1, ), (1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder, control_placeholder],
name='enqueue_op')
image_batch, label_batch = facenet.create_input_pipeline(
input_queue, image_size, nrof_preprocess_threads,
batch_size_placeholder)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Number of examples in training set: %d' % len(image_list))
print('Number of classes in validation set: %d' % len(val_set))
print('Number of examples in validation set: %d' % len(val_image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
config_.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=config_.embedding_size,
weight_decay=config_.weight_decay)
# logits needed for training, we can ignore this in testing phase
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=slim.initializers.xavier_initializer(),
weights_regularizer=slim.l2_regularizer(config_.weight_decay),
scope='Logits',
reuse=False)
# learned embeddings, will be used to assess the similarity between two images
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Norm for the prelogits
eps = 1e-4
prelogits_norm = tf.reduce_mean(
tf.norm(tf.abs(prelogits) + eps,
ord=config_.prelogits_norm_p,
axis=1))
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_norm * config_.prelogits_norm_loss_factor)
# Add center loss
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, config_.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * config_.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
config_.learning_rate_decay_epochs * config_.epoch_size,
config_.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
correct_prediction = tf.cast(
tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)),
tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, config_.optimizer,
learning_rate, config_.moving_average_decay,
tf.global_variables(), config_.log_histograms)
# Create a saver
all_vars = tf.trainable_variables()
var_to_restore = [
v for v in all_vars if not v.name.startswith('Logits')
]
saver = tf.train.Saver(var_to_restore,
max_to_keep=5,
keep_checkpoint_every_n_hours=1.0)
#saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=config_.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
try:
print('Restoring pretrained model: %s' % pretrained_model)
#saver.restore(sess, pretrained_model)
facenet.load_model(pretrained_model)
except:
print(
'Error loading model(maybe checkpoint path is wrong).. training from scratch'
)
# Training and validation loop
print('Running training')
nrof_steps = config_.max_nrof_epochs * config_.epoch_size
nrof_val_samples = int(
math.ceil(config_.max_nrof_epochs /
config_.validate_every_n_epochs)
) # Validate every validate_every_n_epochs as well as in the last epoch
stat = {
'loss':
np.zeros((nrof_steps, ), np.float32),
'center_loss':
np.zeros((nrof_steps, ), np.float32),
'reg_loss':
np.zeros((nrof_steps, ), np.float32),
'xent_loss':
np.zeros((nrof_steps, ), np.float32),
'prelogits_norm':
np.zeros((nrof_steps, ), np.float32),
'accuracy':
np.zeros((nrof_steps, ), np.float32),
'val_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_xent_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_accuracy':
np.zeros((nrof_val_samples, ), np.float32),
'lfw_accuracy':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'lfw_valrate':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'learning_rate':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'time_train':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'time_validate':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'time_evaluate':
np.zeros((config_.max_nrof_epochs, ), np.float32),
'prelogits_hist':
np.zeros((config_.max_nrof_epochs, 1000), np.float32),
}
for epoch in range(1, config_.max_nrof_epochs + 1):
step = sess.run(global_step, feed_dict=None)
# Train for one epoch
t = time.time()
cont = train(
config_, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
control_placeholder, global_step, total_loss, train_op,
summary_op, summary_writer, regularization_losses,
config_.learning_rate_schedule_file, stat,
cross_entropy_mean, accuracy, learning_rate, prelogits,
prelogits_center_loss, config_.random_rotate,
config_.random_crop, config_.random_flip, prelogits_norm,
config_.prelogits_hist_max,
config_.use_fixed_image_standardization)
stat['time_train'][epoch - 1] = time.time() - t
if not cont:
break
t = time.time()
if len(val_image_list) > 0 and (
(epoch - 1) % config_.validate_every_n_epochs
== config_.validate_every_n_epochs - 1
or epoch == config_.max_nrof_epochs):
validate(config_, sess, epoch, val_image_list,
val_label_list, enqueue_op,
image_paths_placeholder, labels_placeholder,
control_placeholder, phase_train_placeholder,
batch_size_placeholder, stat, total_loss,
regularization_losses, cross_entropy_mean,
accuracy, config_.validate_every_n_epochs,
config_.use_fixed_image_standardization)
stat['time_validate'][epoch - 1] = time.time() - t
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, epoch)
# Evaluate on LFW
t = time.time()
if config_.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, control_placeholder,
embeddings, label_batch, lfw_paths, actual_issame,
config_.lfw_batch_size, config_.lfw_nrof_folds,
log_dir, step, summary_writer, stat, epoch,
config_.lfw_distance_metric,
config_.lfw_subtract_mean,
config_.lfw_use_flipped_images,
config_.use_fixed_image_standardization)
stat['time_evaluate'][epoch - 1] = time.time() - t
print('Saving statistics')
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir, 'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
# 预训练模型
if args.pretrained_model:
print('Pre-trained model: %s' % os.path.expanduser(args.pretrained_model))
# # lfw 数据集的位置
# if args.lfw_dir:
# print('LFW directory: %s' % args.lfw_dir)
# # Read the file containing the pairs used for testing
# pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# # Get the paths for the corresponding images
# lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False, name='global_step')
# Placeholders for the learning rate, batch_size, phase_train, image_path, labels
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None, 3), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None, 3), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3,), (3,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder])
# 读取数据的线程数，将下面改为 multi_gpu 运行的版本
nrof_preprocess_threads = 4
images_and_labels_all = []
for _ in range(nrof_preprocess_threads):
for gpu in range(args.num_gpus):
images_and_labels = []
# 每次都从 queue 中将数据 dequeue 出来
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
images_and_labels_all.append(images_and_labels)
# 将数据整理，并适用于下面的多 gpu 运行
image_batch_split = []
label_batch_split = []
# label_extend = []
for i in range(args.num_gpus):
image_batch, labels_batch = tf.train.batch_join(
images_and_labels_all[i], batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
labels_batch = tf.identity(labels_batch, 'label_batch')
image_batch_split.append(image_batch)
label_batch_split.append(labels_batch)
# label_extend.extend(labels_batch)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# print('Using optimizer: {}'.format(args.optimizer))
# if args.optimizer == 'ADAGRAD':
# opt = tf.train.AdagradOptimizer(learning_rate)
# elif args.optimizer == 'MOM':
# opt = tf.train.MomentumOptimizer(learning_rate, rho=0.9, epsilon=1e-6)
# elif args.optimizer=='ADAM':
# opt = tf.train.AdamOptimizer(learning_rate, beta1=0.9, beta2=0.999, epsilon=0.1)
# elif args.optimizer=='RMSPROP':
# opt = tf.train.RMSPropOptimizer(learning_rate, decay=0.9, momentum=0.9, epsilon=1.0)
# elif args.optimizer=='MOM':
# opt = tf.train.MomentumOptimizer(learning_rate, 0.9, use_nesterov=True)
# else:
# raise Exception('Not supported optimizer: {}'.format(args.optimizer))
# 在这部分进行 multi_gpu
print('Building training graph....')
tower_losses = []
tower_triplet = []
tower_reg= []
# embeddings_extend = []
embeddings_split = []
for i in range(args.num_gpus):
with tf.device("/gpu:" + str(i)):
with tf.name_scope("tower_" + str(i)) as scope:
with tf.variable_scope(tf.get_variable_scope()) as var_scope:
# Build the inference graph
with tf.variable_scope(name_or_scope='', reuse=tf.AUTO_REUSE):
# # Dequeues one batch for one tower
# image_batch_de, label_batch_de = batch_queue.dequeue()
prelogits, _ = network.inference(image_batch_split[i], args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# embeddings_extend.extend(embeddings)
embeddings_split.append(embeddings)
# Split embeddings into anchor, postive and negative and calculate triplet loss
anchor, positive, negative = tf.unstack(tf.reshape(embeddings, [-1,3,args.embedding_size]), 3, 1)
triplet_loss_split = facenet.triplet_loss(anchor, positive, negative, args.alpha)
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
tower_triplet.append(triplet_loss_split)
loss = triplet_loss_split + tf.add_n(regularization_losses)
tower_losses.append(loss)
tower_reg.append(regularization_losses)
# 同名变量可以重用
tf.get_variable_scope().reuse_variables()
total_loss = tf.reduce_mean(tower_losses)
total_reg = tf.reduce_mean(tower_reg)
losses = {}
losses['total_loss'] = total_loss
losses['total_reg'] = total_reg
# # 计算 embeddings 的均值
# tmp_embeddings = None
# for j in range(arg.num_gpus):
# if j > 0:
# tmp_embeddings += embeddings_split[j]
# else:
# tmp_embeddings = embeddings_split[j]
# embeddings = tmp_embeddings / args.num_gpus
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables())
# grads = opt.compute_gradients(total_loss,tf.trainable_variables(),colocate_gradients_with_ops=True)
# apply_gradient_op = opt.apply_gradients(grads,global_step=global_step)
# update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
# with tf.control_dependencies(update_ops):
# train_op = tf.group(apply_gradient_op)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer(), feed_dict={phase_train_placeholder:True})
sess.run(tf.local_variables_initializer(), feed_dict={phase_train_placeholder:True})
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
saver.restore(sess, os.path.expanduser(args.pretrained_model))
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train multi gpus for one epoch
epoch_start_time = time.time()
# for i in range(args.num_gpus):
train_multi_gpu(args, sess, train_set, epoch, image_paths_placeholder, labels_placeholder,
label_batch_split[0], label_batch_split[1], batch_size_placeholder, learning_rate_placeholder,
phase_train_placeholder, enqueue_op, input_queue, global_step, embeddings_split[0], embeddings_split[1], losses['total_loss'], losses['total_reg'], train_op,
summary_op, summary_writer, args.learning_rate_schedule_file, args.embedding_size)
print('The %dth epoch running time is %.3f seconds!!!' %(epoch, time.time()-epoch_start_time))
# # Train for one epoch
# train(args, sess, epoch,
# learning_rate_placeholder, phase_train_placeholder, global_step,
# losses, train_op, summary_op, summary_writer, args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
return model_dir
def main(argv=None): # pylint: disable=unused-argument
if FLAGS.model_name:
subdir = FLAGS.model_name
preload_model = True
else:
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
preload_model = False
log_dir = os.path.join(os.path.expanduser(FLAGS.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.mkdir(log_dir)
model_dir = os.path.join(os.path.expanduser(FLAGS.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.mkdir(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
store_training_info(src_path, log_dir, ' '.join(argv))
np.random.seed(seed=FLAGS.seed)
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, validation_set = facenet.split_dataset(dataset,
FLAGS.train_set_fraction,
FLAGS.split_mode)
print('Model directory: %s' % model_dir)
with tf.Graph().as_default():
tf.set_random_seed(FLAGS.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32,
shape=(FLAGS.batch_size,
FLAGS.image_size,
FLAGS.image_size, 3),
name='input')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Build the inference graph
embeddings = facenet.inference_nn4_max_pool_96(
images_placeholder,
FLAGS.pool_type,
FLAGS.use_lrn,
FLAGS.keep_probability,
phase_train=phase_train_placeholder)
# Split example embeddings into anchor, positive and negative
anchor, positive, negative = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(anchor, positive, negative, FLAGS.alpha)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, _ = facenet.train(loss, global_step, FLAGS.optimizer,
FLAGS.learning_rate,
FLAGS.moving_average_decay)
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=0)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(
log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
with sess.as_default():
if preload_model:
ckpt = tf.train.get_checkpoint_state(model_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
else:
raise ValueError('Checkpoint not found')
# Training and validation loop
for epoch in range(FLAGS.max_nrof_epochs):
# Train for one epoch
step = train(sess, train_set, epoch, images_placeholder,
phase_train_placeholder, global_step, embeddings,
loss, train_op, summary_op, summary_writer)
# Test on validation set
validate(sess, validation_set, epoch, images_placeholder,
phase_train_placeholder, global_step, embeddings,
loss, 'validation', summary_writer)
# Test on training set
validate(sess, train_set, epoch, images_placeholder,
phase_train_placeholder, global_step, embeddings,
loss, 'training', summary_writer)
if (epoch % FLAGS.checkpoint_period
== 0) or (epoch == FLAGS.max_nrof_epochs - 1):
# Save the model checkpoint
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
# Save the model if it hasn't been saved before
graphdef_dir = os.path.join(model_dir, 'graphdef')
graphdef_filename = 'graph_def.pb'
if (not os.path.exists(
os.path.join(graphdef_dir, graphdef_filename))):
print('Saving graph definition')
tf.train.write_graph(sess.graph_def, graphdef_dir,
graphdef_filename, False)
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir):
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir):
os.makedirs(model_dir)
np.random.seed(seed=args.seed)
random.seed(args.seed)
nrof_classes = 8631
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
# images_lfw = np.load(args.lfw_dir+'.npy')
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
tfrecords_list = glob.glob(args.tfrecord_dir)
filename_queue = tf.train.string_input_producer(tfrecords_list,
shuffle=True)
reader = tf.TFRecordReader()
key, records = reader.read(filename_queue)
features = tf.parse_single_example(records,
features={
'label':
tf.FixedLenFeature([],
tf.int64),
"image_raw":
tf.FixedLenFeature([],
tf.string)
})
label = tf.cast(features['label'], tf.int32)
image = tf.cast(
tf.image.decode_jpeg(features["image_raw"], channels=3),
tf.float32)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
image.set_shape((112, 96, 3))
image = tf.subtract(image, 127.5) * 0.0078125
images_and_labels = [image, label]
image_batch, label_batch = tf.train.shuffle_batch(
images_and_labels,
batch_size=args.batch_size,
shapes=[(112, 96, 3), ()],
capacity=40000,
min_after_dequeue=10000,
allow_smaller_final_batch=True)
# image_batch.set_shape([args.batch_size, 112, 112, 3])
# label_batch.set_shape([args.batch_size])
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label')
image_batch.set_shape([None, 112, 96, 3])
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: 3000000')
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
prelogits = tf.identity(prelogits, 'prelogits')
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
cross_entropy_mean, logits = cosSoftmax_loss(embeddings,
label_batch,
args.batch_size,
nrof_classes,
m=0.35,
s=30,
name='softmax')
# cross_entropy_mean, accuracy = adaptive_loss(embeddings, label_batch, args.batch_size, nrof_classes, m=0.35, s=30, name='softmax')
# AM_logits, logits = AM_logits_compute(embeddings, label_batch, args, nrof_classes)
#AM_logits = Arc_logits(embeddings, label_batch, args, nrof_classes)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
# labels=label_batch, logits=AM_logits, name='cross_entropy_per_example')
# cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
correct_prediction = tf.cast(
tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)),
tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
for weights in slim.get_variables_by_name('kernel'):
kernel_regularization = tf.contrib.layers.l2_regularizer(
args.weight_decay)(weights)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
kernel_regularization)
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
if args.weight_decay == 0:
total_loss = tf.add_n([cross_entropy_mean], name='total_loss')
else:
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
tf.add_to_collection('losses', total_loss)
#define two saver in case under 'finetuning on different dataset' situation
saver_save = tf.train.Saver(tf.trainable_variables(), max_to_keep=1)
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.trainable_variables(), args.log_histograms)
# train_op = tf.train.AdamOptimizer(learning_rate).minimize(total_loss,global_step = global_step,var_list=tf.trainable_variables())
# train_op = tf.train.MomentumOptimizer(learning_rate,momentum=0.9).minimize(total_loss,global_step=global_step,var_list=tf.trainable_variables())
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
# gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
gpu_options = tf.GPUOptions(allow_growth=True)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
exclusions = []
except_exclusions = slim.get_variables_to_restore(
exclude=exclusions)
restore_variables = [
v for v in tf.trainable_variables()
if v in except_exclusions
]
saver_load = tf.train.Saver(restore_variables)
print('Restoring pretrained model: %s' % pretrained_model)
saver_load.restore(sess, pretrained_model)
best_accuracy = evaluate_double(sess, phase_train_placeholder,
image_batch, embeddings, lfw_paths,
actual_issame, log_dir, 0,
summary_writer, 0.0, saver_save,
model_dir, subdir, args)
print('Running training')
epoch = 0
best_accuracy = 0.0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
train(args, sess, epoch, learning_rate_placeholder,
phase_train_placeholder, global_step, total_loss,
train_op, summary_op, summary_writer,
regularization_losses, accuracy, learning_rate)
print('validation running...')
