def stem(x):
"""
Stem fo the pure InceptionV4 and Inception-ResNet-V2
"""
out = conv3x3(x, 32, stride=2, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 32, stride=1, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 64, stride=1, padding="SAME", batch_norm=use_batch_norm)
out1 = max_pool_3x3(out, stride=2, padding="VALID")
out2 = conv3x3(out, 96, stride=2, padding="VALID", batch_norm=use_batch_norm)
out = concat(3, [out1, out2])
out1 = conv1x1(out, 64, batch_norm=use_batch_norm)
out1 = conv3x3(out1, 96, padding="VALID", batch_norm=use_batch_norm)
out2 = conv1x1(out, 64, batch_norm=use_batch_norm)
out2 = conv1x7(out2, 64, batch_norm=use_batch_norm)
out2 = conv7x1(out2, 64, batch_norm=use_batch_norm)
out2 = conv3x3(out2, 96, padding="VALID", batch_norm=use_batch_norm)
out = concat(3, [out1, out2])
out1 = conv3x3(out, 192, stride=2, padding="VALID", batch_norm=use_batch_norm)
out2 = max_pool_3x3(out, stride=2, padding="VALID")
out = concat(3, [out1, out2])
return out
def inception7B(out, num_3x3, num_d3x3_red, num_d3x3_1, num_d3x3_2):
out1 = conv3x3(out, num_3x3, stride=2, padding="VALID", batch_norm=use_batch_norm)
out2 = conv1x1(out, num_d3x3_red, batch_norm=use_batch_norm)
out2 = conv3x3(out2, num_d3x3_1, batch_norm=use_batch_norm)
out2 = conv3x3(out2, num_d3x3_2, stride=2, padding="VALID", batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def reductionA(out, k=0, l=0, m=0, n=0):
out1 = conv3x3(out, n, stride=2, padding="VALID", batch_norm=use_batch_norm)
out2 = conv1x1(out, k, batch_norm=use_batch_norm)
out2 = conv3x3(out2, l, batch_norm=use_batch_norm)
out2 = conv3x3(out2, m, stride=2, padding="VALID", batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def stem(x):
"""
Stem fo the pure InceptionV4 and Inception-ResNet-V2
"""
out = conv3x3(x, 32, stride=2, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out,
32,
stride=1,
padding="VALID",
batch_norm=use_batch_norm)
out = conv3x3(out, 64, stride=1, padding="SAME", batch_norm=use_batch_norm)
out1 = max_pool_3x3(out, stride=2, padding="VALID")
out2 = conv3x3(out,
96,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out = concat(3, [out1, out2])
out1 = conv1x1(out, 64, batch_norm=use_batch_norm)
out1 = conv3x3(out1, 96, padding="VALID", batch_norm=use_batch_norm)
out2 = conv1x1(out, 64, batch_norm=use_batch_norm)
out2 = conv1x7(out2, 64, batch_norm=use_batch_norm)
out2 = conv7x1(out2, 64, batch_norm=use_batch_norm)
out2 = conv3x3(out2, 96, padding="VALID", batch_norm=use_batch_norm)
out = concat(3, [out1, out2])
out1 = conv3x3(out,
192,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out2 = max_pool_3x3(out, stride=2, padding="VALID")
out = concat(3, [out1, out2])
return out
def inception7D(out, num_3x3_red, num_3x3, num_d7_3x3_red, num_d7_1, num_d7_2, num_d7_3x3):
out1 = conv1x1(out, num_3x3_red, batch_norm=use_batch_norm)
out1 = conv3x3(out1, num_3x3, stride=2, batch_norm=use_batch_norm)
out2 = conv1x1(out, num_d7_3x3_red, batch_norm=use_batch_norm)
out2 = conv1x7(out2, num_d7_1, batch_norm=use_batch_norm)
out2 = conv7x1(out2, num_d7_2, batch_norm=use_batch_norm)
out2 = conv3x3(out2, num_d7_3x3, stride=2, batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2)
