def vgg_net(weights, image, is_train):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name='%s_w' % name)
current = utils.conv2d_basic(current, kernels, None)
current = utils.batch_norm(current,
kernels.get_shape()[3], is_train,
'%s_bn' % name)
elif kind == 'relu':
current = tf.nn.relu(current, name='%s' % name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
if name in [
'conv3_4', 'relu3_4', 'conv4_4', 'relu4_4', 'conv5_4',
'relu5_4'
]:
continue
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
# Convolutin 레이어일 경우
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
# MATLAB 파일의 행렬 순서를 tensorflow 행렬의 순서로 변환합니다.
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
# Activation 레이어일 경우
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
# Pooling 레이어리 경우
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4] #网络的类型,是conv还是relu或者pool.
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels,
# out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
# if FLAGS.debug:
# util.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
# added for resume better
global_iter_counter = tf.Variable(0, name='global_step', trainable=False)
net['global_step'] = global_iter_counter
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = np.transpose(kernels, (1, 0, 2, 3))
bias = bias.reshape(-1)
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
assert len(net) == len(layers)
return net
def vgg_net(weights, image):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3' # 'conv3_4', 'relu3_4', 'pool3',
# 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
# 'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
#
# 'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
# 'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = np.transpose(kernels, (1, 0, 2, 3))
bias = bias.reshape(-1)
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
assert len(net) == len(layers)
return net
def vgg_net(self, weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3', 'conv4_1',
'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3', 'relu4_3',
'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4',
'relu5_4')
self.net = {}
self.var_net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(
kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
self.var_net[name] = kernels
self.var_net[name + 'b'] = bias
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
self.net[name] = current
return self.net, self.var_net
def vgg_net(weights, image):
layers = (
'conv1_1',
'relu1_1',
'conv1_2',
'relu1_2',
'pool1', # output: 112*112*64
'conv2_1',
'relu2_1',
'conv2_2',
'relu2_2',
'pool2', # output: 56*56*128
'conv3_1',
'relu3_1',
'conv3_2',
'relu3_2',
'conv3_3', # output: 28*28*256
'relu3_3',
'conv3_4',
'relu3_4',
'pool3',
'conv4_1',
'relu4_1',
'conv4_2',
'relu4_2',
'conv4_3', # output: 14*14*512
'relu4_3',
'conv4_4',
'relu4_4',
'pool4',
'conv5_1',
'relu5_1',
'conv5_2',
'relu5_2',
'conv5_3', # output: 7*7*512
'relu5_3',
'conv5_4',
'relu5_4') # followed by three FN
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
'''
weights[i][0][0][0][0]:
<tf.Variable 'inference/conv1_1_w:0' shape=(3, 3, 3, 64) dtype=float32_ref>
<tf.Variable 'inference/conv1_1_b:0' shape=(64,) dtype=float32_ref>
<tf.Variable 'inference/conv1_2_w:0' shape=(3, 3, 64, 64) dtype=float32_ref>
<tf.Variable 'inference/conv1_2_b:0' shape=(64,) dtype=float32_ref>
<tf.Variable 'inference/conv2_1_w:0' shape=(3, 3, 64, 128) dtype=float32_ref>
<tf.Variable 'inference/conv2_1_b:0' shape=(128,) dtype=float32_ref>
<tf.Variable 'inference/conv2_2_w:0' shape=(3, 3, 128, 128) dtype=float32_ref>
<tf.Variable 'inference/conv2_2_b:0' shape=(128,) dtype=float32_ref>
'''
net = {}
current = image
for i, name in enumerate(
layers