if args.lfw_dir:
best_accuracy = evaluate_double(
sess, phase_train_placeholder, image_batch, embeddings,
lfw_paths, actual_issame, log_dir,
step + args.epoch_size, summary_writer, best_accuracy,
saver_save, model_dir, subdir, args)
return model_dir
def main(argv=None): # pylint: disable=unused-argument
if FLAGS.model_name:
subdir = FLAGS.model_name
preload_model = True
else:
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
preload_model = False
log_dir = os.path.join(os.path.expanduser(FLAGS.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.mkdir(log_dir)
model_dir = os.path.join(os.path.expanduser(FLAGS.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.mkdir(model_dir)
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(argv))
np.random.seed(seed=FLAGS.seed)
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, validation_set = facenet.split_dataset(dataset, FLAGS.train_set_fraction, FLAGS.split_mode)
print('Model directory: %s' % model_dir)
with tf.Graph().as_default():
tf.set_random_seed(FLAGS.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(None, FLAGS.image_size, FLAGS.image_size, 3), name='input')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Build the inference graph
embeddings = network.inference(images_placeholder, FLAGS.pool_type, FLAGS.use_lrn,
FLAGS.keep_probability, phase_train=phase_train_placeholder)
# Split example embeddings into anchor, positive and negative
anchor, positive, negative = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(anchor, positive, negative, FLAGS.alpha)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, _ = facenet.train(loss, global_step, FLAGS.optimizer, FLAGS.learning_rate, FLAGS.moving_average_decay)
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=0)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
with sess.as_default():
if preload_model:
ckpt = tf.train.get_checkpoint_state(model_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
else:
raise ValueError('Checkpoint not found')
# Training and validation loop
for epoch in range(FLAGS.max_nrof_epochs):
# Train for one epoch
step = train(sess, train_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, train_op, summary_op, summary_writer)
# Store the state of the random number generator
rng_state = np.random.get_state()
# Test on validation set
np.random.seed(seed=FLAGS.seed)
validate(sess, validation_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, 'validation', summary_writer)
# Test on training set
np.random.seed(seed=FLAGS.seed)
validate(sess, train_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, 'training', summary_writer)
# Restore state of the random number generator
np.random.set_state(rng_state)
if (epoch % FLAGS.checkpoint_period == 0) or (epoch==FLAGS.max_nrof_epochs-1):
# Save the model checkpoint
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
# Read data and apply label preserving distortions
image_batch, label_batch = facenet.read_and_augument_data(
image_list, label_list, args.image_size, args.batch_size,
args.max_nrof_epochs, args.random_crop, args.random_flip,
args.nrof_preprocess_threads)
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
# Build the inference graph
prelogits, _ = network.inference(image_batch,
args.keep_probability,
phase_train=True,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
# Add DeCov regularization loss
if args.decov_loss_factor > 0.0:
logits_decov_loss = facenet.decov_loss(
logits) * args.decov_loss_factor
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
logits_decov_loss)
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.scalar_summary('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
logits, label_batch, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.all_variables(), args.log_histograms)
# Evaluation
print('Building evaluation graph')
lfw_label_list = range(0, len(lfw_paths))
assert (
len(lfw_paths) % args.lfw_batch_size == 0
), "The number of images in the LFW test set need to be divisible by the lfw_batch_size"
eval_image_batch, eval_label_batch = facenet.read_and_augument_data(
lfw_paths,
lfw_label_list,
args.image_size,
args.lfw_batch_size,
None,
False,
False,
args.nrof_preprocess_threads,
shuffle=False)
# Node for input images
eval_image_batch.set_shape((None, args.image_size, args.image_size, 3))
eval_image_batch = tf.identity(eval_image_batch, name='input')
eval_prelogits, _ = network.inference(eval_image_batch,
1.0,
phase_train=False,
weight_decay=0.0,
reuse=True)
eval_embeddings = tf.nn.l2_normalize(eval_prelogits,
1,
1e-10,
name='embeddings')
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.initialize_all_variables())
sess.run(tf.initialize_local_variables())
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, learning_rate_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, eval_embeddings, eval_label_batch,
actual_issame, args.lfw_batch_size, args.seed,
args.lfw_nrof_folds, log_dir, step,
summary_writer)
return model_dir
def main(args):
image_size = (args.image_size, args.image_size)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
stat_file_name = os.path.join(log_dir, 'stat.h5')
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
dataset = facenet.get_dataset(args.data_dir)
nrof_classes = len(dataset)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 3),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 3),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=200000,
dtypes=[tf.string, tf.int64],
shapes=[(3, ), (3, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
# image_batch, label_batch = facenet.create_input_pipeline(input_queue, image_size, nrof_preprocess_threads, batch_size_placeholder)
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(labels_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Building training graph')
with tf.contrib.slim.arg_scope(
mobilenet_v2.training_scope(
is_training=True,
dropout_keep_prob=args.keep_probability,
weight_decay=args.weight_decay)):
logits, end_points = mobilenet_v2.mobilenet(
image_batch, num_classes=nrof_classes)
prelogits = tf.squeeze(end_points['global_pool'], [1, 2])
print('After mobilenet ')
logits = slim.fully_connected(
prelogits,
args.embedding_size,
activation_fn=None,
weights_initializer=slim.initializers.xavier_initializer(),
reuse=False)
embeddings = tf.identity(logits, 'embeddings')
"""
Tensor("Squeeze:0", shape=(?, 1280), dtype=float32)
Tensor("fully_connected/BiasAdd:0", shape=(?, 512), dtype=float32)
Tensor("label_batch:0", shape=(?,), dtype=int64)
"""
print('logits node')
print(prelogits)
print(logits)
print(label_batch)
# g = tf.get_default_graph()
# tf.contrib.quantize.create_training_graph(input_graph=g)
# Norm for the prelogits
eps = 1e-4
# Split embeddings into anchor, positive and negative and calculate triplet loss
anchor, positive, negative = tf.unstack(
tf.reshape(embeddings, [-1, 3, args.embedding_size]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative,
args.alpha)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables())
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# saver_new = tf.train.Saver(max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
# gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, dataset, epoch, image_paths_placeholder,
labels_placeholder, label_batch, batch_size_placeholder,
learning_rate_placeholder, enqueue_op, input_queue,
global_step, embeddings, total_loss, train_op,
summary_op, summary_writer,
args.learning_rate_schedule_file, args.embedding_size,
anchor, positive, negative, triplet_loss)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, lfw_paths, embeddings, label_batch,
image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder,
enqueue_op, actual_issame, args.batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer, args.embedding_size)
return model_dir
def main(args):
project_dir = os.path.dirname(os.getcwd())
network = importlib.import_module(args.model_def)
with open(join(project_dir, 'config.yaml'), 'r') as f:
cfg = yaml.load(f)
if cfg['specs']['set_gpu']:
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = str(cfg['base_conf']['gpu_num'])
subdir = '%s_center_loss_factor_%1.2f' % (args.data_dir,
args.center_loss_factor)
# test = os.path.expanduser(args.logs_base_dir)
log_dir = os.path.join(project_dir, 'fine_tuning_process', 'logs', subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(project_dir, 'fine_tuning_process', 'models',
subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
data_dir = os.path.join(project_dir, 'fine_tuning_process', 'data',
args.data_dir, 'train')
train_set = facenet.get_dataset(data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = os.path.join(project_dir, 'fine_tuning_process',
'models',
cfg['model_map'][args.embedding_size])
print('Pre-trained model: %s' % pretrained_model)
lfw_dir = os.path.join(project_dir, 'fine_tuning_process', 'data',
args.data_dir, 'test')
print('LFW directory: %s' % lfw_dir)
# Read the file containing the pairs used for testing
lfw_pairs = os.path.join(project_dir, 'fine_tuning_process', 'data',
args.data_dir, 'pairs.txt')
pairs = lfw.read_pairs(lfw_pairs)
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths_personal(lfw_dir, pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# get soft labels
with open(join(data_dir, 'soft_label.pk'), 'rb') as f:
confidence_score = pickle.load(f)
image_list, soft_labels_list = facenet.get_image_paths_and_soft_labels(
train_set, confidence_score)
soft_labels_array = np.array(soft_labels_list)
soft_labels = ops.convert_to_tensor(soft_labels_array,
dtype=tf.float32)
assert len(image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
range_size = array_ops.shape(soft_labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
hard_labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='hard_labels')
soft_labels_placeholder = tf.placeholder(tf.float32,
shape=(None, nrof_classes),
name='soft_labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.float32],
shapes=[(1, ), (nrof_classes, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, soft_labels_placeholder],
name='enqueue_op')
nrof_preprocess_threads = 4
images_and_softlabels = []
for _ in range(nrof_preprocess_threads):
filenames, soft_labels = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_softlabels.append([images, soft_labels])
image_batch, soft_label_batch = tf.train.batch_join(
images_and_softlabels, batch_size=batch_size_placeholder)
image_batch = tf.squeeze(image_batch, 1)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
soft_label_batch = tf.identity(soft_label_batch, 'soft_label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
# fine_tuning = slim.fully_connected(prelogits, args.embedding_size, activation_fn=None,
# scope='FineTuning', reuse=False, trainable=True)
logits = slim.fully_connected(
prelogits,
nrof_classes,
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.fuzzy_center_loss(
prelogits, soft_label_batch, args.center_loss_alfa,
args.fuzzier, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
tf.summary.scalar('prelogits_center_loss', prelogits_center_loss)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.softmax_cross_entropy_with_logits_v2(
labels=soft_label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
all_vars = tf.trainable_variables()
var_to_restore = [
v for v in all_vars if not v.name.startswith('Logits')
]
saver = tf.train.Saver(var_to_restore, max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
result = sess.graph.get_tensor_by_name(
"InceptionResnetV1/Bottleneck/weights:0")
pre = sess.graph.get_tensor_by_name(
"InceptionResnetV1/Block8/Branch_1/Conv2d_0c_3x1/weights:0"
)
# tf.stop_gradient(persisted_result)
# print(result.eval())
# print("======")
# print(pre.eval())
# Training and validation loop
print('Running training')
epoch = 0
pre_acc = -1
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, soft_labels_array,
index_dequeue_op, enqueue_op, image_paths_placeholder,
soft_labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file, logits)
# print(result.eval())
# print("======")
# print(pre.eval())
# Save variables and the metagraph if it doesn't exist already
# Evaluate on LFW
if args.lfw_dir:
acc = evaluate(sess, enqueue_op, image_paths_placeholder,
soft_labels_placeholder,
phase_train_placeholder,
batch_size_placeholder, embeddings,
soft_label_batch, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_nrof_folds,
log_dir, step, summary_writer, nrof_classes,
prelogits_center_loss)
if acc > pre_acc:
save_variables_and_metagraph(sess, saver,
summary_writer, model_dir,
subdir, step)
pre_acc = acc
return model_dir
def main(args):
# 主函数入口
# 从model模块中选择要使用的训练结构
network = importlib.import_module(args.model_def)
# 日志路径和模型保存路径的子目录命名结构
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
# 当前脚本的的绝对路径，存储相关信息到log文件中
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
# 用参数初始化numpy随机器
np.random.seed(seed=args.seed)
random.seed(args.seed)
# 获取训练数据
train_set = facenet.get_dataset(args.data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set, os.path.expanduser(args.filter_filename),
args.filter_percentile, args.filter_min_nrof_images_per_class)
# 训练数据子类别数目
nrof_classes = len(train_set)
# 打印模型和log保存路径，如果有预训练模型需要加载即加载
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
# 如果指定了lfw验证数据路径即打印，并且加载lfw数据配对文件及相关信息
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
# 新建一个图，将该图作为默认图，进入上下文管理器
with tf.Graph().as_default():
# 设置随机种子和全局
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
# 得到训练数据的列表和相对应的标签列表
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list)>0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
# 将标签列表转化为tensor形式的列表，并获取tensor的范围
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
#
index_queue = tf.train.range_input_producer(range_size, num_epochs=None,
shuffle=True, seed=None, capacity=32)
# 构建训练标签队列，并设置训练中的参数
index_dequeue_op = index_queue.dequeue_many(args.batch_size*args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,1), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None,1), name='labels')
# 构建训练数据的队列，训练数据，训练标签配对
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1,), (1,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder], name='enqueue_op')
# 数据预处理线程数
nrof_preprocess_threads = 4
images_and_labels = []
# 使用线程去并行处理数据
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
# 读取队列中的配对数据，读取训练图像和训练标签，并对数据进行预处理
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_png(file_contents)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
# 如果没开启随机截取的开关，即选择图像正中的位置进行训练
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
# 再一次确认图像分辨率
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
# 配置训练batch的参数
image_batch, label_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
# 打印最终处理完的数据信息
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
# 构建TensorFlow数据流图
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
# 连接全连接层
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
# 对特征维度（128维）做一个L2正则化处理
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
# 添加center loss
if args.center_loss_factor>0.0:
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
#
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
# 计算交叉熵
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch, logits=logits, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
# 计算总的loss
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
# 构建总的训练方式
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables(), args.log_histograms)
# Create a saver
# 构建一个保存对象
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
# 构建一个总的训练摘要对象
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
# 开始训练，配置GPU使用参数
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