return concat(3, [out1, out2, out3])
def reductionB(out):
out1 = conv1x1(out, 192, batch_norm=use_batch_norm)
out1 = conv3x3(out1, 192, stride=2, padding="VALID", batch_norm=use_batch_norm)
out2 = conv1x1(out, 256, batch_norm=use_batch_norm)
out2 = conv1x7(out2, 256, batch_norm=use_batch_norm)
out2 = conv7x1(out2, 320, batch_norm=use_batch_norm)
out2 = conv3x3(out2, 320, stride=2, padding="VALID", batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def inception7D(out, num_3x3_red, num_3x3, num_d7_3x3_red, num_d7_1, num_d7_2,
num_d7_3x3):
out1 = conv1x1(out, num_3x3_red, batch_norm=use_batch_norm)
out1 = conv3x3(out1, num_3x3, stride=2, batch_norm=use_batch_norm)
out2 = conv1x1(out, num_d7_3x3_red, batch_norm=use_batch_norm)
out2 = conv1x7(out2, num_d7_1, batch_norm=use_batch_norm)
out2 = conv7x1(out2, num_d7_2, batch_norm=use_batch_norm)
out2 = conv3x3(out2, num_d7_3x3, stride=2, batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2)
return concat(3, [out1, out2, out3])
def reductionA(out, k=0, l=0, m=0, n=0):
out1 = conv3x3(out,
n,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out2 = conv1x1(out, k, batch_norm=use_batch_norm)
out2 = conv3x3(out2, l, batch_norm=use_batch_norm)
out2 = conv3x3(out2,
m,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def inception7B(out, num_3x3, num_d3x3_red, num_d3x3_1, num_d3x3_2):
out1 = conv3x3(out,
num_3x3,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out2 = conv1x1(out, num_d3x3_red, batch_norm=use_batch_norm)
out2 = conv3x3(out2, num_d3x3_1, batch_norm=use_batch_norm)
out2 = conv3x3(out2,
num_d3x3_2,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def reductionB(out):
out1 = conv1x1(out, 192, batch_norm=use_batch_norm)
out1 = conv3x3(out1,
192,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out2 = conv1x1(out, 256, batch_norm=use_batch_norm)
out2 = conv1x7(out2, 256, batch_norm=use_batch_norm)
out2 = conv7x1(out2, 320, batch_norm=use_batch_norm)
out2 = conv3x3(out2,
320,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out3 = max_pool_3x3(out, stride=2, padding="VALID")
return concat(3, [out1, out2, out3])
def __init__(self, width=299, height=299, channels=3, classes=10):
super(InceptionV3, self).__init__()
self._number_of_classes = classes
size = width * height * channels
x = tf.placeholder(tf.float32, [None, size], name="x")
self._inputs = x
with tf.variable_scope("reshape1"):
x = tf.reshape(x, [-1, width, height, channels])
max_w = 299
min_w = 299 // 3
with tf.variable_scope("resize1"):
if width < min_w:
x = tf.image.resize_images(x, [min_w, min_w])
elif width > 299:
x = tf.image.resize_images(x, [max_w, max_w])
# Stage 1
out = conv3x3(x,
32,
stride=2,
padding="VALID",
batch_norm=use_batch_norm)
out = conv3x3(out, 32, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 64, batch_norm=use_batch_norm)
out = max_pool_3x3(out, padding="VALID")
# Stage 2
out = conv3x3(out, 80, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 192, padding="VALID", batch_norm=use_batch_norm)
out = max_pool_3x3(out, padding="VALID")
# Stage 3
out = inception7A(out, 64, 64, 96, 96, 48, 64, 32)
out = inception7A(out, 64, 64, 96, 96, 48, 64, 64)