): # 对于一个可迭代/可遍历的对象（如列表、字符串），enumerate将其组成一个索引序列，利用它可以同时获得索引和值
kind = name[:4]
num = name[4:]
if kind == 'conv' and num == '1_1':
W = utils.weight_variable(
[3, 3, 4, 64],
name=name + "_w") # [patch 7*7,insize 512, outsize 4096]
b = utils.bias_variable([64], name=name + "_b")
current = utils.conv2d_basic(current, W, b)
elif kind == 'conv' and num != '1_1':
kernels, bias = weights[i][0][0][0][0]
# print("kernels:",i,kernels)
# print kernels
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
# print(kernels)
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
# print(bias)
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
# def vgg_net(image):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image # [n,224,224,3]
for i, name in enumerate(layers):
kind = name[:4]
# kind2=name[:5]
if kind=='conv':#kind2 == 'conv1':
# '''
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)), name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
'''
current=tf.layers.conv2d(current,64,3,padding='same')
elif kind2=='conv2':
current = tf.layers.conv2d(current, 128, 3, padding='same')
elif kind2=='conv3':
current = tf.layers.conv2d(current, 256, 3, padding='same')
elif kind2=='conv4':
current = tf.layers.conv2d(current, 512, 3, padding='same')
elif kind2=='conv5':
current = tf.layers.conv2d(current, 512, 3, padding='same')
else:
pass
# '''
if kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net # [n,14,14,512]
def vgg_net(weights, image):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels,
# out_channels]
kernels = Utils.get_variable(np.transpose(
kernels, (1, 0, 2, 3)), name=name + "_w")
bias = Utils.get_variable(bias.reshape(-1), name=name + "_b")
current = Utils.conv2d_basic(current, kernels, bias) # 前向传播结果 current
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
# if FLAGS.debug:
# util.add_activation_summary(current)
elif kind == 'pool':
# vgg 的前5层的stride都是2，也就是前5层的size依次减小1倍
# 这里处理了前4层的stride，用的是平均池化
# 第5层的pool在下文的外部处理了，用的是最大池化
# pool1 size缩小2倍
# pool2 size缩小4倍
# pool3 size缩小8倍
# pool4 size缩小16倍
current = Utils.avg_pool_2x2(current)
net[name] = current
# added for resume better
global_iter_counter = tf.Variable(0, name='global_step', trainable=False)
net['global_step'] = global_iter_counter # 每层前向传播结果放在net中， 是一个字典
return net
def vgg_net(weights, image):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
if (name == 'conv1_1'):
kernel_shape = kernels.shape[:2] + (4, ) + kernels.shape[3:]
else:
kernel_shape = kernels.shape
bias_shape = bias.shape
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
new_kernel = np.zeros(kernel_shape)
new_kernel_shape = np.transpose(new_kernel, (1, 0, 2, 3)).shape
# print(f"new kernel shape: {new_kernel_shape}")
new_bias = np.zeros(bias_shape)
new_bias_shape = new_bias.reshape(-1).shape[0]
# print(f"new bias shape: {new_bias_shape}")
kernels = utils.weight_variable(shape=new_kernel_shape, name=name + "_w" )
bias = utils.bias_variable(shape=[new_bias_shape], name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
# print(f"VGG-19 {name} layer: {current.shape}")
return net
def vgg_net(weights, image):
layers = ('relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1', 'relu2_1',
'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1', 'conv3_2',
'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3', 'relu4_3',
'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1', 'conv5_2',
'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
#grayiterstart = 0
if GRAY_MODE:
i = 0
name = 'conv1_1' # layers[0]
kind = name[:4] #=conv
bias = weights[0][0][0][0][0][1]
kernels = weights[0][0][0][0][0][0]
#put the in_channels first, select 1st channel
#[w,h,in_channels,out_channels]
kernelstrans = np.array([np.transpose(kernels, (2, 0, 1, 3))[0]])