# 开始训练循环
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder, phase_train_placeholder, batch_size_placeholder, global_step,
total_loss, train_op, summary_op, summary_writer, regularization_losses, args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder,
embeddings, label_batch, lfw_paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, log_dir, step, summary_writer)
sess.close()
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S') + args.experiment_name
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir, 'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
class_name = ['smile','oval_face','5_ocloc_shadow','bald','archied_eyebrows','Big_lips', 'Big_Nose']
class_num = len(class_name)
class_index = [31,25,0,4,1,6,7]
all_image_list = []
all_label_list = []
for i in range(class_num):
image_list = []
label_list = []
train_set = facenet.get_sub_category_dataset(args.data_dir, class_index[i])
image_list_p, label_list_p = facenet.get_image_paths_and_labels_triplet(train_set[0], args)
image_list_n, label_list_n = facenet.get_image_paths_and_labels_triplet(train_set[1], args)
image_list.append(image_list_p)
image_list.append(image_list_n)
label_list.append(label_list_p)
label_list.append(label_list_n)
all_image_list.append(image_list)
all_label_list.append(label_list)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model:
print('Pre-trained model: %s' % os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
image_size = args.image_size
batch_size = args.batch_size
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# image_paths_placeholder = tf.placeholder(tf.string, shape=(None,3), name='image_paths')
image_placeholder = tf.placeholder(tf.float32, shape=(batch_size, image_size, image_size,3), name='images')
labels_placeholder = tf.placeholder(tf.int64, shape=(batch_size,3), name='labels')
code_placeholder = tf.placeholder(tf.float32, shape=(batch_size,class_num,1,1), name='code')
image_batch = normalized_image(image_placeholder)
code_batch = code_placeholder
control_code = tf.tile(code_placeholder,[1,1,args.embedding_size,1])
mask_array = np.ones((1 ,class_num,args.embedding_size,1),np.float32)
# for i in range(class_num):
# mask_array[:,i,(args.embedding_size/class_num)*i:(args.embedding_size/class_num)*(i+1)] = 1
mask_tensor = tf.get_variable('mask', dtype=tf.float32, trainable=args.learned_mask, initializer=tf.constant(mask_array))
mask_tensor = tf.tile(mask_tensor,[batch_size,1,1,1])
control_code = tf.tile(code_placeholder,[1,1,args.embedding_size,1])
mask_out = tf.multiply(mask_tensor, control_code)
mask_out = tf.reduce_sum(mask_out,axis=1)
mask_out = tf.squeeze(mask_out)
mask_out = tf.nn.relu(mask_out)
mask0_array = np.ones((1, class_num, 128, 1), np.float32)
mask0_tensor = tf.get_variable('mask0', dtype=tf.float32, trainable=args.learned_mask,
initializer=tf.constant(mask0_array))
mask0_tensor = tf.tile(mask0_tensor, [batch_size, 1, 1, 1])
control0_code = tf.tile(code_placeholder,[1,1,128,1])
mask0_out = tf.multiply(mask0_tensor, control0_code)
mask0_out = tf.reduce_sum(mask0_out, axis=1)
mask0_out = tf.squeeze(mask0_out)
mask0_out = tf.nn.relu(mask0_out)
mask0_out = tf.expand_dims(mask0_out,1)
mask0_out = tf.expand_dims(mask0_out,1)
mask1_array = np.ones((1, class_num, 128, 1), np.float32)
mask1_tensor = tf.get_variable('mask1', dtype=tf.float32, trainable=args.learned_mask,
initializer=tf.constant(mask1_array))
mask1_tensor = tf.tile(mask1_tensor, [batch_size, 1, 1, 1])
control1_code = tf.tile(code_placeholder,[1,1,128,1])
mask1_out = tf.multiply(mask1_tensor, control1_code)
mask1_out = tf.reduce_sum(mask1_out, axis=1)
mask1_out = tf.squeeze(mask1_out)
mask1_out = tf.nn.relu(mask1_out)
mask1_out = tf.expand_dims(mask1_out,1)
mask1_out = tf.expand_dims(mask1_out,1)
# Build the inference graph
prelogits, _ = network.inference(image_batch, mask0_out, mask1_out, args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
embeddings_pre = tf.multiply(mask_out, prelogits)
embeddings = tf.nn.l2_normalize(embeddings_pre, 1, 1e-10, name='embeddings')
anchor_index = list(range(0,batch_size,3))
positive_index = list(range(1,batch_size,3))
negative_index = list(range(2,batch_size,3))
a_indice = tf.constant(np.array(anchor_index))
p_indice = tf.constant(np.array(positive_index))
n_indice = tf.constant(np.array(negative_index))
anchor = tf.gather(embeddings,a_indice)
positive = tf.gather(embeddings,p_indice)
negative = tf.gather(embeddings,n_indice)
triplet_loss = facenet.triplet_loss(anchor, positive, negative, args.alpha)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables())
# Create a saver
trainable_variables = tf.global_variables()
saver = tf.train.Saver(trainable_variables, max_to_keep=35)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer(), feed_dict={phase_train_placeholder:True})
sess.run(tf.local_variables_initializer(), feed_dict={phase_train_placeholder:True})
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
saver.restore(sess, os.path.expanduser(args.pretrained_model))
# Training and validation loop
epoch = 0
Accuracy = [0]
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
# epoch = step // args.epoch_size
# Train for one epoch
code_list = []
triplets_list = []
max_num = 32768
if (epoch+1)%args.lr_epoch == 0:
args.learning_rate = 0.1*args.learning_rate
if args.random_trip:
for i in range(class_num):
code = np.zeros((batch_size, class_num, 1, 1), np.float32)
_class = i
code[:, _class, :, :] = 1
Triplets = triplet_random(args, sess, all_image_list[i], all_image_list, epoch, image_placeholder,
batch_size_placeholder, learning_rate_placeholder,
phase_train_placeholder, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer,
args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss, max_num)
triplets_list.append(Triplets)
code_list.append(code)
else:
for i in range(class_num):
code = np.zeros((batch_size, 1, 1, 1), np.float32)
_class = i
if _class > 3:
_class = 3
code[:, :, :, :] = _class
print(class_name[i])
Triplets = triplet(args, sess, all_image_list[i], epoch, image_placeholder, code_placeholder,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer, args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss,code)
triplets_num = len(Triplets)
triplets_list.append(Triplets)
code_list.append(code)
train(args, sess, image_list, epoch, image_placeholder, code_placeholder,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer, args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss, triplets_list, code_list, model_dir, Accuracy)
if (epoch+1)%2 == 0:
Accuracy = test(args, sess, image_list, epoch, image_placeholder,code_placeholder,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer, args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss, triplets_list, Accuracy)
# Save variables and the metagraph if it doesn't exist already
model_name = 'epoch' + str(epoch+1)
print(model_dir)
if (epoch+1) > 0 :
if (epoch +1)%2 == 0:
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, model_name, step)
print('models are saved in ', os.path.join(model_dir, model_name))
epoch = epoch + 1
sess.close()
return model_dir
def main(args):
# 导入网络架构模型
network = importlib.import_module(args['model_def'])
# 用当前日期来命名模型
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
# 日志保存目录
log_dir = os.path.join(os.path.expanduser(args['logs_base_dir']), subdir)
# 没有日志文件就创建一个
if not os.path.isdir(log_dir):
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args['models_base_dir']),
subdir)
# 没有模型保存目录就创建一个
if not os.path.isdir(model_dir):
os.makedirs(model_dir)
# 保存参数日志
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# 设置随机数种子
np.random.seed(seed=args['seed'])
# 获取数据集，train_set是包含文件路径与标签的集合
# 包含图片地址的（image_paths）以及对应的人名(name)
train_set = facenet.get_dataset(args['data_dir'])
print('模型目录: %s' % model_dir)
print('log目录: %s' % log_dir)
# 判断是否有预训练模型
if args['pretrained_model']:
print('Pre-trained model: %s' %
os.path.expanduser(args['pretrained_model']))
if args['lfw_dir']:
print('LFW目录: %s' % args['lfw_dir'])
# 读取用于测试的pairs文件
pairs = lfw.read_pairs(os.path.expanduser(args['lfw_pairs']))
shuffle(pairs)
# 获取对应的路径
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args['lfw_dir']), pairs, args['lfw_file_ext'])
# 建立图
with tf.Graph().as_default():
tf.set_random_seed(args['seed'])
global_step = tf.Variable(0, trainable=False)
# 学习率
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
# 批大小
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
# 用于判断是训练还是测试
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# 图像路径
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 3),
name='image_paths')
# 图像标签
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 3),
name='labels')
# 新建一个队列，数据流操作，先入先出
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3, ), (3, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder])
preprocess_threads = 4
images_and_labels = []
for _ in range(preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
# 随机水平反转
if args['random_flip']:
image = tf.image.random_flip_left_right(image)
image.set_shape((args['image_size'], args['image_size'], 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args['image_size'], args['image_size'], 3), ()],
enqueue_many=True,
capacity=4 * preprocess_threads * args['batch_size'],
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
labels_batch = tf.identity(labels_batch, 'label_batch')
# 构造计算图
# 其中prelogits是最后一层的输出
prelogits, _ = network.inference(
image_batch,
args['keep_probability'],
phase_train=phase_train_placeholder,
bottleneck_layer_size=args['embedding_size'],
weight_decay=args['weight_decay'])
# L2正则化
# embeddings = tf.nn.l2_normalize
# 输入向量, L2范化的维数（取0（列L2范化）或1（行L2范化））
# 泛化的最小值边界, name='embeddings')
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# 计算 triplet_loss
anchor, positive, negative = tf.unstack(
tf.reshape(embeddings, [-1, 3, args['embedding_size']]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative,
args['alpha'])
# 将指数衰减应用在学习率上
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args['learning_rate_decay_epochs']\
* args['epoch_size'],
args['learning_rate_decay_factor'],
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# 计算损失
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# 构建L2正则化
total_loss = tf.add_n([triplet_loss] + regularization_losses,
name='total_loss')
# 确定优化方法并根据损失函求梯度，每更新一次参数，global_step 会加 1
train_op = facenet.train(total_loss, global_step, args['optimizer'],
learning_rate, args['moving_average_decay'],
tf.global_variables())
# 创建一个saver用来保存或者从内存中读取一个模型参数
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
summary_op = tf.summary.merge_all()
# 设置显存比例
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args['gpu_memory_fraction'])
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# 初始化变量
sess.run(tf.global_variables_initializer(),
feed_dict={phase_train_placeholder: True})
sess.run(tf.local_variables_initializer(),
feed_dict={phase_train_placeholder: True})
# 写log文件
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
# 获取线程
coord = tf.train.Coordinator()
# 将队列中的多用sunner开始执行
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
# 读入预训练模型（如果有）
if args['pretrained_model']:
print('载入预训练模型: %s' % args['pretrained_model'])
# saver.restore(sess, os.path.expanduser(args['pretrained_model']))
facenet.load_model(args['pretrained_model'])
epoch = 0
# 将所有数据过一遍的次数
while epoch < args['max_nrof_epochs']:
step = sess.run(global_step, feed_dict=None)
# epoch_size是一个epoch中批的个数
# epoch是全局的批处理个数以一个epoch中。。。这个epoch将用于求学习率
epoch = step // args['epoch_size']
# 训练一个epoch
train(args, sess, train_set, epoch, image_paths_placeholder,
labels_placeholder, labels_batch, batch_size_placeholder,
learning_rate_placeholder, phase_train_placeholder,
enqueue_op, input_queue, global_step, embeddings,
total_loss, train_op, summary_op, summary_writer,
args['learning_rate_schedule_file'],
args['embedding_size'], anchor, positive, negative,
triplet_loss)
# 保存变量和metagraph（如果不存在）
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# 使用lfw评价当前模型
if args['lfw_dir']:
evaluate(sess, lfw_paths, embeddings, labels_batch,
image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder,
phase_train_placeholder, enqueue_op,
actual_issame, args['batch_size'],
args['lfw_nrof_folds'], log_dir, step,
summary_writer, args['embedding_size'])
return model_dir
def train():
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, test_set = facenet.split_dataset(dataset, 0.9)
fileName = "/home/david/debug4.h5"
f = h5py.File(fileName, "r")
for item in f.values():
print(item)
w1 = f['1w'][:]
b1 = f['1b'][:]
f.close()
print(w1.shape)
print(b1.shape)
"""Train CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32,
shape=(FLAGS.batch_size, 96, 96,
3),
name='Input')
# Build a Graph that computes the logits predictions from the inference model
#embeddings = facenet.inference_nn4_max_pool_96(images_placeholder, phase_train=True)
conv1 = _conv(images_placeholder,
3,
64,
7,
7,
2,
2,
'SAME',
'conv1_7x7',
phase_train=False,
use_batch_norm=False,
init_weight=w1,
init_bias=b1)
resh1 = tf.reshape(conv1, [-1, 294912])
embeddings = _affine(resh1, 294912, 128)
# Split example embeddings into anchor, positive and negative
a, p, n = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(a, p, n)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, grads = facenet.train(loss, global_step)
# Create a saver
saver = tf.train.Saver(tf.all_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(
log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
graph_def=sess.graph_def)
epoch = 0
with sess.as_default():
while epoch < FLAGS.max_nrof_epochs:
batch_number = 0
while batch_number < FLAGS.epoch_size:
print('Loading new data')
image_data, num_per_class, image_paths = facenet.load_data(
train_set)
print('Selecting suitable triplets for training')
start_time = time.time()
emb_list = []
# Run a forward pass for the sampled images
nrof_examples_per_epoch = FLAGS.people_per_batch * FLAGS.images_per_person
nrof_batches_per_epoch = int(
np.floor(nrof_examples_per_epoch / FLAGS.batch_size))
if True:
for i in xrange(nrof_batches_per_epoch):
feed_dict, _ = facenet.get_batch(
images_placeholder, image_data, i)
emb_list += sess.run([embeddings],
feed_dict=feed_dict)
emb_array = np.vstack(
emb_list
) # Stack the embeddings to a nrof_examples_per_epoch x 128 matrix
# Select triplets based on the embeddings
apn, nrof_random_negs, nrof_triplets = facenet.select_triplets(
emb_array, num_per_class, image_data)
duration = time.time() - start_time
print(
'(nrof_random_negs, nrof_triplets) = (%d, %d): time=%.3f seconds'
% (nrof_random_negs, nrof_triplets, duration))
count = 0
while count < nrof_triplets * 3 and batch_number < FLAGS.epoch_size:
start_time = time.time()
feed_dict, batch = facenet.get_batch(
images_placeholder, apn, batch_number)
if (batch_number % 20 == 0):
err, summary_str, _ = sess.run(
[loss, summary_op, train_op],
feed_dict=feed_dict)
summary_writer.add_summary(
summary_str,
FLAGS.epoch_size * epoch + batch_number)
else:
err, _ = sess.run([loss, train_op],
feed_dict=feed_dict)
duration = time.time() - start_time
print(
'Epoch: [%d][%d/%d]\tTime %.3f\ttripErr %2.3f'
% (epoch, batch_number, FLAGS.epoch_size,
duration, err))
batch_number += 1
count += FLAGS.batch_size
else:
while batch_number < FLAGS.epoch_size:
start_time = time.time()
feed_dict, _ = facenet.get_batch(
images_placeholder, image_data, batch_number)
grad_tensors, grad_vars = zip(*grads)
eval_list = (train_op, loss) + grad_tensors
result = sess.run(eval_list, feed_dict=feed_dict)
grads_eval = result[2:]
nrof_parameters = 0
for gt, gv in zip(grads_eval, grad_vars):
print('%40s: %6d' % (gv.op.name, np.size(gt)))
nrof_parameters += np.size(gt)
print('Total number of parameters: %d' %
nrof_parameters)
err = result[1]