out = inception7A(out, 64, 64, 96, 96, 48, 64, 64)
out = inception7B(out, 384, 64, 96, 96)
# Stage 4
out = inception7C(out, 192, 128, 128, 192, 128, 128, 128, 128, 192,
192)
out = inception7C(out, 192, 160, 160, 192, 160, 160, 160, 160, 192,
192)
out = inception7C(out, 192, 160, 160, 192, 160, 160, 160, 160, 192,
192)
out = inception7C(out, 192, 192, 192, 192, 192, 192, 192, 192, 192,
192)
out = inception7D(out, 192, 320, 192, 192, 192, 192)
# Stage 5
out = inception7E(out, 320, 384, 384, 384, 448, 384, 384, 384, 192,
avg_pool_3x3)
out = inception7E(out, 320, 384, 384, 384, 448, 384, 384, 384, 192,
max_pool_3x3)
out = tf.nn.avg_pool(out,
ksize=[1, 8, 8, 1],
strides=[1, 1, 1, 1],
padding='SAME')
dims = out.get_shape().as_list()
flatten_size = 1
for d in dims[1:]:
flatten_size *= d
with tf.variable_scope("reshape2"):
out = tf.reshape(out, [-1, int(flatten_size)])
with tf.variable_scope("fc1"):
out = fc(out, classes)
self._logits = out
if classes > 1:
self._predictions = tf.nn.softmax(self._logits)
else:
self._predictions = self._logits
def __init__(self, width, height, channels, classes):
super(AlexNet, self).__init__()
size = width * height * channels
x = tf.placeholder(tf.float32, [None, size], name="x")
self._inputs = x
with tf.variable_scope("reshape1"):
x = tf.reshape(self._inputs, [-1, width, height, channels])
self._number_of_classes = classes
with tf.variable_scope("resize1"):
if width < 224:
x = tf.image.resize_images(x, [48, 48])
else:
x = tf.image.resize_images(x, [224, 224])
with tf.variable_scope("conv1"):
conv1 = conv11x11(x, 96, stride=4, padding='VALID')
pool1 = max_pool_3x3(conv1, padding='VALID')
lrn1 = tf.nn.lrn(pool1, 5, bias=1.0, alpha=0.0001, beta=0.75)
with tf.variable_scope("conv2"):
conv2 = conv5x5(lrn1, 256)
pool2 = max_pool_3x3(conv2, padding='VALID')
lrn2 = tf.nn.lrn(pool2, 5, bias=1.0, alpha=0.0001, beta=0.75)
with tf.variable_scope("conv3"):
conv3 = conv3x3(lrn2, 384)
with tf.variable_scope("conv4"):
conv4 = conv3x3(conv3, 384)
with tf.variable_scope("conv5"):
conv5 = conv3x3(conv4, 256)
pool3 = max_pool_3x3(conv5)
lrn3 = tf.nn.lrn(pool3, 5, bias=1.0, alpha=0.001 / 9.0, beta=0.75)
dims = lrn3.get_shape().as_list()
flatten_size = 1
for d in dims[1:]:
flatten_size *= d
with tf.variable_scope("reshape2"):
flatten = tf.reshape(lrn3, [-1, int(flatten_size)])
with tf.variable_scope("fc1"):
fc1 = fc(flatten, 4096)
relu1 = tf.nn.relu(fc1)
with tf.variable_scope("fc2"):
fc2 = fc(relu1, 4096)
relu2 = tf.nn.relu(fc2)
with tf.variable_scope("fc3"):
y = fc(relu2, classes)
self._logits = y
if classes > 1:
self._predictions = tf.nn.softmax(y)
else:
self._predictions = self._logits
def __init__(self, width, height, channels, classes):
super(AlexNet, self).__init__()
size = width * height * channels
x = tf.placeholder(tf.float32, [None, size], name="x")
self._inputs = x
with tf.variable_scope("reshape1"):
x = tf.reshape(self._inputs, [-1, width, height, channels])
self._number_of_classes = classes
with tf.variable_scope("resize1"):
if width < 224:
x = tf.image.resize_images(x, [48, 48])
else:
x = tf.image.resize_images(x, [224, 224])
with tf.variable_scope("conv1"):