#[in_channels,w,h,out_channels]
kernels = np.transpose(kernelstrans, (2, 1, 0, 3))
#after line above : [h,w,in_channels,out_channels]
kernels = utils.get_variable(kernels, name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
#
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i + 1][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def inference_fully_convolutional(dataset):
'''
Fully convolutional inference on notMNIST dataset
:param datset: [batch_size, 28*28*1] tensor
:return: logits
'''
dataset_reshaped = tf.reshape(dataset, [-1, 28, 28, 1])
with tf.name_scope("conv1") as scope:
W_conv1 = utils.weight_variable_xavier_initialized([3, 3, 1, 32],
name="W_conv1")
b_conv1 = utils.bias_variable([32], name="b_conv1")
h_conv1 = tf.nn.relu(
utils.conv2d_strided(dataset_reshaped, W_conv1, b_conv1))
with tf.name_scope("conv2") as scope:
W_conv2 = utils.weight_variable_xavier_initialized([3, 3, 32, 64],
name="W_conv2")
b_conv2 = utils.bias_variable([64], name="b_conv2")
h_conv2 = tf.nn.relu(utils.conv2d_strided(h_conv1, W_conv2, b_conv2))
with tf.name_scope("conv3") as scope:
W_conv3 = utils.weight_variable_xavier_initialized([3, 3, 64, 128],
name="W_conv3")
b_conv3 = utils.bias_variable([128], name="b_conv3")
h_conv3 = tf.nn.relu(utils.conv2d_strided(h_conv2, W_conv3, b_conv3))
with tf.name_scope("conv4") as scope:
W_conv4 = utils.weight_variable_xavier_initialized([3, 3, 128, 256],
name="W_conv4")
b_conv4 = utils.bias_variable([256], name="b_conv4")
h_conv4 = tf.nn.relu(utils.conv2d_strided(h_conv3, W_conv4, b_conv4))
with tf.name_scope("conv5") as scope:
# W_conv5 = utils.weight_variable_xavier_initialized([2, 2, 256, 512], name="W_conv5")
# b_conv5 = utils.bias_variable([512], name="b_conv5")
# h_conv5 = tf.nn.relu(utils.conv2d_strided(h_conv4, W_conv5, b_conv5))
h_conv5 = utils.avg_pool_2x2(h_conv4)
with tf.name_scope("conv6") as scope:
W_conv6 = utils.weight_variable_xavier_initialized([1, 1, 256, 10],
name="W_conv6")
b_conv6 = utils.bias_variable([10], name="b_conv6")
logits = tf.nn.relu(utils.conv2d_basic(h_conv5, W_conv6, b_conv6))
print logits.get_shape()
logits = tf.reshape(logits, [-1, 10])
return logits
def vgg_dilate_net(weights, image):
layers = ('convDilate1_1', 'reluDilate1_1', 'convDilate1_2',
'reluDilate1_2', 'poolDilate1', 'convDilate2_1', 'reluDilate2_1',
'convDilate2_2', 'reluDilate2_2', 'poolDilate2', 'convDilate3_1',
'reluDilate3_1', 'convDilate3_2', 'reluDilate3_2',
'convDilate3_3', 'reluDilate3_3', 'convDilate3_4',
'reluDilate3_4', 'convDilate4_1', 'reluDilate4_1',
'convDilate4_2', 'reluDilate4_2', 'convDilate4_3',
'reluDilate4_3', 'convDilate4_4', 'reluDilate4_4',
'convDilate5_1', 'reluDilate5_1', 'convDilate5_2',
'reluDilate5_2', 'convDilate5_3', 'reluDilate5_3',
'convDilate5_4', 'reluDilate5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
rate = int(name[10:11])
if rate <= 2:
kernels, bias = weights[i][0][0][0][0]
elif rate > 2:
kernels, bias = weights[7][0][0][0][0]
if rate > 4:
rate = 3
if int(name[12:13]) == 4:
rate = 2
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.weight_variable(np.shape(
np.transpose(kernels, (1, 0, 2, 3))),
name=name + "_wDilate")
bias = utils.bias_variable(np.shape(bias.reshape(-1)),
name=name + "_bDilate")
current = utils.conv2d_dilate(current, 2**(rate - 1), kernels,
bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def inference_fully_convolutional(dataset):
'''
Fully convolutional inference on notMNIST dataset
:param datset: [batch_size, 28*28*1] tensor
:return: logits
'''
dataset_reshaped = tf.reshape(dataset, [-1, 28, 28, 1])
with tf.name_scope("conv1") as scope:
W_conv1 = utils.weight_variable_xavier_initialized([3, 3, 1, 32], name="W_conv1")
b_conv1 = utils.bias_variable([32], name="b_conv1")