batch_number += 1
epoch += 1
# Save the model checkpoint periodically.
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess,
checkpoint_path,
global_step=epoch * FLAGS.epoch_size + batch_number)
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model:
print('Pre-trained model: %s' % os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,3), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None,3), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3,), (3,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder])
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unpack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_png(file_contents)
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, weight_decay=args.weight_decay)
pre_embeddings = slim.fully_connected(prelogits, args.embedding_size, activation_fn=None, scope='Embeddings', reuse=False)
#embedding_size = 1792
embeddings = tf.nn.l2_normalize(pre_embeddings, 1, 1e-10, name='embeddings')
# Split embeddings into anchor, positive and negative and calculate triplet loss
anchor, positive, negative = tf.unpack(tf.reshape(embeddings, [-1,3,args.embedding_size]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative, args.alpha)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses, name='total_loss')
# Create list with variables to restore
restore_vars = []
update_gradient_vars = []
if args.pretrained_model:
update_gradient_vars = tf.global_variables()
for var in tf.global_variables():
if not 'Embeddings/' in var.op.name:
restore_vars.append(var)
#else:
#update_gradient_vars.append(var)
else:
restore_vars = tf.global_variables()
update_gradient_vars = tf.global_variables()
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, update_gradient_vars)
# Create a saver
restore_saver = tf.train.Saver(restore_vars)
saver = tf.train.Saver(tf.global_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
restore_saver.restore(sess, os.path.expanduser(args.pretrained_model))
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, train_set, epoch, image_paths_placeholder, labels_placeholder, labels_batch,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, enqueue_op, input_queue, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer, args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, lfw_paths, embeddings, labels_batch, image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, enqueue_op, actual_issame, args.batch_size,
args.seed, args.lfw_nrof_folds, log_dir, step, summary_writer, args.embedding_size)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
#-------------------------------------------------------------------------------------------------------------------
test_set = facenet.get_dataset(args.lfw_dir)
if args.filter_filename:
test_set = filter_dataset(test_set,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
nrof_classes = len(test_set)
#----------------------------------------------------------------------------------------------------------------
image_test_list, label_test_list = facenet.get_image_paths_and_labels(
test_set)
assert len(image_test_list) > 0, 'The dataset should not be empty'
#-----------------------------------------------------------------------------------------------------------------
#-------------------------------------------------------------------------------------------------------------------
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
#------------------------------------------------------------------------------------------------------------------------
test_labels = ops.convert_to_tensor(label_test_list, dtype=tf.int32)
test_range_size = array_ops.shape(test_labels)[0]
test_index_queue = tf.train.range_input_producer(test_range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
test_index_dequeue_op = test_index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
test_image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='test_image_paths')
test_labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='test_labels')
input_test_queue = data_flow_ops.FIFOQueue(
capacity=8000000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
vali_enqueue_op = input_test_queue.enqueue_many(
[test_image_paths_placeholder, test_labels_placeholder],
name='vali_enqueue_op')
#-----------------------------------------------------------------------------------------------------------------------
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
#--------------------------------------------------------------------------------------------------------------------------------
test_images_and_labels = []
test_filenames, test_label = input_test_queue.dequeue()
test_images = []
for test_filename in tf.unstack(test_filenames):
test_file_contents = tf.read_file(test_filename)
test_image = tf.image.decode_image(test_file_contents, channels=3)
test_image.set_shape((args.image_size, args.image_size, 3))
test_images.append(tf.image.per_image_standardization(test_image))
test_images_and_labels.append([test_images, test_label])
test_image_batch, test_label_batch = tf.train.batch_join(
test_images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * args.batch_size,
allow_smaller_final_batch=True)
test_image_batch = tf.identity(test_image_batch, 'test_image_batch')
test_image_batch = tf.identity(test_image_batch, 'test_input')
test_label_batch = tf.identity(test_label_batch, 'test_label_batch')
#---------------------------------------------------------------------------------------------------------------------------------
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
print(image_batch)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
logits = tf.identity(logits, 'final_Logits')
#trainable=False,
#----------------------------------------------------------------------------------------------------------------------------------
test_prelogits, _ = network.inference(
test_image_batch,
1,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay,
reuse=True)
test_logits = slim.fully_connected(
test_prelogits,
len(test_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=True)
with tf.name_scope('accuracy'):
onehot_labels = tf.one_hot(test_label_batch, 7)
correct_prediction = tf.equal(tf.argmax(test_logits, 1),
tf.argmax(onehot_labels, 1))
correct_prediction = tf.cast(correct_prediction, tf.float32)
accuracy_ = tf.cast(tf.reduce_mean(correct_prediction), tf.float32)
#print(accuracy_)
with tf.name_scope('confusion'):
confusion = tf.contrib.metrics.confusion_matrix(test_label_batch,
tf.argmax(
test_logits,
1),
num_classes=7)
#print(confusion)
#-----------------------------------------------------------------------------------------------------------------------------------
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
# cross_entropy_mean = focal_loss(onehot_labels=label_batch, cls_preds=logits,
# alpha=0.25, gamma=2, name=None, scope=None)
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver1 = tf.train.Saver(tf.trainable_variables()[0:-1],
max_to_keep=200)
saver = tf.train.Saver(max_to_keep=200)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
# init_0 = tf.global_variables_initializer()
# sess.run(init_0)
# var_names = []
# name2var = dict(zip(map(lambda x:x.name.split(':')[0], tf.global_variables()),tf.global_variables()))
# with tf.variable_scope('',reuse=True):
# for var_name, saved_var_name in var_names[:-1]:
# curr_var = name2var[saved_var_name]
# var_shape = curr_var.get_shape().as_list()
# if var_shape == saved_shapes[saved_var_name]:
# restore_vars.append(curr_var)
# saver = tf.train.Saver(restore_vars,max_to_keep=3)
saver1.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# print(step)
# Train for one epoch
train(args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file)
sum_ = 0
confusion_m = tf.zeros((7, 7))
for i in range(7):
acc, m = test(
confusion, accuracy_, args, sess, epoch,
image_test_list, label_test_list, index_dequeue_op,
vali_enqueue_op, test_image_paths_placeholder,
test_labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file)
sum_ = sum_ + acc
confusion_m = confusion_m + m
avg = sum_ / 7
confusion_m = confusion_m / tf.reduce_sum(confusion_m, 1)
print(avg)
np.set_printoptions(precision=4)
print(confusion_m.eval())
summary = tf.Summary()
#pylint: disable=maybe-no-member
summary.value.add(tag='lfw/accuracy', simple_value=avg)
#summary.value.add(tag='lfw/c_matrix', simple_value=m)
summary_writer.add_summary(summary, step)
with open(os.path.join(log_dir, 'lfw_result.txt'), 'at') as f:
f.write('\n%d\t%.5f\t\n' % (step, avg))
f.write(np.array_str(confusion_m.eval()))
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# # Evaluate on LFW
# if args.lfw_dir:
# evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder,
# embeddings, label_batch, lfw_paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, log_dir, step, summary_writer)
return model_dir
def main(args):
logging.info('###### all args #####: %s' % args)
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
if not args.no_store_revision_info:
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set, args.filter_filename,
args.filter_percentile,
args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
logging.info('Model directory: %s' % model_dir)
logging.info('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
logging.info('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
logging.info('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
logging.info('image_list size %d, label_list size %d' %
(len(image_list), len(label_list)))
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
logging.info('labels shape %s, range size: %s' %
(labels.shape, str(range_size)))
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
logging.info('batch size:%d, epoch size:%d' %
(args.batch_size, args.epoch_size))
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
logging.info('# filenames len:%s' % filenames.shape)
images = []
for filename in tf.unstack(filenames):
# logging.info('#file:%s' % filename)
file_contents = tf.read_file(filename)
image = tf.image.decode_jpeg(file_contents)
# logging.info('#image shape:%s' % image.shape)
# if args.random_rotate:
# image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
# # if args.random_crop:
# # image = tf.random_crop(image, [args.image_size, args.image_size, 3])
# else:
# image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
# if args.random_flip:
# image = tf.image.random_flip_left_right(image)
image = porcessor.preprocess_image(image,
args.image_size,
args.image_size,
is_training=True)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
logging.info('Total number of classes: %d' % nrof_classes)
logging.info('Total number of examples: %d' % len(image_list))
logging.info('Building training graph')
batch_norm_params = {
# Decay for the moving averages
'decay': 0.995,
# epsilon to prevent 0s in variance
'epsilon': 0.001,
# force in-place updates of mean and variance estimates
'updates_collections': None,
# Moving averages ends up in the trainable variables collection
'variables_collections': [tf.GraphKeys.TRAINABLE_VARIABLES],
# Only update statistics during training mode
'is_training': phase_train_placeholder
}
# Build the inference graph
# with tf.device('/GPU:0'):
prelogits, _ = network.inference(image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
weight_decay=args.weight_decay)
bottleneck = slim.fully_connected(
prelogits,
args.embedding_size,
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
normalizer_fn=slim.batch_norm,
normalizer_params=batch_norm_params,
scope='Bottleneck',
reuse=False)
logits = slim.fully_connected(
bottleneck,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(bottleneck,
1,
1e-10,
name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=10)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default(), tf.device('/gpu:1'):
if pretrained_model:
logging.info('Restoring pretrained model: %s' %
pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
logging.info('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, embeddings, label_batch,
lfw_paths, actual_issame, args.lfw_batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer)
sess.close()
return model_dir
def main(args):
print('args used to train the model:')
print(args)
# =========================================================================== #
# 数据预处理,路径问题等
# =========================================================================== #
network = importlib.import_module(args.model_def)
image_size = (args.image_size, args.image_size)
subdir = datetime.strftime(datetime.now(),
'%Y%m%d-%H%M%S') # '20180614-100359'
subdir += args.dir_postfix
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir),
subdir) # 本地保存log的路径
log_dir += args.dir_postfix
val_log_dir = log_dir # 本地保存validation result的路径
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir),
subdir) # 本地保存model的路径,最后一层目录用上面生成的日期命名
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
stat_file_name = os.path.join(log_dir, 'stat.h5')
# Write arguments to a text file, 把训练时用的所有参数保存到txt中
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
# dataset中是多个ImageClass对象,每个对象中有类名(name)和该类所有图片的路径(image_paths)
dataset = facenet.get_dataset(args.data_dir)
# filter_name default='', 这个训练的时候没用到, 先不管了
if args.filter_filename:
dataset = filter_dataset(dataset,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
# validation_set_split_ratio default=0.0, 如果需要验证操作的话设置这个
# 划分验证集有两种方式:按类划分和按图片划分(默认)
if args.validation_set_split_ratio > 0.0:
train_set, val_set = facenet.split_dataset(
dataset, args.validation_set_split_ratio,
args.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
# pairs[0] = ['Abel_Pacheco', '1', '4']
# args.lfw_pairs: data/pairs.txt, 这个文件中每一行或有三个字段或有四个字段
# 如果有三个字段表示从一个人的所有图片中选择两个(正样本, actual_issame=True)
# 如果有四个字段表示从两个人的所有图片中各选一个(负样本, actual_issame=False)
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
# lfw_paths里的每个值都是一个元组(path0, path1), actual_issame里对应的是这个元组是正样本还是副样本
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
# =========================================================================== #
# 下面开始处理TensorFlow相关的东西
# =========================================================================== #
with tf.Graph().as_default():
# =========================================================================== #
# Input pipeline
# =========================================================================== #
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
# 这两个list是训练集中的所有数据
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(
val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
# num_epochs=None时,循环生成0到range_size-1的数
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
# 这个op是用来生成索引值的,每次从队列里取出一个epoch中所有图片数量的下标
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='labels')
control_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='control')
nrof_preprocess_threads = 4
input_queue = data_flow_ops.FIFOQueue(
capacity=2000000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1, ), (1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder, control_placeholder],
name='enqueue_op')
# create input pipeline, 将上面image path中的文件取出来
image_batch, label_batch = facenet.create_input_pipeline(
input_queue, image_size, nrof_preprocess_threads,
batch_size_placeholder)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Number of examples in training set: %d' % len(image_list))
print('Number of classes in validation set: %d' % len(val_set))
print('Number of examples in validation set: %d' % len(val_image_list))
print('Building training graph')
### Build the inference graph
# prelogits是128维的向量
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
# TODO: Inject some code here for fine tuning
# logits的shape: [None, len(train_set)]
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=slim.initializers.xavier_initializer(),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
# 这步操作之后在测试时就可以拿到这个值了
logits = tf.identity(logits, name='logits')
# axis=1, 是把每行进行normalize, 每行是一个样本, 128维(或512维)的向量
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Norm for the prelogits
eps = 1e-4
prelogits_norm = tf.reduce_mean(
tf.norm(tf.abs(prelogits) + eps, ord=args.prelogits_norm_p,
axis=1))
# 把prelogits进行norm操作得到的是一个数,这个数乘个weight作为loss的一部分,--prelogits_norm_loss_factor 5e-4
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_norm * args.prelogits_norm_loss_factor)
# Add center loss, 这个center loss作为正则化loss的一部分, 出自一篇人脸识别的论文, 先不管
# 这个loss默认权重是0, 文档中给出的训练脚本并没有用这个loss
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch,
args.center_loss_alfa,
nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
correct_prediction = tf.cast(
tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)),
tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# 这个total_loss是ce和regular loss的和,应该是已经把所有loss的值都加起来了
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=20)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
### Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
# =========================================================================== #
# Start training
# =========================================================================== #
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
# saver.restore(sess, pretrained_model)
if args.checkpoint_exclude_scopes:
exclusions = [args.checkpoint_exclude_scopes.split(',')]
else:
exclusions = []
# load from pre-trained model
tf.contrib.framework.assign_from_checkpoint_fn(
pretrained_model, [
var for var in tf.trainable_variables()
if all(not var.op.name.startswith(exclusion)
for exclusion in exclusions)
],
ignore_missing_vars=args.ignore_missing_vars)(sess)
# Training and validation loop
print('Running training')
# max_nrof_epochs: number of epochs to run, default=500
# epoch_size: number of batches per epoch,default=1000
nrof_steps = args.max_nrof_epochs * args.epoch_size # total batch number
# 在验证集上验证的次数
nrof_val_samples = int(
math.ceil(args.max_nrof_epochs / args.validate_every_n_epochs)
) # Validate every validate_every_n_epochs as well as in the last epoch
stat = {
# 每一个batch记录一次
'loss':
np.zeros((nrof_steps, ), np.float32),
'center_loss':
np.zeros((nrof_steps, ), np.float32),
'reg_loss':
np.zeros((nrof_steps, ), np.float32),
'xent_loss':
np.zeros((nrof_steps, ), np.float32),
'prelogits_norm':
np.zeros((nrof_steps, ), np.float32),
'accuracy':
np.zeros((nrof_steps, ), np.float32),
# 每验证一次记录一次
'val_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_xent_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_accuracy':
np.zeros((nrof_val_samples, ), np.float32),
# 每个epoch记录一次
'lfw_accuracy':
np.zeros((args.max_nrof_epochs, ), np.float32),
'lfw_valrate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'learning_rate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_train':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_validate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_evaluate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'prelogits_hist':
np.zeros((args.max_nrof_epochs, 1000), np.float32),
}
for epoch in range(1, args.max_nrof_epochs + 1):
