def inference(x, y_, keep_prob, phase_train):
with tf.variable_scope('conv_1'):
conv1 = Convolution2D(x, (28, 28), 1, 32, (5, 5), activation='none')
conv1_bn = batch_norm(conv1.output(), 32, phase_train)
conv1_out = tf.nn.relu(conv1_bn)
pool1 = MaxPooling2D(conv1_out)
pool1_out = pool1.output()
with tf.variable_scope('conv_2'):
conv2 = Convolution2D(pool1_out, (14, 14),
32,
64, (5, 5),
activation='none')
conv2_bn = batch_norm(conv2.output(), 64, phase_train)
conv2_out = tf.nn.relu(conv2_bn)
pool2 = MaxPooling2D(conv2_out)
pool2_out = pool2.output()
pool2_flat = tf.reshape(pool2_out, [-1, 7 * 7 * 64])
with tf.variable_scope('fc1'):
fc1 = FullConnected(pool2_flat, 7 * 7 * 64, 1024)
fc1_out = fc1.output()
fc1_dropped = tf.nn.dropout(fc1_out, keep_prob)
y_pred = ReadOutLayer(fc1_dropped, 1024, 10).output()
loss = tf.reduce_mean(
-tf.reduce_sum(y_ * tf.log(y_pred), reduction_indices=[1]))
accuracy = evaluation(y_pred, y_)
return loss, accuracy, y_pred
def create_decoder_fc(self, conf, final_shape):
for i in range(len(conf)):
if i > 1:
drop = tf.nn.dropout(self.layers[-1], self.keep_prob)
self.layers.append(drop)
fc = FullConnected(self.layers[-1],
self.layers[-1].get_shape().as_list()[1],
conf[i],
activation='sigmoid')
self.layers.append(fc.output())
self.layers.append(tf.reshape(self.layers[-1], final_shape))
def create_encoder_fc(self, conf):
size = self.layers[-1].get_shape().as_list()
self.layers.append(
tf.reshape(self.layers[-1], [-1, size[1] * size[2] * size[3]]))
for i in range(len(conf)):
if i > 1:
drop = tf.nn.dropout(self.layers[-1], self.keep_prob)
self.layers.append(drop)
fc = FullConnected(self.layers[-1],
self.layers[-1].get_shape().as_list()[1],
conf[i],
activation='sigmoid')
self.layers.append(fc.output())
def mk_nn_model(x, y_):
# Encoding phase
x_image = tf.reshape(x, [-1, 28, 28, 1])
conv1 = Convolution2D(x_image, (28, 28), 1, 8,
(9, 9), activation='sigmoid')
conv1_out = conv1.output()
# pool1 = MaxPooling2D(conv1_out)
# pool1_out = pool1.output()
# pool1_out = tf.nn.dropout(pool1_out,keep_prob=0.2)
conv2 = Convolution2D(conv1_out, (28, 28), 8, 4,
(9, 9), activation='sigmoid')
conv2_out = conv2.output()
# pool2 = MaxPooling2D(conv2_out)
# pool2_out = pool2.output()
# pool2_out = tf.nn.dropout(pool2_out,keep_prob=0.2)
# at this point the representation is (4, 28, 28) i.e. 128*16-dimensional
po = tf.reshape(conv2_out,[-1,4*28*28])
fc = FullConnected(po, 4*28*28, 256, activation='sigmoid')
fc_out = fc.output()
# fc2 = FullConnected(fc_out, 256, 2, activation='sigmoid')
# fc2_out = fc2.output()
fo = FullConnected(fc_out, 256, 10, activation='sigmoid')
fo_out = fo.output()
# Decoding phase
dfc1 = FullConnected(fo_out, 10, 256, activation='sigmoid')
dfc1_out = dfc1.output()
# reshape
deconvin = tf.reshape(dfc1_out, [-1,16,16,1])
#resize_images(images, size, method=ResizeMethod.BILINEAR, align_corners=False)
deconvin = tf.image.resize_images(deconvin, (28,28),method=tf.image.ResizeMethod.BILINEAR, align_corners=False)
conv_t1 = Conv2Dtranspose(deconvin, (28, 28), 1, 4,
(12, 12), activation='sigmoid')
conv_t1_out = conv_t1.output()
conv_t2 = Conv2Dtranspose(conv_t1_out, (28, 28), 4, 4,
(17, 17), activation='sigmoid')
conv_t2_out = conv_t2.output()
conv_t3 = Conv2Dtranspose(conv_t2_out, (28, 28), 4, 1,
(1, 1), activation='sigmoid')
decoded = conv_t3.output()
decoded = tf.reshape(decoded, [-1, 784])
cross_entropy = -1. *x *tf.log(decoded) - (1. - x) *tf.log(1. - decoded)
loss = tf.reduce_mean(cross_entropy)
# crossentry for classifier
cross_entropy_acc = tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=fo_out)
lossacc = tf.reduce_mean(cross_entropy_acc)
# accuracy of the trained model, between 0 (worst) and 1 (best)
correct_prediction = tf.equal(tf.argmax(fo_out, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
return loss, decoded, lossacc, fo_out, accuracy
def mk_nn_model(x, y_):
# Encoding phase
x_image = tf.reshape(x, [-1, 227, 227, 3])
#1st conv.
conv1 = Convolution2D(x_image, (227, 227),
3,
96, (7, 7),
activation='relu',
S=4)
conv1_out = conv1.output()
#1st pooling
pool1 = MaxPooling2D(conv1_out, ksize=[1, 3, 3, 1], S=2)
pool1_out = pool1.output()
#dropout?
# pool1_out = tf.nn.dropout(pool1_out,keep_prob=0.2)
#LRN1
norm1 = tf.nn.local_response_normalization(pool1_out,
depth_radius=5,
alpha=0.0001,
beta=0.75)
#2nd conv.
conv2 = Convolution2D(norm1, (29, 29),
96,
256, (5, 5),
activation='relu',
S=1) # pad=2 - how to do it????
conv2_out = conv2.output()
#2nd pooling
pool2 = MaxPooling2D(conv2_out, ksize=(1, 3, 3, 1), S=2)
pool2_out = pool2.output()
# pool2_out = tf.nn.dropout(pool2_out,keep_prob=0.2)
norm2 = tf.nn.local_response_normalization(pool2_out,
depth_radius=5,
alpha=0.0001,
beta=0.75)
#3rd conv.
conv3 = Convolution2D(norm2, (15, 15),
256,
384, (3, 3),
activation='relu',
S=1) # pad=2 - how to do it????
conv3_out = conv3.output()
#3rd pooling
pool3 = MaxPooling2D(conv3_out, ksize=(1, 3, 3, 1), S=2)
pool3_out = pool3.output()
# at this point the representation is (4, 28, 28) i.e. 128*16-dimensional
po = tf.reshape(pool3_out, [-1, 384 * 8 * 8])
fc6 = FullConnected(po, 384 * 8 * 8, 512, activation='relu')
fc6_out = fc6.output()
drop6 = tf.nn.dropout(fc6_out, keep_prob=0.5)
fc7 = FullConnected(drop6, 512, 512, activation='relu')
fc7_out = fc7.output()
drop7 = tf.nn.dropout(fc7_out, keep_prob=0.5)
fc8 = FullConnected(drop7, 512, 8, activation='relu')
fc8_out = fc8.output()
# crossentry for classifier
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(labels=y_,
logits=fc8_out)
loss = tf.reduce_mean(cross_entropy)
# accuracy of the trained model, between 0 (worst) and 1 (best)
correct_prediction = tf.equal(tf.argmax(fc8_out, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
return loss, accuracy, fc8_out