h_conv1 = tf.nn.relu(utils.conv2d_strided(dataset_reshaped, W_conv1, b_conv1))
with tf.name_scope("conv2") as scope:
W_conv2 = utils.weight_variable_xavier_initialized([3, 3, 32, 64], name="W_conv2")
b_conv2 = utils.bias_variable([64], name="b_conv2")
h_conv2 = tf.nn.relu(utils.conv2d_strided(h_conv1, W_conv2, b_conv2))
with tf.name_scope("conv3") as scope:
W_conv3 = utils.weight_variable_xavier_initialized([3, 3, 64, 128], name="W_conv3")
b_conv3 = utils.bias_variable([128], name="b_conv3")
h_conv3 = tf.nn.relu(utils.conv2d_strided(h_conv2, W_conv3, b_conv3))
with tf.name_scope("conv4") as scope:
W_conv4 = utils.weight_variable_xavier_initialized([3, 3, 128, 256], name="W_conv4")
b_conv4 = utils.bias_variable([256], name="b_conv4")
h_conv4 = tf.nn.relu(utils.conv2d_strided(h_conv3, W_conv4, b_conv4))
with tf.name_scope("conv5") as scope:
# W_conv5 = utils.weight_variable_xavier_initialized([2, 2, 256, 512], name="W_conv5")
# b_conv5 = utils.bias_variable([512], name="b_conv5")
# h_conv5 = tf.nn.relu(utils.conv2d_strided(h_conv4, W_conv5, b_conv5))
h_conv5 = utils.avg_pool_2x2(h_conv4)
with tf.name_scope("conv6") as scope:
W_conv6 = utils.weight_variable_xavier_initialized([1, 1, 256, 10], name="W_conv6")
b_conv6 = utils.bias_variable([10], name="b_conv6")
logits = tf.nn.relu(utils.conv2d_basic(h_conv5, W_conv6, b_conv6))
print logits.get_shape()
logits = tf.reshape(logits, [-1, 10])
return logits
def vgg_net(weights, image, debug):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image
restore_vars = []
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
# Utilizing stored weights of ImageNet pretrained network to provide the correct shapes
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.weight_variable(np.transpose(kernels, (1, 0, 2, 3)).shape, name=name + "_w")
bias = utils.bias_variable(bias.reshape(-1).shape, name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
restore_vars += [kernels, bias]
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net, restore_vars
def vgg_net(weights, image):
# VGG网络前五大部分
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image # 预测图像
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w") # conv1_1_w
bias = utils.get_variable(bias.reshape(-1),
name=name + "_b") # conv1_1_b
current = utils.conv2d_basic(current, kernels,
bias) # 前向传播结果 current
elif kind == 'relu':
current = tf.nn.relu(current, name=name) # relu1_1
if FLAGS.debug: # 是否开启debug模式 true / false
utils.add_activation_summary(current) # 画图
elif kind == 'pool':
# vgg 的前5层的stride都是2，也就是前5层的size依次减小1倍
# 这里处理了前4层的stride，用的是平均池化
# 第5层的pool在下文的外部处理了，用的是最大池化
# pool1 size缩小2倍
# pool2 size缩小4倍
# pool3 size缩小8倍
# pool4 size缩小16倍
current = utils.avg_pool_2x2(current)
net[name] = current # 每层前向传播结果放在net中， 是一个字典
return net
def vgg_net(weights, image):
layers = (
'conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1',
'conv2_1', 'relu2_1', 'conv2_2', 'relu2_2', 'pool2',
'conv3_1', 'relu3_1', 'conv3_2', 'relu3_2', 'conv3_3',
'relu3_3', 'conv3_4', 'relu3_4', 'pool3',
'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4',
'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
'relu5_3', 'conv5_4', 'relu5_4'
)
net = {}
current = image
for i, name in enumerate(layers):
if name in ['conv3_4', 'relu3_4', 'conv4_4', 'relu4_4', 'conv5_4', 'relu5_4']:
continue
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)), name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
"""
:param weights: np matrix
:param image: tf place holder <- fed with np arrays
:return: a dict. key is the name of layer, value is the corresponding opration node in tf graph.
"""
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {}
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