step = sess.run(global_step, feed_dict=None) # step = 0
### Train for one epoch
t = time.time()
# =========================================================================== #
# 一次训练的过程,学习率是在这里设置的,如果参数中的学习率小于0,就用调度文件中的学习率
# =========================================================================== #
cont = train(
args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
control_placeholder, global_step, total_loss, train_op,
summary_op, summary_writer, regularization_losses,
args.learning_rate_schedule_file, stat, cross_entropy_mean,
accuracy, learning_rate, prelogits, prelogits_center_loss,
args.random_rotate, args.random_crop, args.random_flip,
prelogits_norm, args.prelogits_hist_max,
args.use_fixed_image_standardization)
stat['time_train'][epoch - 1] = time.time() - t
if not cont:
# 每次结束一轮训练,train函数返回True
break
### validate on val_set
t = time.time()
if len(val_image_list) > 0 and \
((
epoch - 1) % args.validate_every_n_epochs == args.validate_every_n_epochs - 1 or epoch == args.max_nrof_epochs):
# =========================================================================== #
# 如果验证集不为空 并且 到了设置的每n个epoch验证一次或者到了最后一个epoch, 则进行一次验证
# =========================================================================== #
validate(args, sess, epoch, val_image_list, val_label_list,
enqueue_op, image_paths_placeholder,
labels_placeholder, control_placeholder,
phase_train_placeholder, batch_size_placeholder,
stat, total_loss, regularization_losses,
cross_entropy_mean, accuracy,
args.validate_every_n_epochs,
args.use_fixed_image_standardization, val_log_dir)
stat['time_validate'][epoch - 1] = time.time() - t
# Save variables and the metagraph if it doesn't exist
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, epoch)
### Evaluate on LFW
t = time.time()
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, control_placeholder,
args.lfw_nrof_folds, log_dir, step,
summary_writer, stat, epoch, embeddings,
label_batch, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_distance_metric,
args.lfw_subtract_mean,
args.lfw_use_flipped_images,
args.use_fixed_image_standardization)
stat['time_evaluate'][epoch - 1] = time.time() - t
print('Saving statistics')
with h5py.File(stat_file_name, 'w') as f:
for key, value in stat.items():
f.create_dataset(key, data=value)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')#以时间方式命名一个子路径
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)#最终的日志的地址
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)#如果路径不存在就建立
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)#模型的路径
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir, 'arguments.txt'))#将argument的具体信息记录在log目录下的argument.txt文件中
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)#使用seed可以保证随机数部分每次出现的结果都一样
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set, os.path.expanduser(args.filter_filename),
args.filter_percentile, args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)#nrof_classes为train_set的类别数
print('Model directory: %s' % model_dir)#打印出来，model和log的地址
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)#如果有提前训练的model就在载入（？）这里没有写载入的代码，但是大概应该是需要载入
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)#打印LFW数据集的地址
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)#获取到lfw数据集的地址，issame是什么还是不知道
with tf.Graph().as_default():#开始tensorflow的会话
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)#图片和标签的list
assert len(image_list)>0, 'The dataset should not be empty'
# 创建一个队列实现将label和picture输入网络，因为数据太大需要使用到队列
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]#label的个数
index_queue = tf.train.range_input_producer(range_size, num_epochs=None,
shuffle=True, seed=None, capacity=32)#相当于一个一个往外出
index_dequeue_op = index_queue.dequeue_many(args.batch_size*args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,1), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None,1), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1,), (1,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder], name='enqueue_op')
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, label_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)#通过googlenet网络中之后输出的结果，详细见models.inception_v1的文件，结果就是构建了网络
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)#在全连接之后初始化参数
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')#进行正则化
# Add center loss
if args.center_loss_factor>0.0:
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_center_loss * args.center_loss_factor)#将引入center loss factor 之后的loss引入到纳入到graph中的正则化loss类别中
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)#设定学习率的初始值和变化方式
tf.summary.scalar('learning_rate', learning_rate)#可视化中离散的学习率的图表
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch, logits=logits, name='cross_entropy_per_example')#每个训练结果的交叉熵损失
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')#得到所有的交叉熵损失的平均值
tf.add_to_collection('losses', cross_entropy_mean)#将平均交叉熵的值纳入到损失类别中
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')#最后的loss由center_loss和cross_entropy两部分组成
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables(), args.log_histograms)#训练的设定
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)#保存图表到log中
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()#tensorboard中显示所有信息
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))#设置每个gpu分配多少的内存用于该process
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())#初始化所有的variable
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)#将graph保存到log的路径下面
coord = tf.train.Coordinator()#创立一个线程管理器
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:#训练的轮次
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder, phase_train_placeholder, batch_size_placeholder, global_step,
total_loss, train_op, summary_op, summary_writer, regularization_losses, args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:#在lfw数据集中进行验证
evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder,
embeddings, label_batch, lfw_paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, log_dir, step, summary_writer)
return model_dir
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
if args.minutiae_pairs_dir is not None:
minutiae_pairs = args.minutiae_pairs_dir.split(':')
evaluate_paths = []
evaluate_paths.append(glob.glob(minutiae_pairs[0]+'*.jpeg'))
evaluate_paths[0].sort()
evaluate_paths.append(glob.glob(minutiae_pairs[1]+'*.jpeg'))
evaluate_paths[1].sort()
with open(os.path.join(model_dir, 'args.txt'), 'w') as f:
for arg in vars(args):
f.write(arg + ' ' + str(getattr(args, arg)) + '\n')
# Store some git revision info in a text file in the log directory
if not args.no_store_revision_info:
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model:
print('Pre-trained model: %s' % os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,3), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None,3), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3,), (3,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder])
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
#image = tf.image.decode_png(file_contents)
image = tf.image.decode_jpeg(file_contents, channels=3)
#image = facenet.crop(image, None,args.image_size+30)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
image = tf.cast(image,tf.float32)
#image = tf.image.per_image_standardization(image)
distorted_image = tf.image.random_brightness(image, max_delta=32)
image = tf.image.random_contrast(distorted_image, lower=0.5, upper=1.5)
#images.append(tf.image.per_image_standardization(image))
images.append(image)
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
batch_norm_params = {
# Decay for the moving averages
'decay': 0.995,
# epsilon to prevent 0s in variance
'epsilon': 0.001,
# force in-place updates of mean and variance estimates
'updates_collections': None,
# Moving averages ends up in the trainable variables collection
'variables_collections': [ tf.GraphKeys.TRAINABLE_VARIABLES ],
# Only update statistics during training mode
'is_training': phase_train_placeholder
}
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, weight_decay=args.weight_decay)
#pre_embeddings = slim.fully_connected(prelogits, args.embedding_size, activation_fn=None,
# weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
# weights_regularizer=slim.l2_regularizer(args.weight_decay),
# normalizer_fn=slim.batch_norm,
# normalizer_params=batch_norm_params,
# scope='Bottleneck', reuse=False)
pre_embeddings = _fully_connected(prelogits, args.embedding_size, name='Bottleneck')
embeddings = tf.nn.l2_normalize(pre_embeddings, 1, 1e-10, name='embeddings')
# Split embeddings into anchor, positive and negative and calculate triplet loss
anchor, positive, negative = tf.unstack(tf.reshape(embeddings, [-1,3,args.embedding_size]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative, args.alpha)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables())
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer(), feed_dict={phase_train_placeholder:True})
sess.run(tf.local_variables_initializer(), feed_dict={phase_train_placeholder:True})
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
saver.restore(sess, os.path.expanduser(args.pretrained_model))
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, train_set, epoch, image_paths_placeholder, labels_placeholder, labels_batch,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, enqueue_op, input_queue, global_step,
embeddings, total_loss, train_op, summary_op, summary_writer, args.learning_rate_schedule_file,
args.embedding_size, anchor, positive, negative, triplet_loss)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
if evaluate_paths:
evaluate_NISTSD27(sess, evaluate_paths, embeddings, labels_batch, image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, enqueue_op,
args.batch_size, log_dir, step, summary_writer, args.embedding_size)
# Evaluate on LFW
#if args.lfw_dir:
# evaluate(sess, lfw_paths, embeddings, labels_batch, image_paths_placeholder, labels_placeholder,
# batch_size_placeholder, learning_rate_placeholder, phase_train_placeholder, enqueue_op, actual_issame, args.batch_size,
# args.lfw_nrof_folds, log_dir, step, summary_writer, args.embedding_size)
sess.close()
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
image_size = (args.image_size, args.image_size)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir),
subdir) #日志保存地址
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
# os.path.realpath(__file__)代表返回当前模块真实路径
# os.path.split(os.path.realpath(__file__))代表返回路径的目录和文件名(元组形式)
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
# dataset是列表，列表元素为每一个类的ImageClass对象，定义见facenet.py文件
dataset = facenet.get_dataset(args.data_dir)
if args.filter_filename:
dataset = filter_dataset(dataset,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
# 划分训练集和测试集，train_set和val_set的形式和dataset一样
if args.validation_set_split_ratio > 0.0:
train_set, val_set = facenet.split_dataset(
dataset, args.validation_set_split_ratio,
args.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set) #类目数(即有多少人)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
# pairs也是列表，子元素也是列表，每个子列表包含pairs.txt的每一行
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
#lfw_paths为两两比较的列表,列表的元素为有2个元素的元祖,actual_issame为列表,表示每个元素是否为同一个人
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
# global_step对应的是全局批的个数，根据这个参数可以更新学习率
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(
val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
# range_size的大小和样本个数一样
range_size = array_ops.shape(labels)[0]
#QueueRunner：保存的是队列中的入列操作，保存在一个list当中，其中每个enqueue运行在一个线程当中
#range_input_producer：返回的是一个队列，队列中有打乱的整数，范围是从0到range size
#range_size大小和总的样本个数一样
#并将一个QueueRunner添加到当前图的QUEUE_RUNNER集合中
index_queue = tf.train.range_input_producer(
range_size, num_epochs=None, shuffle=True, seed=None,
capacity=32) #capacity代表队列容量
# 返回的是一个出列操作，每次出列一个epoch需要用到的样本个数
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='labels')
control_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='control')
# 上面的队列是一个样本索引的出队队列，只用来出列
# 用来每次出列一个epoch中用到的样本
# 这里是第二个队列，这个队列用来入列，每个元素的大小为shape=[(1,),(1,),(1,)]
nrof_preprocess_threads = 4
# 这里的shape代表每一个元素的维度
input_queue = data_flow_ops.FIFOQueue(
capacity=600000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1, ), (1, ), (1, )],
shared_name=None,
name=None)
# 这时一个入列的操作，这个操作将在session run的时候用到
# 每次入列的是image_paths_placeholder, labels_placeholder对
# 注意，这里只有2个队列，一个用来出列打乱的元素序号
# 一个根据对应的需要读取指定的文件
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder, control_placeholder],
name='enqueue_op')
image_batch, label_batch = facenet.create_input_pipeline(
input_queue, image_size, nrof_preprocess_threads,
batch_size_placeholder)
# image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# arcface_logits = arcface_logits_compute(embeddings, label_batch, args, nrof_classes)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
with tf.variable_scope('Logits'):
cosface_logits = AM_logits_compute(embeddings, label_batch, args,
nrof_classes)
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=cosface_logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
correct_prediction = tf.cast(
tf.equal(tf.argmax(cosface_logits, 1),
tf.cast(label_batch, tf.int64)), tf.float32)
accuracy = tf.reduce_mean(correct_prediction)
# for weights in slim.get_variables_by_name('embedding_weights'):
# weight_regularization = tf.contrib.layers.l2_regularizer(args.weight_decay)(weights)
# tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, weight_regularization)
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
if args.weight_decay == 0:
total_loss = tf.add_n([cross_entropy_mean], name='total_loss')
else:
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
#define two saver in case under 'finetuning on different dataset' situation
saver_save = tf.train.Saver(tf.trainable_variables(), max_to_keep=1)
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
facenet.load_model(pretrained_model)
print('Running cosface training')
best_accuracy = 0.0
for epoch in range(1, args.max_nrof_epochs + 1):
step = sess.run(global_step, feed_dict=None)
cont = train(
args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
global_step, total_loss, accuracy, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file, args.random_rotate,
args.random_crop, args.random_flip,
args.use_fixed_image_standardization, control_placeholder)
if not cont:
break
print('validation running...')
if args.lfw_dir:
best_accuracy = evaluate(
sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, control_placeholder,
embeddings, label_batch, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_nrof_folds, log_dir,
step, summary_writer, best_accuracy, saver_save,
model_dir, subdir, args.lfw_subtract_mean,
args.lfw_use_flipped_images,
args.use_fixed_image_standardization,
args.lfw_distance_metric)
return model_dir
def main(args):
# 动态import python模块, 这里指的是外部参数指定的网络结构
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
# os.path.expanduser跨平台支持替换路径中的user路径~
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir, 'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
# args.data_dir can contain more datasets, separated by comma
# TODO: no logic for name conflict?
data_dirs = args.data_dir.split(",")
train_set = []
for data_dir in data_dirs:
if len(data_dir) > 0:
train_set.extend(facenet.get_dataset(data_dir))
if args.filter_filename:
train_set = filter_dataset(train_set, os.path.expanduser(args.filter_filename),
args.filter_percentile, args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
continue_ckpt_dir = None
if args.continue_ckpt_dir:
continue_ckpt_dir = os.path.expanduser(args.continue_ckpt_dir)
print('Continue training from the checkpoint: %s' % continue_ckpt_dir)
snapshot_at_step = None
if args.snapshot_at_step:
snapshot_at_step = int(args.snapshot_at_step)
print('Will take a snapshot checkpoint at step', snapshot_at_step)
nrof_preprocess_threads = 4
if args.nrof_preprocess_threads:
nrof_preprocess_threads = int(args.nrof_preprocess_threads)
print('Number of preprocess threads', nrof_preprocess_threads)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
# https://www.tensorflow.org/api_guides/python/threading_and_queues
# This function converts Python objects of various types to Tensor objects.
# It accepts Tensor objects, numpy arrays, Python lists, and Python scalars.
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
# This operation returns a 1-D integer tensor representing the shape of input.
range_size = array_ops.shape(labels)[0]
# Produces the integers from 0 to limit-1 in a queue.
index_queue = tf.train.range_input_producer(range_size, num_epochs=None,
shuffle=True, seed=None, capacity=32)
index_dequeue_op = index_queue.dequeue_many(args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None, 1), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None, 1), name='labels')
# Creates a queue that dequeues elements in a first-in first-out order.
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1,), (1,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder], name='enqueue_op')
# 读取图片文件, 将图片转换成tensor并且做ensembling处理, 结果存入images_and_labels数组
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
# Unpacks the given dimension of a rank-R tensor into rank-(R-1) tensors.
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
# Detects whether an image is a GIF, JPEG, or PNG, and performs the appropriate operation
# to convert the input bytes string into a Tensor of type uint8.
# Note: decode_gif returns a 4-D array [num_frames, height, width, 3],
# as opposed to decode_jpeg and decode_png, which return 3-D arrays [height, width, num_channels].
image = tf.image.decode_image(file_contents, channels=3)
# 对训练图片做ensembling
# https://www.tensorflow.org/api_docs/python/tf/image
# https://www.tensorflow.org/api_docs/python/tf/contrib/image
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
if args.random_crop:
# 训练数据的图片(182)比参数传进来的大小(160)略大, 不做缩放而是直接随机切成160的
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
if args.random_brightness:
image = tf.image.random_brightness(image, max_delta=0.2)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
# Runs a list of tensors to fill a queue to create batches of examples.
image_batch, label_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
# https://stackoverflow.com/questions/34877523/in-tensorflow-what-is-tf-identity-used-for
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
# Normalizes along dimension dim using an L2 norm.
# For a 1-D tensor with dim = 0, computes output = x / sqrt(max(sum(x**2), epsilon))
# For x with more dimensions, independently normalizes each 1-D slice along dimension dim.
# 人脸图片对应的最终编码, 也是算法的核心输出
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
# Wrapper for Graph.add_to_collection() using the default graph.
# Stores value in the collection with the given name.
# Note that collections are not sets, so it is possible to add a value to a collection several times.
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
# args.center_loss_factor center loss论文里的lambda
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch, logits=logits, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
