def graphDef():
x = tf.placeholder(name='img_inputs', shape=[None, 112, 112, 3], dtype=tf.float32)
w_init_method = tf.contrib.layers.xavier_initializer(uniform=False)
w_init = w_init_method
net = get_resnet(x, 100, type='ir', w_init=w_init_method, trainable=False, keep_rate=1.0)
with tf.Session() as sess:
graph_def = sess.graph.as_graph_def()
with tf.gfile.FastGFile('./model/test.pb', 'wb') as f:
f.write(graph_def.SerializeToString())
name='lr_schedule')
#dp_rate = tf.train.piecewise_constant(global_step, boundaries=lr_steps, values=[0.4, 0.6, 0.8], name='dp_schedule')
# 3.3 define the optimize method
opt = tf.train.MomentumOptimizer(learning_rate=lr, momentum=args.momentum)
# Calculate the gradients for each model tower.
tower_grads = []
tl.layers.set_name_reuse(True)
loss_dict = {}
drop_dict = {}
loss_keys = []
with tf.variable_scope(tf.get_variable_scope()):
for i in range(args.num_gpus):
with tf.device('/gpu:%d' % i):
with tf.name_scope('%s_%d' % (args.tower_name, i)) as scope:
net = get_resnet(images_s[i], args.net_depth, type='ir', w_init=w_init_method, trainable=trainable, keep_rate= args.dropout_rate)
logit = arcface_loss(embedding=net.outputs, labels=labels_s[i], w_init=w_init_method, out_num=args.num_output)
# Reuse variables for the next tower.
tf.get_variable_scope().reuse_variables()
# define the cross entropy
inference_loss = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logit, labels=labels_s[i]))
# define weight deacy losses
wd_loss = 0
for weights in tl.layers.get_variables_with_name('W_conv2d', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(weights)
for W in tl.layers.get_variables_with_name('resnet_v1_50/E_DenseLayer/W', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(W)
for weights in tl.layers.get_variables_with_name('embedding_weights', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(weights)
for gamma in tl.layers.get_variables_with_name('gamma', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(gamma)
# 3.3 define the optimize method
opt = tf.train.MomentumOptimizer(learning_rate=lr, momentum=args.momentum)
# Calculate the gradients for each model tower.
tower_grads = []
tl.layers.set_name_reuse(True)
loss_dict = {}
drop_dict = {}
loss_keys = []
with tf.variable_scope(tf.get_variable_scope()):
for i in range(args.num_gpus):
with tf.device('/gpu:%d' % i):
with tf.name_scope('%s_%d' % (args.tower_name, i)) as scope:
net = get_resnet(images_s[i],
args.net_depth,
type='ir',
w_init=w_init_method,
trainable=trainable)
logit = arcface_loss(embedding=net.outputs,
labels=labels_s[i],
w_init=w_init_method,
out_num=args.num_output)
# Reuse variables for the next tower.
tf.get_variable_scope().reuse_variables()
# define the cross entropy
inference_loss = tf.reduce_mean(
tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=logit, labels=labels_s[i]))
# define weight deacy losses
wd_loss = 0
for weights in tl.layers.get_variables_with_name(
# 3.3 define the optimize method
opt = tf.train.MomentumOptimizer(learning_rate=lr, momentum=args.momentum)
# Calculate the gradients for each model tower.
tower_grads = []
tl.layers.set_name_reuse(True)
loss_dict = {}
drop_dict = {}
loss_keys = []
with tf.variable_scope(tf.get_variable_scope()):
for i in range(args.num_gpus):
with tf.device('/cpu:0'):
with tf.name_scope('%s_%d' % (args.tower_name, i)) as scope:
net = get_resnet(images_s[i],
args.net_depth,
type='ir',
w_init=w_init_method,
trainable=True,
keep_rate=dropout_rate)
logit = arcface_loss(embedding=net.outputs,
labels=labels_s[i],
w_init=w_init_method,
out_num=args.num_output)
# Reuse variables for the next tower.
tf.get_variable_scope().reuse_variables()
# define the cross entropy
inference_loss = tf.reduce_mean(
tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=logit, labels=labels_s[i]))
# define weight deacy losses
wd_loss = 0
for weights in tl.layers.get_variables_with_name(
lr = tf.train.piecewise_constant(global_step, boundaries=lr_steps, values=[0.001, 0.0005, 0.0003, 0.0001],
name='lr_schedule')
# 3.3 define the optimize method
opt = tf.train.MomentumOptimizer(learning_rate=lr, momentum=args.momentum)