# weight tf var
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
# bias tf var
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
# the output tf layer node
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
# average pooling is this correct?????
current = utils.avg_pool_2x2(current)
net[name] = current
return net
def vgg_net(weights, image):
## fcn的前五层网络就是vgg网络
layers = ('conv1_1', 'relu1_1', 'conv1_2', 'relu1_2', 'pool1', 'conv2_1',
'relu2_1', 'conv2_2', 'relu2_2', 'pool2', 'conv3_1', 'relu3_1',
'conv3_2', 'relu3_2', 'conv3_3', 'relu3_3', 'conv3_4', 'relu3_4',
'pool3', 'conv4_1', 'relu4_1', 'conv4_2', 'relu4_2', 'conv4_3',
'relu4_3', 'conv4_4', 'relu4_4', 'pool4', 'conv5_1', 'relu5_1',
'conv5_2', 'relu5_2', 'conv5_3', 'relu5_3', 'conv5_4', 'relu5_4')
net = {} #字典
current = image
for i, name in enumerate(layers):
kind = name[:4]
if kind == 'conv':
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)),
name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if FLAGS.debug:
utils.add_activation_summary(current)
elif kind == 'pool':
## vgg 的前5层的stride都是2，也就是前5层的size依次减小1倍
## 这里处理了前4层的stride，用的是平均池化
## 第5层的pool在下文的外部处理了，用的是最大池化
## pool1 size缩小2倍
## pool2 size缩小4倍
## pool3 size缩小8倍
## pool4 size缩小16倍
current = utils.avg_pool_2x2(current) ## 平均池化
net[name] = current
return net ## vgg每层的结果都保存再net中了
def compute_energy(image, keep_prob_conv, input_channels, output_channels,
scope):
with tf.variable_scope(scope):
#########################
# downsample using VGG #
#########################
conv1_1 = conv_layer(image, 3, input_channels, 64, "1_1")
conv1_2 = conv_layer(conv1_1, 3, 64, 64, "1_2")
pool1 = utils.max_pool_2x2(conv1_2)
drop1 = tf.nn.dropout(pool1, keep_prob=keep_prob_conv)
conv2_1 = conv_layer(drop1, 3, 64, 128, "2_1")
conv2_2 = conv_layer(conv2_1, 3, 128, 128, "2_2")
pool2 = utils.max_pool_2x2(conv2_2)
drop2 = tf.nn.dropout(pool2, keep_prob=keep_prob_conv)
conv3_1 = conv_layer(drop2, 3, 128, 256, "3_1")
conv3_2 = conv_layer(conv3_1, 3, 256, 256, "3_2")
conv3_3 = conv_layer(conv3_2, 3, 256, 256, "3_3")
pool3 = utils.avg_pool_2x2(conv3_3)
drop3 = tf.nn.dropout(pool3, keep_prob=keep_prob_conv)
conv4_1 = conv_layer(drop3, 3, 256, 512, "4_1")
conv4_2 = conv_layer(conv4_1, 3, 512, 512, "4_2")
conv4_3 = conv_layer(conv4_2, 3, 512, 512, "4_3")
pool4 = utils.avg_pool_2x2(conv4_3)
drop4 = tf.nn.dropout(pool4, keep_prob=keep_prob_conv)
conv5_1 = conv_layer(drop4, 3, 512, 512, "5_1")
conv5_2 = conv_layer(conv5_1, 3, 512, 512, "5_2")
conv5_3 = conv_layer(conv5_2, 3, 512, 512, "5_3")
#######################
# Upsample
#######################
# fcn conv5 5x5, 1x1, 1x1
fcn5_1 = conv_layer(conv5_3, 5, 512, 512, "f5_1")
fcn5_2 = conv_layer(fcn5_1, 1, 512, 512, "f5_2")
fcn5_3 = conv_layer(fcn5_2, 1, 512, 512, "f5_3")
# fcn conv 4 5x5, 1x1, 1x1
fcn4_1 = conv_layer(drop4, 5, 512, 512, "f4_1")
fcn4_2 = conv_layer(fcn4_1, 1, 512, 512, "f4_2")
fcn4_3 = conv_layer(fcn4_2, 1, 512, 512, "f4_3")
# fcn conv 3 5x5, 1x1, 1x1
fcn3_1 = conv_layer(drop3, 5, 256, 256, "f3_1")
fcn3_2 = conv_layer(fcn3_1, 1, 256, 256, "f3_2")
fcn3_3 = conv_layer(fcn3_2, 1, 256, 256, "f3_3")
# upsample conv5
shape_fc5_1 = tf.shape(fcn5_3)
deconv_shape5_1 = tf.stack(
[shape_fc5_1[0], shape_fc5_1[1] * 2, shape_fc5_1[2] * 2, 256])
deconv5_1 = deconv_layer(fcn5_3, 4, 512, 256, deconv_shape5_1, "d5_1")
# upsample conv4
shape_fc4_3 = tf.shape(fcn4_3)
deconv_shape4_1 = tf.stack(
[shape_fc4_3[0], shape_fc4_3[1] * 2, shape_fc4_3[2] * 2, 256])
deconv4_1 = deconv_layer(fcn4_3, 4, 512, 256, deconv_shape4_1, "d4_1")
# stack
stacked = tf.concat([fcn3_3, deconv4_1, deconv5_1], -1)
# three 1by1 fuses
fuse1 = conv_layer(stacked, 1, 3 * 256, 2 * 256, "fuse1")
fuse2 = conv_layer(fuse1, 1, 2 * 256, 256, "fuse2")
fuse3 = conv_layer(fuse2, 1, 256, 256, "fuse3")