# Adds all input tensors element-wise.
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss,
global_step,
args.optimizer,
learning_rate,
args.moving_average_decay,
tf.global_variables(),
args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
elif continue_ckpt_dir:
files = os.listdir(continue_ckpt_dir)
meta_files = [s for s in files if s.endswith('.meta')]
if len(meta_files) == 0:
raise ValueError('No meta file found in %s' % continue_ckpt_dir)
elif len(meta_files) > 1:
raise ValueError(
'There should not be more than one meta file in %s' % continue_ckpt_dir)
saver = tf.train.import_meta_graph(continue_ckpt_dir + "/" + meta_files[0])
latest_checkpoint = tf.train.latest_checkpoint(continue_ckpt_dir)
print('Restoring checkpoint: %s' % latest_checkpoint)
saver.restore(sess, latest_checkpoint)
# TODO: don't know why global_step is not saved. get it from the filename
last_step = int(os.path.basename(latest_checkpoint).split('-')[-1])
print('Checkpoint restored, last step is ', str(last_step))
sess.run(global_step.assign(last_step))
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch_size = args.epoch_size
epoch = step // epoch_size
if args.learning_rate > 0.0:
lr = args.learning_rate
else:
# Read the schedule file each epoch, you can change the file content during running
lr = facenet.get_learning_rate_from_file(args.learning_rate_schedule_file, epoch)
# Special value means stop
if lr == 0.0:
break
# Train for one epoch
train(args,
sess,
epoch,
image_list,
label_list,
index_dequeue_op,
enqueue_op,
image_paths_placeholder,
labels_placeholder,
learning_rate_placeholder,
phase_train_placeholder,
batch_size_placeholder,
global_step,
total_loss,
train_op,
summary_op,
summary_writer,
regularization_losses,
lr,
snapshot_at_step,
saver,
model_dir,
subdir
)
# Save variables and the metagraph if it doesn't exist already (step in filename is the next step after restore)
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step + epoch_size)
# Evaluate on LFW
if args.lfw_dir and args.lfw_epoch_interval > 0:
if epoch % args.lfw_epoch_interval == 0:
evaluate(sess,
enqueue_op,
image_paths_placeholder,
labels_placeholder,
phase_train_placeholder,
batch_size_placeholder,
embeddings,
label_batch,
lfw_paths,
actual_issame,
args.lfw_batch_size,
args.lfw_nrof_folds,
log_dir,
step,
summary_writer)
# Print current time
print("Current date time:", datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S'))
return model_dir
def main(args):
# args从文件系统中导入对象
network = importlib.import_module(
args.model_def) # 导入模型，default = 'models.inception_resnet_v1'
image_size = (args.image_size, args.image_size) # 导入尺寸，default = 160
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir),
subdir) # 导入运行日志目录，default = '~/logs/facenet'
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir),
subdir) # ，导入模型目录，default = '~/models/facenet'
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
stat_file_name = os.path.join(log_dir, 'stat.h5') # 统计数据，用h5格式保存
# Write arguments to a text file
facenet.write_arguments_to_file(args,
os.path.join(log_dir,
'arguments.txt')) # 记录命令至运行日志
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(
os.path.realpath(__file__)) # os.path.realpath(__file__)是脚本所在的绝对路径
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed) # default = 666
random.seed(args.seed)
dataset = facenet.get_dataset(
args.data_dir
) # 导入数据集目录，default = '~/datasets/casia/casia_maxpy_mtcnnalign_182_160'
if args.filter_filename: # default = ''
dataset = filter_dataset(
dataset,
os.path.expanduser(args.filter_filename),
# expanduser把path中包含的"~"和"~user"转换成用户目录.
args.filter_percentile,
args.filter_min_nrof_images_per_class
) # default = 100.0, default = 0
if args.validation_set_split_ratio > 0.0: # default = 0.0
train_set, val_set = facenet.split_dataset(
dataset, args.validation_set_split_ratio,
args.min_nrof_val_images_per_class, 'SPLIT_IMAGES')
else:
train_set, val_set = dataset, []
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(
args.lfw_pairs)) # default = 'data/pairs.txt'
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed) # 图级随机数种子
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The training set should not be empty'
val_image_list, val_label_list = facenet.get_image_paths_and_labels(
val_set)
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(
range_size,
num_epochs=None, # 产生1到range_size的对列，num_epochs不给值的话
shuffle=True,
seed=None,
capacity=32) # 则默认无限循环，后面用capacity限制容量
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue') # 使多个元素同时出列并命名
# 创建张量
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='labels')
control_placeholder = tf.placeholder(tf.int32,
shape=(None, 1),
name='control')
nrof_preprocess_threads = 4
# 先进先出对列
input_queue = data_flow_ops.FIFOQueue(
capacity=2000000,
dtypes=[tf.string, tf.int32, tf.int32],
shapes=[(1, ), (1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder, control_placeholder],
name='enqueue_op')
image_batch, label_batch = facenet.create_input_pipeline(
input_queue, image_size, nrof_preprocess_threads,
batch_size_placeholder)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Number of classes in training set: %d' % nrof_classes)
print('Number of examples in training set: %d' % len(image_list))
print('Number of classes in validation set: %d' % len(val_set))
print('Number of examples in validation set: %d' % len(val_image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None, # 相当于Dense
weights_initializer=slim.initializers.xavier_initializer(),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(
prelogits, 1, 1e-10, name='embeddings') # 按行进行L2正则化，同除各行L2范数
# Norm for the prelogits
eps = 1e-4
# 按行计算平均值。norm计算范数，abs取绝对值，ord取1，使用第一范数
prelogits_norm = tf.reduce_mean(
tf.norm(tf.abs(prelogits) + eps, ord=args.prelogits_norm_p,
axis=1))
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_norm *
args.prelogits_norm_loss_factor) # 将该种正则化方式储存至graphkeys
# Add center loss
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch,
args.center_loss_alfa,
nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
# 指数衰减。参数1：原始学习率。参数2：全局步数，每次自增1，如果不写则不增1.参数3：衰减系数。参数4：衰减速度。参数5：是否取整。True时取整
# 具体计算：decayed_learning_rate = learning_rate*decay_rate^(global_step/decay_steps)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate) # 保存数据
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# cast转换数据类型(布尔->浮点数)，equal比较两个向量相同位置的元素是否一样
correct_prediction = tf.cast(
tf.equal(tf.argmax(logits, 1), tf.cast(label_batch, tf.int64)),
tf.float32)
accuracy = tf.reduce_mean(correct_prediction) # 求平均值即准确率
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES) # 从graphkeys中把之前存储的loss计算方式取出来
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all(
) # 将所有summary全部保存到磁盘，以便tensorboard显示
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.
gpu_memory_fraction) # 使用gpu，参数为使用的显存上限
sess = tf.Session(config=tf.ConfigProto(
gpu_options=gpu_options, log_device_placement=False)) # 会话
sess.run(tf.global_variables_initializer()) # 返回所有全局变量
sess.run(tf.local_variables_initializer()) # 返回所有局部变量
summary_writer = tf.summary.FileWriter(
log_dir, sess.graph) # 将训练过程数据保存在filewriter指定的文件中
coord = tf.train.Coordinator() # 管理在Session中的多个线程
"""
启动入队线程，由多个或单个线程，按照设定规则，把文件读入Filename Queue中。函数返回线程ID的列表，一般情况下，系统有多少个核，
就会启动多少个入队线程（入队具体使用多少个线程在tf.train.batch中定义）
"""
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
nrof_steps = args.max_nrof_epochs * args.epoch_size
nrof_val_samples = int(
math.ceil(args.max_nrof_epochs / args.validate_every_n_epochs)
) # Validate every validate_every_n_epochs as well as in the last epoch
# ceil返回上入整数
stat = {
'loss':
np.zeros((nrof_steps, ), np.float32),
'center_loss':
np.zeros((nrof_steps, ), np.float32),
'reg_loss':
np.zeros((nrof_steps, ), np.float32),
'xent_loss':
np.zeros((nrof_steps, ), np.float32),
'prelogits_norm':
np.zeros((nrof_steps, ), np.float32),
'accuracy':
np.zeros((nrof_steps, ), np.float32),
'val_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_xent_loss':
np.zeros((nrof_val_samples, ), np.float32),
'val_accuracy':
np.zeros((nrof_val_samples, ), np.float32),
'lfw_accuracy':
np.zeros((args.max_nrof_epochs, ), np.float32),
'lfw_valrate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'learning_rate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_train':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_validate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'time_evaluate':
np.zeros((args.max_nrof_epochs, ), np.float32),
'prelogits_hist':
np.zeros((args.max_nrof_epochs, 1000), np.float32),
}
for epoch in range(1, args.max_nrof_epochs + 1):
step = sess.run(global_step, feed_dict=None)
# Train for one epoch
t = time.time()
cont = train(
args, sess, epoch, image_list, label_list,
index_dequeue_op, enqueue_op, image_paths_placeholder,
labels_placeholder, learning_rate_placeholder,
phase_train_placeholder, batch_size_placeholder,
control_placeholder, global_step, total_loss, train_op,
summary_op, summary_writer, regularization_losses,
args.learning_rate_schedule_file, stat, cross_entropy_mean,
accuracy, learning_rate, prelogits, prelogits_center_loss,
args.random_rotate, args.random_crop, args.random_flip,
prelogits_norm, args.prelogits_hist_max,
args.use_fixed_image_standardization)
stat['time_train'][epoch - 1] = time.time() - t
if not cont:
break
t = time.time()
if len(val_image_list) > 0 and (
(epoch - 1) % args.validate_every_n_epochs
== args.validate_every_n_epochs - 1
or epoch == args.max_nrof_epochs):
validate(args, sess, epoch, val_image_list, val_label_list,
enqueue_op, image_paths_placeholder,
labels_placeholder, control_placeholder,
phase_train_placeholder, batch_size_placeholder,
stat, total_loss, regularization_losses,
cross_entropy_mean, accuracy,
args.validate_every_n_epochs,
args.use_fixed_image_standardization)
stat['time_validate'][epoch - 1] = time.time() - t
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, epoch)
# Evaluate on LFW
t = time.time()
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, control_placeholder,
embeddings, label_batch, lfw_paths, actual_issame,
args.lfw_batch_size, args.lfw_nrof_folds, log_dir,
step, summary_writer, stat, epoch,
args.lfw_distance_metric, args.lfw_subtract_mean,
args.lfw_use_flipped_images,
args.use_fixed_image_standardization)
stat['time_evaluate'][epoch - 1] = time.time() - t
print('Saving statistics')
with h5py.File(stat_file_name, 'w') as f:
for key, value in stat.iteritems():
f.create_dataset(key, data=value)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
if args.filter_filename:
train_set = filter_dataset(train_set, os.path.expanduser(args.filter_filename),
args.filter_percentile, args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list)>0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size, num_epochs=None,
shuffle=True, seed=None, capacity=32)
index_dequeue_op = index_queue.dequeue_many(args.batch_size*args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string, shape=(None,1), name='image_paths')
labels_placeholder = tf.placeholder(tf.int64, shape=(None,1), name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1,), (1,)],
shared_name=None, name=None)
enqueue_op = input_queue.enqueue_many([image_paths_placeholder, labels_placeholder], name='enqueue_op')
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_png(file_contents)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image], tf.uint8)
if args.random_crop:
image = tf.random_crop(image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, label_batch = tf.train.batch_join(
images_and_labels, batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()], enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Build the inference graph
prelogits, _ = network.inference(image_batch, args.keep_probability,
phase_train=phase_train_placeholder, bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor>0.0:
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch, logits=logits, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.global_variables(), args.log_histograms)
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder, phase_train_placeholder, batch_size_placeholder, global_step,
total_loss, train_op, summary_op, summary_writer, regularization_losses, args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder,
embeddings, label_batch, lfw_paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, log_dir, step, summary_writer)
sess.close()
return model_dir
def main(args):
#network = importlib.import_module(args.model_def, 'inception_v3')
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
with open(os.path.join(model_dir, 'args.txt'), 'w') as f:
for arg in vars(args):
f.write(arg + ' ' + str(getattr(args, arg)) + '\n')
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
#train_set = facenet.get_dataset(args.data_dir)
train_set = facenet.get_dataset_with_enhanced(args.data_dir)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
# Read data and apply label preserving distortions
image_batch, label_batch = facenet.read_and_augument_data(
image_list, label_list, args.image_size, args.batch_size,
args.max_nrof_epochs, args.random_rotate, args.random_crop,
args.random_flip, args.nrof_preprocess_threads, args.padding_size,
args.patch_type)
#print('Total number of classes: %d' % len(train_set))
print('Total number of examples: %d' % len(image_list))
# Node for input images
image_batch = tf.identity(image_batch, name='input')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Placeholder for keep probability
keep_probability_placeholder = tf.placeholder(tf.float32,
name='keep_prob')
# Build the inference graph
# prelogits = network.inference(image_batch, keep_probability_placeholder,
# phase_train=phase_train_placeholder, weight_decay=args.weight_decay)
batch_norm_params = {
# Decay for the moving averages
'decay': 0.995,
# epsilon to prevent 0s in variance
'epsilon': 0.001,
# force in-place updates of mean and variance estimates
'updates_collections': None,
# Moving averages ends up in the trainable variables collection
'variables_collections': [tf.GraphKeys.TRAINABLE_VARIABLES],
# Only update statistics during training mode
'is_training': phase_train_placeholder
}
#prelogits, _ = network.inception_v3(image_batch, num_classes=len(train_set),is_training=True)
prelogits, _ = network.inference(image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
weight_decay=args.weight_decay)
#prelogits = tf.identity(prelogits, name="prelogits")
bottleneck = _fully_connected(prelogits,
args.embedding_size,
name='pre_embedding')
#bottleneck = tf.nn.l2_normalize(bottleneck, dim=1,name='embedding')
logits = _fully_connected_classifier(bottleneck,
len(train_set),
name='logits')
"""
bottleneck = slim.fully_connected(prelogits, args.embedding_size, activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
normalizer_fn=slim.batch_norm,
normalizer_params=batch_norm_params,
scope='Bottleneck', reuse=False)
logits = slim.fully_connected(bottleneck, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
logits = tf.identity(logits, name="logits")
"""
# Add DeCov regularization loss
if args.decov_loss_factor > 0.0:
logits_decov_loss = facenet.decov_loss(
logits) * args.decov_loss_factor
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES,
logits_decov_loss)
# Add center loss
update_centers = tf.no_op('update_centers')
if args.center_loss_factor > 0.0:
prelogits_center_loss, update_centers = facenet.center_loss(
bottleneck, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
#embeddings = tf.nn.l2_normalize(bottleneck, 1, 1e-10, name='embeddings')
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
"""
# Multi-label loss: sigmoid loss
sigmoid_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels=label_batch, logits=logits, name='sigmoid_loss_per_example')
sigmoid_loss_mean = tf.reduce_mean(sigmoid_loss, name='sigmoid_loss')
"""
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
#total_loss = tf.add_n([cross_entropy_mean], name='total_loss')
# prediction
prediction = tf.argmax(logits, axis=1, name='prediction')
acc = slim.metrics.accuracy(predictions=tf.cast(prediction,
dtype=tf.int32),
labels=tf.cast(label_batch,
dtype=tf.int32))
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables())
# Create a saver
# save_variables = list(set(tf.all_variables())-set([w])-set([b]))
save_variables = tf.trainable_variables()
saver = tf.train.Saver(save_variables, max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
# gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
# sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if pretrained_model:
saver.restore(sess, pretrained_model)
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, phase_train_placeholder,
learning_rate_placeholder, keep_probability_placeholder,
global_step, total_loss, acc, train_op, summary_op,
summary_writer, regularization_losses,
args.learning_rate_schedule_file, update_centers)
# Evaluate on LFW
if args.lfw_dir:
start_time = time.time()
_, _, accuracy, val, val_std, far = lfw.validate(
sess,
lfw_paths,
actual_issame,
args.seed,
args.batch_size,
image_batch,
phase_train_placeholder,
keep_probability_placeholder,
embeddings,
nrof_folds=args.lfw_nrof_folds)
print('Accuracy: %1.3f+-%1.3f' %
(np.mean(accuracy), np.std(accuracy)))
print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' %
(val, val_std, far))
lfw_time = time.time() - start_time
# Add validation loss and accuracy to summary
summary = tf.Summary()
#pylint: disable=maybe-no-member
summary.value.add(tag='lfw/accuracy',