# Calculate the gradients for each model tower.
tower_grads = []
tl.layers.set_name_reuse(True)
loss_dict = {}
drop_dict = {}
loss_keys = []
with tf.variable_scope(tf.get_variable_scope()):
for i in range(args.num_gpus):
with tf.device('/gpu:%d' % i):
with tf.name_scope('%s_%d' % (args.tower_name, i)) as scope:
net = get_resnet(images_s[i], args.net_depth, type='ir', w_init=w_init_method, trainable=True, keep_rate=dropout_rate)
logit = arcface_loss(embedding=net.outputs, labels=labels_s[i], w_init=w_init_method, out_num=args.num_output)
# Reuse variables for the next tower.
tf.get_variable_scope().reuse_variables()
# define the cross entropy
inference_loss = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logit, labels=labels_s[i]))
# define weight deacy losses
wd_loss = 0
for weights in tl.layers.get_variables_with_name('W_conv2d', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(weights)
for W in tl.layers.get_variables_with_name('resnet_v1_50/E_DenseLayer/W', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(W)
for weights in tl.layers.get_variables_with_name('embedding_weights', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(weights)
for gamma in tl.layers.get_variables_with_name('gamma', True, True):
wd_loss += tf.contrib.layers.l2_regularizer(args.weight_deacy)(gamma)
data_set = load_bin(db, args.image_size, args)
ver_list.append(data_set)
ver_name_list.append(db)
images = tf.placeholder(name='img_inputs',
shape=[None, *args.image_size, 3],
dtype=tf.float32)
labels = tf.placeholder(name='img_labels', shape=[
None,
], dtype=tf.int64)
dropout_rate = tf.placeholder(name='dropout_rate', dtype=tf.float32)
w_init_method = tf.contrib.layers.xavier_initializer(uniform=False)
net = get_resnet(images,
args.net_depth,
type='ir',
w_init=w_init_method,
trainable=False,
keep_rate=0.5)
embedding_tensor = net.outputs
# mv_mean = tl.layers.get_variables_with_name('resnet_v1_50/bn0/moving_mean', False, True)[0]
# 3.2 get arcface loss
logit = arcface_loss(embedding=net.outputs,
labels=labels,
w_init=w_init_method,
out_num=args.num_output)
sess = tf.Session()
saver = tf.train.Saver()
result_index = []
for file_index in args.ckpt_index_list:
ver_name_list.append(db)
images = tf.placeholder(name='img_inputs',
shape=[None, *args.image_size, 3],
dtype=tf.float32) # (?, 112, 112, 3)
labels = tf.placeholder(name='img_labels', shape=[
None,
], dtype=tf.int64) # (?, )
dropout_rate = tf.placeholder(name='dropout_rate', dtype=tf.float32)
w_init_method = tf.contrib.layers.xavier_initializer(
uniform=False) # 随机初始化权重先把空架子搭起来，后续再往里面restore train好的权重
net = get_resnet(
images,
args.net_depth,
type='ir',
w_init=w_init_method,
trainable=False,
keep_rate=dropout_rate) # L_Resnet_E_IR (?, 112,112,3)>(?, 512)
embedding_tensor = net.outputs
# mv_mean = tl.layers.get_variables_with_name('resnet_v1_50/bn0/moving_mean', False, True)[0]
# 3.2 get arcface loss
logit = arcface_loss(
embedding=net.outputs,
labels=labels,
w_init=w_init_method,
out_num=args.num_output) # (?, 512)，(?,)，initializer ，85164
gpu_config = tf.ConfigProto()
gpu_config.gpu_options.allow_growth = True
sess = tf.Session(config=gpu_config)
else:
is_seblock = 'ir'
# MGPU-start
# Calculate the gradients for each model tower.
tower_grads = [] # 保存来自不同GPU计算出的梯度、loss列表
tl.layers.set_name_reuse(True)
loss_dict = {} # 保存来自不同GPU计算出的梯度、loss列表
drop_dict = {}
loss_keys = []
with tf.variable_scope(tf.get_variable_scope()):
for iter_gpus in num_gpus:
with tf.device('/gpu:%d' % iter_gpus):
with tf.name_scope('%s_%d' % (tower_name, iter_gpus)) as scope:
net = get_resnet(images_s[iter_gpus],
net_depth,
type=is_seblock,
w_init=w_init_method,
trainable=True,
keep_rate=dropout_rate)
if is_arcloss == 1:
logit = arcface_loss(embedding=net.outputs,
labels=labels[iter_gpus],
w_init=w_init_method,
out_num=num_output,
s=loss_s,
m=loss_m)
else:
logit = cosineface_losses(embedding=net.outputs,
labels=labels_s[iter_gpus],
w_init=w_init_method,
out_num=num_output,
s=loss_s,