# final upsampling
shape_final1 = tf.shape(fuse3)
deconv_shape_final1 = tf.stack([
shape_final1[0], shape_final1[1] * 2, shape_final1[2] * 2,
output_channels
])
deconv_final1 = deconv_layer(fuse3, 4, 256, output_channels,
deconv_shape_final1, "d_final1")
shape_final2 = tf.shape(deconv_final1)
deconv_shape_final2 = tf.stack([
shape_final2[0], shape_final2[1] * 2, shape_final2[2] * 2,
output_channels
])
deconv_final2 = deconv_layer(deconv_final1, 4, output_channels,
output_channels, deconv_shape_final2,
"d_final2")
shape_final3 = tf.shape(deconv_final2)
deconv_shape_final3 = tf.stack([
shape_final3[0], shape_final3[1] * 2, shape_final3[2] * 2,
output_channels
])
deconv_final3 = deconv_layer(deconv_final2, 4, output_channels,
output_channels, deconv_shape_final3,
"d_final3")
shape_final4 = tf.shape(deconv_final3)
deconv_shape_final4 = tf.stack([
shape_final4[0], shape_final4[1] * 2, shape_final4[2] * 2,
output_channels
])
deconv_final4 = deconv_layer(deconv_final3, 4, output_channels,
output_channels, deconv_shape_final4,
"d_final4")
annotation_pred = tf.argmax(deconv_final3, dimension=3, name="prediction")
return tf.expand_dims(annotation_pred, dim=3), deconv_final3
def vgg_net_rgb(weights, image, debug, keep_prob):
layers = (
'rgb_conv1_1', 'rgb_relu1_1', 'rgb_conv1_2', 'rgb_relu1_2', 'rgb_pool1',
'rgb_conv2_1', 'rgb_relu2_1', 'rgb_conv2_2', 'rgb_relu2_2', 'rgb_pool2',
'rgb_conv3_1', 'rgb_relu3_1', 'rgb_conv3_2', 'rgb_relu3_2', 'rgb_conv3_3',
'rgb_relu3_3', 'rgb_pool3',
'rgb_conv4_1', 'rgb_relu4_1', 'rgb_conv4_2', 'rgb_relu4_2', 'rgb_conv4_3',
'rgb_relu4_3'
# 'conv5_1', 'relu5_1', 'conv5_2', 'relu5_2', 'conv5_3',
# 'relu5_3'
)
# output of retrained layer for vgg
net = {}
current = image
for i, name in enumerate(layers):
kind = name[4:8]
if kind == 'conv':
if name == 'rgb_conv4_1':
# Modify the senventh conv layer
# pdb.set_trace()
kernels, bias = weights[i][0][0][0][0]
kernels = np.transpose(kernels, (1, 0, 2, 3))
sample_index = random.sample(range(512), 256)
kernels = kernels[:, :, :, sample_index]
bias = bias[:, sample_index]
kernels = utils.get_variable(kernels, name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias, keep_prob)
elif name == 'rgb_conv4_2':
# Modify the eighth conv layer
# pdb.set_trace()
kernels, bias = weights[i][0][0][0][0]
kernels = np.transpose(kernels, (1, 0, 2, 3))
sample_index_1 = random.sample(range(512), 256)
sample_index_2 = random.sample(range(512), 256)
kernels = kernels[:, :, sample_index_1, :]
kernels = kernels[:, :, :, sample_index_2]
bias = bias[:, sample_index_2]
kernels = utils.get_variable(kernels, name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias, keep_prob)
elif name == 'rgb_conv4_3':
# pdb.set_trace()
# Modify the ninth conv layer
kernels, bias = weights[i][0][0][0][0]
kernels = np.transpose(kernels, (1, 0, 2, 3))
sample_index_1 = random.sample(range(512), 256)
sample_index_2 = random.sample(range(512), 256)
kernels = kernels[:, :, sample_index_1, :]
kernels = kernels[:, :, :, sample_index_2]
bias = bias[:, sample_index_2]
kernels = utils.get_variable(kernels, name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias, keep_prob)
else:
kernels, bias = weights[i][0][0][0][0]
# matconvnet: weights are [width, height, in_channels, out_channels]
# tensorflow: weights are [height, width, in_channels, out_channels]
kernels = utils.get_variable(np.transpose(kernels, (1, 0, 2, 3)), name=name + "_w")
bias = utils.get_variable(bias.reshape(-1), name=name + "_b")
current = utils.conv2d_basic(current, kernels, bias, keep_prob)
elif kind == 'relu':
current = tf.nn.relu(current, name=name)
if debug:
utils.add_activation_summary(current)
elif kind == 'pool':
current = utils.avg_pool_2x2(current)
net[name] = current
# pdb.set_trace()
return net