simple_value=np.mean(accuracy))
summary.value.add(tag='lfw/val_rate', simple_value=val)
summary.value.add(tag='time/lfw', simple_value=lfw_time)
summary_writer.add_summary(summary, step)
with open(os.path.join(log_dir, 'lfw_result.txt'),
'at') as f:
f.write('%d\t%.5f\t%.5f\n' %
(step, np.mean(accuracy), val))
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
return model_dir
def main(args):
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model:
print('Pre-trained model: %s' %
os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 3),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 3),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3, ), (3, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder])
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
labels_batch = tf.identity(labels_batch, 'label_batch')
# Build the inference graph
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Split embeddings into anchor, positive and negative and calculate triplet loss
anchor, positive, negative = tf.unstack(
tf.reshape(embeddings, [-1, 3, args.embedding_size]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative,
args.alpha)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables())
# Create a saver
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer(),
feed_dict={phase_train_placeholder: True})
sess.run(tf.local_variables_initializer(),
feed_dict={phase_train_placeholder: True})
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
saver.restore(sess, os.path.expanduser(args.pretrained_model))
# facenet.load_model(args.pretrained_model)
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, train_set, epoch, image_paths_placeholder,
labels_placeholder, labels_batch, batch_size_placeholder,
learning_rate_placeholder, phase_train_placeholder,
enqueue_op, input_queue, global_step, embeddings,
total_loss, train_op, summary_op, summary_writer,
args.learning_rate_schedule_file, args.embedding_size,
anchor, positive, negative, triplet_loss)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, lfw_paths, embeddings, labels_batch,
image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder,
phase_train_placeholder, enqueue_op,
actual_issame, args.batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer, args.embedding_size)
return model_dir
def main(args):
#此处导入的是：models.inception_resnet_v1模型，以后再看怎么更改模型
network = importlib.import_module(args.model_def)
#用当前日期来命名模型
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
#日志保存在c:\\users\\Administrator\logs\facenet\ 文件夹里
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir) #没有日志文件就创建一个
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# 把参数写在日志文件中
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
#arg_string:'E:/facenet/train_tripletloss.py' output_dir:'C:\\Users\\Administrator/logs/facenet\\20180314-181556'
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
#获取数据集，train_set是包含文件路径与标签的集合
#先输入一个父路径 path：'E:/facenet/data/lfw_160'，接着输入每个子路径
# 输出：一个list，每个元素是一个ImageClass，里边包含图片地址的list（image_paths）以及对应的人名(name)[以后可能会直接调用这几个属性]
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model: #用在判断是否有预训练模型，但是如果有，怎么加载呢？
print('Pre-trained model: %s' %
os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
#建立图
#with语句适用于对资源进行访问的场合，确保使用过程中是否发生异常都会执行必要嘚瑟“清理”操作
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# 学习率 Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
#批大小
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
#用于判断是训练还是测试
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
#图像路径
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 3),
name='image_paths')
# 图像标签
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 3),
name='labels')
#新建一个队列，数据流操作，fifo先入先出
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(3, ), (3, )],
shared_name=None,
name=None)
#enqueue_many返回的是一个操作
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder])
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, labels_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
labels_batch = tf.identity(labels_batch, 'label_batch')
# Build the inference (构造计算图)
#其中prelogits是最后一层的输出
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
#L2正则化（范化）函数
# embeddings = tf.nn.l2_normalize(输入向量, L2范化的维数（取0（列L2范化）或1（行L2范化））, 泛化的最小值边界, name='embeddings')
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Split embeddings into anchor, positive and negative and calculate triplet loss
anchor, positive, negative = tf.unstack(
tf.reshape(embeddings, [-1, 3, args.embedding_size]), 3, 1)
triplet_loss = facenet.triplet_loss(anchor, positive, negative,
args.alpha)
#将指数衰减应用在学习率上
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# 计算损失
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
#构建L2正则化
total_loss = tf.add_n([triplet_loss] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
# 确定优化方法并根据损失函数求梯度，在这里，每更行一次参数，global_step会加1
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables())
# Create a saver创建一个saver用来保存或者从内存中回复一个模型参数
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.能够在GPU上分配的最大内存
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.global_variables_initializer(),
feed_dict={phase_train_placeholder: True})
sess.run(tf.local_variables_initializer(),
feed_dict={phase_train_placeholder: True})
#写log文件
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
#获取线程坐标
coord = tf.train.Coordinator()
#将队列中的多用sunner开始执行
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if args.pretrained_model:
print('Restoring pretrained model: %s' % args.pretrained_model)
#saver.restore(sess, os.path.expanduser(args.pretrained_model))
facenet.load_model(args.pretrained_model)
# Training and validation loop
epoch = 0
#将所有数据过一遍的次数 默认500
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
#epoch_size是一个epoch中批的个数，这个epoch是全局的批处理个数以一个epoch中。。。这个epoch将用于求学习率
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, train_set, epoch, image_paths_placeholder,
labels_placeholder, labels_batch, batch_size_placeholder,
learning_rate_placeholder, phase_train_placeholder,
enqueue_op, input_queue, global_step, embeddings,
total_loss, train_op, summary_op, summary_writer,
args.learning_rate_schedule_file, args.embedding_size,
anchor, positive, negative, triplet_loss)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, lfw_paths, embeddings, labels_batch,
image_paths_placeholder, labels_placeholder,
batch_size_placeholder, learning_rate_placeholder,
phase_train_placeholder, enqueue_op,
actual_issame, args.batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer, args.embedding_size)
return model_dir
def main(args):
network = importlib.import_module(args.model_def, 'inference')
if args.model_name:
subdir = args.model_name
preload_model = True
else:
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
preload_model = False
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir),
pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32,
shape=(None, args.image_size,
args.image_size, 3),
name='input')
# Placeholder for the learning rate
labels_placeholder = tf.placeholder(tf.int64, name='labels')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learing_rate')
# Build the inference graph
logits1, _ = network.inference(images_placeholder,
[128, len(train_set)],
args.keep_probability,
phase_train=phase_train_placeholder,
weight_decay=args.weight_decay)
# Split example embeddings into anchor, positive and negative and calculate triplet loss
embeddings = tf.nn.l2_normalize(logits1, 1, 1e-10, name='embeddings')
anchor, positive, negative = tf.split(0, 3, embeddings)
triplet_loss = facenet.triplet_loss(anchor, positive, negative,
args.alpha)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.scalar_summary('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_triplet_loss = tf.add_n([triplet_loss] + regularization_losses,
name='total_triplet_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
triplet_loss_train_op = facenet.train('tripletloss_',
total_triplet_loss, global_step,
args.optimizer, learning_rate,
args.moving_average_decay)
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=0)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
sess.run(init)
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
with sess.as_default():
if preload_model:
ckpt = tf.train.get_checkpoint_state(model_dir)
#pylint: disable=maybe-no-member
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
else:
raise ValueError('Checkpoint not found')
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
epoch = sess.run(global_step,
feed_dict=None) // args.epoch_size
# Train for one epoch
step = train_triplet_loss(
args, sess, train_set, epoch, images_placeholder,
labels_placeholder, phase_train_placeholder,
learning_rate_placeholder, global_step, embeddings,
total_triplet_loss, triplet_loss_train_op, summary_op,
summary_writer)
if args.lfw_dir:
_, _, accuracy, val, val_std, far = lfw.validate(
sess,
paths,
actual_issame,
args.seed,
args.batch_size,
images_placeholder,
phase_train_placeholder,
embeddings,
nrof_folds=args.lfw_nrof_folds)
print('Accuracy: %1.3f+-%1.3f' %
(np.mean(accuracy), np.std(accuracy)))
print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' %
(val, val_std, far))
# Add validation loss and accuracy to summary
summary = tf.Summary()
#pylint: disable=maybe-no-member
summary.value.add(tag='lfw/accuracy',
simple_value=np.mean(accuracy))
summary.value.add(tag='lfw/val_rate', simple_value=val)
summary_writer.add_summary(summary, step)
if (epoch % args.checkpoint_period
== 0) or (epoch == args.max_nrof_epochs - 1):
# Save the model checkpoint
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
return model_dir
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
# 创建模型文件夹
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
nrof_classes = len(train_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
if args.baihe_pack_file:
print('load baihe dataset')
lfw_paths, actual_issame = msgpack_numpy.load(open(args.baihe_pack_file))
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
# 迭代轮数, 不同的轮数可以使用不同的学习率
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
# Read data and apply label preserving distortions
image_batch, label_batch = facenet.read_and_augment_data(image_list, label_list, args.image_size,
args.batch_size, args.max_nrof_epochs, args.random_crop, args.random_flip, args.random_rotate,
args.nrof_preprocess_threads)
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
print('Building training graph')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
# Build the inference graph, 返回的是网络结构
prelogits, _ = network.inference(image_batch, args.keep_probability, phase_train=True,
weight_decay=args.weight_decay)
# 初始化采用截断的正态分布噪声, 标准差为0.1
# tf.truncated_normal_initializer(stddev=0.1)
logits = slim.fully_connected(prelogits, len(train_set), activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits', reuse=False)
# Add DeCov regularization loss
if args.decov_loss_factor > 0.0:
logits_decov_loss = facenet.decov_loss(logits) * args.decov_loss_factor
# 将decov_loss加入到名字为tf.GraphKeys.REGULARIZATION_LOSSES的集合当中来
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, logits_decov_loss)
# Add center loss (center_loss作为一个正则项加入到collections)
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(prelogits, label_batch, args.center_loss_alfa, nrof_classes)
# 将center加入到名字为tf.GraphKeys.REGULARIZATION_LOSSES的集合当中来
tf.add_to_collection(tf.GraphKeys.REGULARIZATION_LOSSES, prelogits_center_loss * args.center_loss_factor)
# 对学习率进行指数衰退
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.scalar_summary('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
# 将softmax和交叉熵一起做,得到最后的损失函数,提高效率
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
logits, label_batch, name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
# 获取正则loss
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, tf.all_variables(), args.log_histograms)
# Evaluation
print('Building evaluation graph')
lfw_label_list = range(0, len(lfw_paths))
assert (len(lfw_paths) % args.lfw_batch_size == 0), \
"The number of images in the LFW test set need to be divisible by the lfw_batch_size"
eval_image_batch, eval_label_batch = facenet.read_and_augment_data(lfw_paths, lfw_label_list, args.image_size,
args.lfw_batch_size, None, False, False,
False, args.nrof_preprocess_threads,
shuffle=False)
# Node for input images
eval_image_batch.set_shape((None, args.image_size, args.image_size, 3))
eval_image_batch = tf.identity(eval_image_batch, name='input')
eval_prelogits, _ = network.inference(eval_image_batch, 1.0,
phase_train=False, weight_decay=0.0, reuse=True)
eval_embeddings = tf.nn.l2_normalize(eval_prelogits, 1, 1e-10, name='embeddings')
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=10)
# saver = tf.train.Saver(tf.global_variables(), max_to_keep=10)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options, log_device_placement=False))
# sess.run(tf.global_variables_initializer())
# sess.run(tf.local_variables_initializer())
sess.run(tf.initialize_all_variables())
sess.run(tf.initialize_local_variables())
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
# 将队列runner启动，队列就开始运行，返回启动的线程
# 注意input_queue是先入列，再出列，由于入列的时候输入是place holder，因此到后的线程的时候，会阻塞，
# 直到下train中sess run （enqueue_op）的时候， 会向队列中载入值，后面的出列才有对象，才在各自的队列中开始执行
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
try:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
train(args, sess, epoch, learning_rate_placeholder, global_step, total_loss, train_op, summary_op,
summary_writer, regularization_losses, args.learning_rate_schedule_file)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer, model_dir, subdir, step)
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, eval_embeddings, eval_label_batch, actual_issame, args.lfw_batch_size, args.seed,
args.lfw_nrof_folds, log_dir, step, summary_writer)
# Evaluate on baihe_data
if args.baihe_pack_file:
evaluate(sess, eval_embeddings, eval_label_batch, actual_issame, args.lfw_batch_size, args.seed,
args.lfw_nrof_folds, log_dir, step, summary_writer)
except:
traceback.print_exc()
continue
return model_dir
def train():
dataset = facenet.get_dataset(FLAGS.data_dir)
train_set, test_set = facenet.split_dataset(dataset, 0.9)
fileName = "/home/david/debug4.h5"
f = h5py.File(fileName, "r")
for item in f.values():
print(item)
w1 = f['1w'][:]
b1 = f['1b'][:]
f.close()
print(w1.shape)
print(b1.shape)
"""Train CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(FLAGS.batch_size, 96, 96, 3), name='Input')
# Build a Graph that computes the logits predictions from the inference model
#embeddings = facenet.inference_nn4_max_pool_96(images_placeholder, phase_train=True)
conv1 = _conv(images_placeholder, 3, 64, 7, 7, 2, 2, 'SAME', 'conv1_7x7', phase_train=False, use_batch_norm=False, init_weight=w1, init_bias=b1)
resh1 = tf.reshape(conv1, [-1, 294912])
embeddings = _affine(resh1, 294912, 128)
# Split example embeddings into anchor, positive and negative
a, p, n = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(a, p, n)
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, grads = facenet.train(loss, global_step)
# Create a saver
saver = tf.train.Saver(tf.all_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess.run(init)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, graph_def=sess.graph_def)
epoch = 0
with sess.as_default():
while epoch<FLAGS.max_nrof_epochs:
batch_number = 0
while batch_number<FLAGS.epoch_size:
print('Loading new data')
image_data, num_per_class, image_paths = facenet.load_data(train_set)
print('Selecting suitable triplets for training')
start_time = time.time()
emb_list = []
# Run a forward pass for the sampled images
nrof_examples_per_epoch = FLAGS.people_per_batch*FLAGS.images_per_person
nrof_batches_per_epoch = int(np.floor(nrof_examples_per_epoch/FLAGS.batch_size))
if True:
for i in xrange(nrof_batches_per_epoch):
feed_dict, _ = facenet.get_batch(images_placeholder, image_data, i)
emb_list += sess.run([embeddings], feed_dict=feed_dict)
emb_array = np.vstack(emb_list) # Stack the embeddings to a nrof_examples_per_epoch x 128 matrix
# Select triplets based on the embeddings
apn, nrof_random_negs, nrof_triplets = facenet.select_triplets(emb_array, num_per_class, image_data)
duration = time.time() - start_time
print('(nrof_random_negs, nrof_triplets) = (%d, %d): time=%.3f seconds' % (nrof_random_negs, nrof_triplets, duration))
count = 0
while count<nrof_triplets*3 and batch_number<FLAGS.epoch_size:
start_time = time.time()
feed_dict, batch = facenet.get_batch(images_placeholder, apn, batch_number)
if (batch_number%20==0):
err, summary_str, _ = sess.run([loss, summary_op, train_op], feed_dict=feed_dict)
summary_writer.add_summary(summary_str, FLAGS.epoch_size*epoch+batch_number)
else:
err, _ = sess.run([loss, train_op], feed_dict=feed_dict)
duration = time.time() - start_time
print('Epoch: [%d][%d/%d]\tTime %.3f\ttripErr %2.3f' % (epoch, batch_number, FLAGS.epoch_size, duration, err))
batch_number+=1
count+=FLAGS.batch_size
else:
while batch_number<FLAGS.epoch_size:
start_time = time.time()
feed_dict, _ = facenet.get_batch(images_placeholder, image_data, batch_number)
grad_tensors, grad_vars = zip(*grads)
eval_list = (train_op, loss) + grad_tensors
result = sess.run(eval_list, feed_dict=feed_dict)
grads_eval = result[2:]
nrof_parameters = 0
for gt, gv in zip(grads_eval, grad_vars):
print('%40s: %6d' % (gv.op.name, np.size(gt)))
nrof_parameters += np.size(gt)
print('Total number of parameters: %d' % nrof_parameters)
err = result[1]