conv1 = conv11x11(x, 96, stride=4, padding='VALID')
pool1 = max_pool_3x3(conv1, padding='VALID')
lrn1 = tf.nn.lrn(pool1, 5, bias=1.0, alpha=0.0001, beta=0.75)
with tf.variable_scope("conv2"):
conv2 = conv5x5(lrn1, 256)
pool2 = max_pool_3x3(conv2, padding='VALID')
lrn2 = tf.nn.lrn(pool2, 5, bias=1.0, alpha=0.0001, beta=0.75)
with tf.variable_scope("conv3"):
conv3 = conv3x3(lrn2, 384)
with tf.variable_scope("conv4"):
conv4 = conv3x3(conv3, 384)
with tf.variable_scope("conv5"):
conv5 = conv3x3(conv4, 256)
pool3 = max_pool_3x3(conv5)
lrn3 = tf.nn.lrn(pool3, 5, bias=1.0, alpha=0.001 / 9.0, beta=0.75)
dims = lrn3.get_shape().as_list()
flatten_size = 1
for d in dims[1:]:
flatten_size *= d
with tf.variable_scope("reshape2"):
flatten = tf.reshape(lrn3, [-1, int(flatten_size)])
with tf.variable_scope("fc1"):
fc1 = fc(flatten, 4096)
relu1 = tf.nn.relu(fc1)
with tf.variable_scope("fc2"):
fc2 = fc(relu1, 4096)
relu2 = tf.nn.relu(fc2)
with tf.variable_scope("fc3"):
y = fc(relu2, classes)
self._logits = y
if classes > 1:
self._predictions = tf.nn.softmax(y)
else:
self._predictions = self._logits
def __init__(self, width=299, height=299, channels=3, classes=10):
super(InceptionV3, self).__init__()
self._number_of_classes = classes
size = width * height * channels
x = tf.placeholder(tf.float32, [None, size], name="x")
self._inputs = x
with tf.variable_scope("reshape1"):
x = tf.reshape(x, [-1, width, height, channels])
max_w = 299
min_w = 299 // 3
with tf.variable_scope("resize1"):
if width < min_w:
x = tf.image.resize_images(x, [min_w, min_w])
elif width > 299:
x = tf.image.resize_images(x, [max_w, max_w])
# Stage 1
out = conv3x3(x, 32, stride=2, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 32, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 64, batch_norm=use_batch_norm)
out = max_pool_3x3(out, padding="VALID")
# Stage 2
out = conv3x3(out, 80, padding="VALID", batch_norm=use_batch_norm)
out = conv3x3(out, 192, padding="VALID", batch_norm=use_batch_norm)
out = max_pool_3x3(out, padding="VALID")
# Stage 3
out = inception7A(out, 64, 64, 96, 96, 48, 64, 32)
out = inception7A(out, 64, 64, 96, 96, 48, 64, 64)
out = inception7A(out, 64, 64, 96, 96, 48, 64, 64)
out = inception7B(out, 384, 64, 96, 96)
# Stage 4
out = inception7C(out, 192, 128, 128, 192, 128, 128, 128, 128, 192, 192)
out = inception7C(out, 192, 160, 160, 192, 160, 160, 160, 160, 192, 192)
out = inception7C(out, 192, 160, 160, 192, 160, 160, 160, 160, 192, 192)
out = inception7C(out, 192, 192, 192, 192, 192, 192, 192, 192, 192, 192)
out = inception7D(out, 192, 320, 192, 192, 192, 192)
# Stage 5
out = inception7E(out, 320, 384, 384, 384, 448, 384, 384, 384, 192, avg_pool_3x3)
out = inception7E(out, 320, 384, 384, 384, 448, 384, 384, 384, 192, max_pool_3x3)
out = tf.nn.avg_pool(out, ksize=[1, 8, 8, 1], strides=[1, 1, 1, 1], padding='SAME')
dims = out.get_shape().as_list()
flatten_size = 1
for d in dims[1:]:
flatten_size *= d
with tf.variable_scope("reshape2"):
out = tf.reshape(out, [-1, int(flatten_size)])
with tf.variable_scope("fc1"):
out = fc(out, classes)
self._logits = out
if classes > 1:
self._predictions = tf.nn.softmax(self._logits)
else:
self._predictions = self._logits