batch_number+=1
epoch+=1
# Save the model checkpoint periodically.
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=epoch*FLAGS.epoch_size+batch_number)
def main(args):
network = importlib.import_module(args.model_def, 'inference')
if args.model_name:
subdir = args.model_name
preload_model = True
else:
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
preload_model = False
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.mkdir(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.mkdir(model_dir)
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(None, args.image_size, args.image_size, 3), name='input')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Build the inference graph
embeddings = network.inference(images_placeholder, args.pool_type, args.use_lrn,
args.keep_probability, phase_train=phase_train_placeholder, weight_decay=args.weight_decay)
# Split example embeddings into anchor, positive and negative
anchor, positive, negative = tf.split(0, 3, embeddings)
# Calculate triplet loss
loss = facenet.triplet_loss(anchor, positive, negative, args.alpha)
# Calculate the total loss
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([loss] + regularization_losses, name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op, _ = facenet.train(total_loss, global_step, args.optimizer, args.learning_rate,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, args.moving_average_decay)
# Create a saver
saver = tf.train.Saver(tf.all_variables(), max_to_keep=0)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
sess = tf.Session()
sess.run(init)
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
with sess.as_default():
if preload_model:
ckpt = tf.train.get_checkpoint_state(model_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
else:
raise ValueError('Checkpoint not found')
# Training and validation loop
for epoch in range(args.max_nrof_epochs):
# Train for one epoch
step = train(args, sess, train_set, epoch, images_placeholder, phase_train_placeholder,
global_step, embeddings, loss, train_op, summary_op, summary_writer)
_, _, accuracy, val, val_std, far = lfw.validate(sess,
paths, actual_issame, args.seed, args.batch_size,
images_placeholder, phase_train_placeholder, embeddings, nrof_folds=args.lfw_nrof_folds)
print('Accuracy: %1.3f+-%1.3f' % (np.mean(accuracy), np.std(accuracy)))
print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' % (val, val_std, far))
# Add validation loss and accuracy to summary
summary = tf.Summary()
summary.value.add(tag='lfw/accuracy', simple_value=np.mean(accuracy))
summary.value.add(tag='lfw/val_rate', simple_value=val)
summary_writer.add_summary(summary, step)
if (epoch % args.checkpoint_period == 0) or (epoch==args.max_nrof_epochs-1):
# Save the model checkpoint
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
return model_dir
def main(args):
os.environ["CUDA_VISIBLE_DEVICES"] = '0'
network = importlib.import_module(args.model_def)
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(
log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(
model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Write arguments to a text file
facenet.write_arguments_to_file(args, os.path.join(log_dir,
'arguments.txt'))
# Store some git revision info in a text file in the log directory
src_path, _ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
random.seed(args.seed)
train_set = facenet.get_dataset(args.data_dir)
train_set, test_set = facenet.split_dataset(train_set,
0.8,
mode='SPLIT_IMAGES')
if args.filter_filename:
train_set = filter_dataset(train_set,
os.path.expanduser(args.filter_filename),
args.filter_percentile,
args.filter_min_nrof_images_per_class)
nrof_classes = len(train_set)
test_nrof_classes = len(test_set)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
pretrained_model = None
if args.pretrained_model:
pretrained_model = os.path.expanduser(args.pretrained_model)
# meta_file, ckpt_model=facenet.get_model_filenames(pretrained_model)
# pretrained_model=ckpt_model
print('Pre-trained model: %s' % pretrained_model)
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(
os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Get a list of image paths and their labels
image_list, label_list = facenet.get_image_paths_and_labels(train_set)
assert len(image_list) > 0, 'The dataset should not be empty'
test_image_list, test_label_list = facenet.get_image_paths_and_labels(
test_set)
assert len(test_image_list) > 0, 'The dataset should not be empty'
# Create a queue that produces indices into the image_list and label_list
labels = ops.convert_to_tensor(label_list, dtype=tf.int32)
range_size = array_ops.shape(labels)[0]
index_queue = tf.train.range_input_producer(range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
index_dequeue_op = index_queue.dequeue_many(
args.batch_size * args.epoch_size, 'index_dequeue')
learning_rate_placeholder = tf.placeholder(tf.float32,
name='learning_rate')
batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
image_paths_placeholder = tf.placeholder(tf.string,
shape=(None, 1),
name='image_paths')
labels_placeholder = tf.placeholder(tf.int64,
shape=(None, 1),
name='labels')
input_queue = data_flow_ops.FIFOQueue(capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
enqueue_op = input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder], name='enqueue_op')
nrof_preprocess_threads = 4
images_and_labels = []
for _ in range(nrof_preprocess_threads):
filenames, label = input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
#pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
images_and_labels.append([images, label])
image_batch, label_batch = tf.train.batch_join(
images_and_labels,
batch_size=batch_size_placeholder,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * nrof_preprocess_threads * args.batch_size,
allow_smaller_final_batch=True)
image_batch = tf.identity(image_batch, 'image_batch')
image_batch = tf.identity(image_batch, 'input')
label_batch = tf.identity(label_batch, 'label_batch')
print('Total number of classes: %d' % nrof_classes)
print('Total number of examples: %d' % len(image_list))
# Create a queue that produces indices into the test_image_list and test_label_list
test_labels = ops.convert_to_tensor(test_label_list, dtype=tf.int32)
test_range_size = array_ops.shape(test_labels)[0]
test_index_queue = tf.train.range_input_producer(test_range_size,
num_epochs=None,
shuffle=True,
seed=None,
capacity=32)
# global test_batch_size
# test_batch_size=args.batch_size
test_index_dequeue_op = test_index_queue.dequeue_many(
test_batch_size * test_epoch_size, 'test_index_dequeue')
test_input_queue = data_flow_ops.FIFOQueue(
capacity=100000,
dtypes=[tf.string, tf.int64],
shapes=[(1, ), (1, )],
shared_name=None,
name=None)
test_enqueue_op = test_input_queue.enqueue_many(
[image_paths_placeholder, labels_placeholder],
name='test_enqueue_op')
test_nrof_preprocess_threads = 4
test_images_and_labels = []
for _ in range(test_nrof_preprocess_threads):
filenames, label = test_input_queue.dequeue()
images = []
for filename in tf.unstack(filenames):
file_contents = tf.read_file(filename)
image = tf.image.decode_image(file_contents, channels=3)
if args.random_rotate:
image = tf.py_func(facenet.random_rotate_image, [image],
tf.uint8)
if args.random_crop:
image = tf.random_crop(
image, [args.image_size, args.image_size, 3])
else:
image = tf.image.resize_image_with_crop_or_pad(
image, args.image_size, args.image_size)
if args.random_flip:
image = tf.image.random_flip_left_right(image)
# pylint: disable=no-member
image.set_shape((args.image_size, args.image_size, 3))
images.append(tf.image.per_image_standardization(image))
test_images_and_labels.append([images, label])
test_image_batch, test_label_batch = tf.train.batch_join(
test_images_and_labels,
batch_size=test_batch_size,
shapes=[(args.image_size, args.image_size, 3), ()],
enqueue_many=True,
capacity=4 * test_nrof_preprocess_threads * test_batch_size,
allow_smaller_final_batch=True)
test_image_batch = tf.identity(test_image_batch, 'test_image_batch')
test_image_batch = tf.identity(test_image_batch, 'test_input')
test_label_batch = tf.identity(test_label_batch, 'test_label_batch')
print('Total number of test classes: %d' % test_nrof_classes)
print('Total number of test examples: %d' % len(test_image_list))
image_input_placeholder = tf.get_default_graph().get_tensor_by_name(
"input:0")
label_input_placeholder = tf.get_default_graph().get_tensor_by_name(
"label_batch:0")
print('Building training graph')
# Build the inference graph
print("embeddings size is: %s" % (str(args.embedding_size)))
prelogits, _ = network.inference(
image_batch,
args.keep_probability,
phase_train=phase_train_placeholder,
bottleneck_layer_size=args.embedding_size,
weight_decay=args.weight_decay)
logits = slim.fully_connected(
prelogits,
len(train_set),
activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=0.1),
weights_regularizer=slim.l2_regularizer(args.weight_decay),
scope='Logits',
reuse=False)
embeddings = tf.nn.l2_normalize(prelogits, 1, 1e-10, name='embeddings')
# Add center loss
if args.center_loss_factor > 0.0:
prelogits_center_loss, _ = facenet.center_loss(
prelogits, label_batch, args.center_loss_alfa, nrof_classes)
tf.add_to_collection(
tf.GraphKeys.REGULARIZATION_LOSSES,
prelogits_center_loss * args.center_loss_factor)
learning_rate = tf.train.exponential_decay(
learning_rate_placeholder,
global_step,
args.learning_rate_decay_epochs * args.epoch_size,
args.learning_rate_decay_factor,
staircase=True)
tf.summary.scalar('learning_rate', learning_rate)
# Calculate the average cross entropy loss across the batch
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=label_batch,
logits=logits,
name='cross_entropy_per_example')
cross_entropy_mean = tf.reduce_mean(cross_entropy,
name='cross_entropy')
tf.add_to_collection('losses', cross_entropy_mean)
# Calculate the total losses
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([cross_entropy_mean] + regularization_losses,
name='total_loss')
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay,
tf.global_variables(), args.log_histograms)
# Create a saver
if args.pretrained_model:
variables = []
for v in tf.trainable_variables():
if v.name.startswith("Logits") or v.name.startswith(
"InceptionResnetV1/Block8") or v.name.startswith(
"InceptionResnetV1/Block17"):
print("skip variable %s" % v.name)
continue
else:
print("var name %s" % v.name)
variables.append(v)
# if not v.name.startswith("Logits") and not v.name.startswith("Block8"):
# variables.append(v)
saver = tf.train.Saver(variables, max_to_keep=3)
else:
saver = tf.train.Saver(tf.trainable_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options,
log_device_placement=False))
sess.run(tf.global_variables_initializer())
sess.run(tf.local_variables_initializer())
summary_writer = tf.summary.FileWriter(log_dir, sess.graph)
coord = tf.train.Coordinator()
tf.train.start_queue_runners(coord=coord, sess=sess)
with sess.as_default():
if pretrained_model:
print('Restoring pretrained model: %s' % pretrained_model)
saver.restore(sess, pretrained_model)
# Training and validation loop
print('Running training')
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
#epoch = step // args.epoch_size
# Train for one epoch
step=train(args, sess, epoch, image_list, label_list, index_dequeue_op, enqueue_op, image_paths_placeholder, labels_placeholder,
learning_rate_placeholder, phase_train_placeholder, batch_size_placeholder, global_step,
total_loss, train_op, summary_op, summary_writer, regularization_losses, args.learning_rate_schedule_file\
,test_image_list, test_label_list, test_index_dequeue_op, test_enqueue_op,image_batch, label_batch)
# Save variables and the metagraph if it doesn't exist already
save_variables_and_metagraph(sess, saver, summary_writer,
model_dir, subdir, step)
epoch += 1
# Evaluate on LFW
if args.lfw_dir:
evaluate(sess, enqueue_op, image_paths_placeholder,
labels_placeholder, phase_train_placeholder,
batch_size_placeholder, embeddings, label_batch,
lfw_paths, actual_issame, args.lfw_batch_size,
args.lfw_nrof_folds, log_dir, step,
summary_writer)
# constant_graph = graph_util.convert_variables_to_constants(sess, sess.graph_def, ["embeddings"])
# with tf.gfile.FastGFile(model_dir + '/video-faces_%d_model.pb'%args.max_nrof_epochs, mode='wb') as f:
# f.write(constant_graph.SerializeToString())
return model_dir
def main(args):
network = importlib.import_module(args.model_def, 'inference')
subdir = datetime.strftime(datetime.now(), '%Y%m%d-%H%M%S')
log_dir = os.path.join(os.path.expanduser(args.logs_base_dir), subdir)
if not os.path.isdir(log_dir): # Create the log directory if it doesn't exist
os.makedirs(log_dir)
model_dir = os.path.join(os.path.expanduser(args.models_base_dir), subdir)
if not os.path.isdir(model_dir): # Create the model directory if it doesn't exist
os.makedirs(model_dir)
# Store some git revision info in a text file in the log directory
src_path,_ = os.path.split(os.path.realpath(__file__))
facenet.store_revision_info(src_path, log_dir, ' '.join(sys.argv))
np.random.seed(seed=args.seed)
train_set = facenet.get_dataset(args.data_dir)
print('Model directory: %s' % model_dir)
print('Log directory: %s' % log_dir)
if args.pretrained_model:
print('Pre-trained model: %s' % os.path.expanduser(args.pretrained_model))
if args.lfw_dir:
print('LFW directory: %s' % args.lfw_dir)
# Read the file containing the pairs used for testing
pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))
# Get the paths for the corresponding images
lfw_paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs, args.lfw_file_ext)
with tf.Graph().as_default():
tf.set_random_seed(args.seed)
global_step = tf.Variable(0, trainable=False)
# Placeholder for input images
images_placeholder = tf.placeholder(tf.float32, shape=(None, args.image_size, args.image_size, 3), name='input')
# Placeholder for phase_train
phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
# Placeholder for the learning rate
learning_rate_placeholder = tf.placeholder(tf.float32, name='learning_rate')
# Build the inference graph
prelogits, _ = network.inference(images_placeholder, args.keep_probability,
phase_train=True, weight_decay=args.weight_decay)
pre_embeddings = slim.fully_connected(prelogits, 128, activation_fn=None, scope='Embeddings', reuse=False)
# Split example embeddings into anchor, positive and negative and calculate triplet loss
embeddings = tf.nn.l2_normalize(pre_embeddings, 1, 1e-10, name='embeddings')
anchor, positive, negative = tf.split(0, 3, embeddings)
triplet_loss = facenet.triplet_loss(anchor, positive, negative, args.alpha)
learning_rate = tf.train.exponential_decay(learning_rate_placeholder, global_step,
args.learning_rate_decay_epochs*args.epoch_size, args.learning_rate_decay_factor, staircase=True)
tf.scalar_summary('learning_rate', learning_rate)
# Calculate the total losses
regularization_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
total_loss = tf.add_n([triplet_loss] + regularization_losses, name='total_loss')
# Create list with variables to restore
restore_vars = []
update_gradient_vars = []
if args.pretrained_model:
for var in tf.all_variables():
if not 'Embeddings/' in var.op.name:
restore_vars.append(var)
else:
update_gradient_vars.append(var)
else:
restore_vars = tf.all_variables()
update_gradient_vars = tf.all_variables()
# Build a Graph that trains the model with one batch of examples and updates the model parameters
train_op = facenet.train(total_loss, global_step, args.optimizer,
learning_rate, args.moving_average_decay, update_gradient_vars)
# Create a saver
restore_saver = tf.train.Saver(restore_vars)
saver = tf.train.Saver(tf.all_variables(), max_to_keep=3)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Start running operations on the Graph.
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=args.gpu_memory_fraction)
sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
# Initialize variables
sess.run(tf.initialize_all_variables())
sess.run(tf.initialize_local_variables())
summary_writer = tf.train.SummaryWriter(log_dir, sess.graph)
tf.train.start_queue_runners(sess=sess)
with sess.as_default():
if args.pretrained_model:
restore_saver.restore(sess, os.path.expanduser(args.pretrained_model))
# Training and validation loop
epoch = 0
while epoch < args.max_nrof_epochs:
step = sess.run(global_step, feed_dict=None)
epoch = step // args.epoch_size
# Train for one epoch
step = train(args, sess, train_set, epoch, images_placeholder,
learning_rate_placeholder, global_step, embeddings, total_loss, train_op, summary_op, summary_writer)
if args.lfw_dir:
_, _, accuracy, val, val_std, far = lfw.validate(sess, lfw_paths,
actual_issame, args.seed, 60, images_placeholder, phase_train_placeholder, embeddings, nrof_folds=args.lfw_nrof_folds)
print('Accuracy: %1.3f+-%1.3f' % (np.mean(accuracy), np.std(accuracy)))
print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' % (val, val_std, far))
# Add validation loss and accuracy to summary
summary = tf.Summary()
#pylint: disable=maybe-no-member
summary.value.add(tag='lfw/accuracy', simple_value=np.mean(accuracy))
summary.value.add(tag='lfw/val_rate', simple_value=val)
summary_writer.add_summary(summary, step)
# Save the model checkpoint
print('Saving checkpoint')
checkpoint_path = os.path.join(model_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
return model_dir
