"common/git@gitlab.lis-lab.fr:franck.dary/macaon.git" did not exist on "f9ac7a417eece1ad85f3a8543b4ddb813788ed4a"
Newer
Older
"""
Benchmark VGG: Benchmarking deepstrom versus other architectures of the VGG network.
Usage:
benchmark_vgg dense [-a value] [-v size] [-e numepoch] [-s batchsize] [-D reprdim]
Options:
--help -h Display help and exit.
-e numepoch --num-epoch=numepoch The number of epoch.
-s batchsize --batch-size=batchsize The number of example in each batch
-v size --validation-size size The size of the validation set [default: 10000]
-a value --seed value The seed value used for all randomization processed [default: 0]
-D reprdim --out-dim=reprdim The dimension of the final representation
"""
import sys
import time as t
import numpy as np
import tensorflow as tf
import docopt
import skluc.main.data.mldatasets as dataset
from skluc.main.tensorflow_.utils import fully_connected, batch_generator, classification_cifar
from skluc.main.utils import logger
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
import keras
from keras.models import Sequential
from keras.layers import Activation
from keras.layers import Conv2D, MaxPooling2D
from keras.initializers import he_normal
from keras.layers.normalization import BatchNormalization
def VGG19(input_shape):
# with tf.variable_scope("block1_conv1"):
# weights = tf.get_variable("weights", (3, 3, 3, 64), initializer=tf.random_normal_initializer(stddev=0.1), trainable=trainable)
# biases = tf.get_variable("biases", (64), initializer=tf.constant_initializer(0.0), trainable=trainable)
# regularizer = tf.contrib.layers.l2_regularizer(scale=0.1)
# conv = tf.nn.conv2d(input_, weights, strides=[1, 1, 1, 1], padding='SAME', kernel_regularizer=regularizer)
# batch_norm = tf.nn.batch_normalization(conv, variance_epsilon=1e-3)
# relu = tf.nn.relu(conv + biases)
# tf.summary.histogram("act", relu)
# in order to reduce dimensionality, use bigger pooling size
# pool = max_pool(relu, pool_size=pool_size)
# with tf.variable_scope("conv_pool_2"):
# conv2 = conv_relu_pool(conv1, [5, 5, 6, 16], [16], pool_size=2, trainable=trainable)
weight_decay = 0.0001
# build model
model = Sequential()
# Block 1
model.add(Conv2D(64, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block1_conv1', input_shape=input_shape))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(64, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block1_conv2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2), name='block1_pool'))
# Block 2
model.add(Conv2D(128, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block2_conv1'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(128, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block2_conv2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2), name='block2_pool'))
# Block 3
model.add(Conv2D(256, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block3_conv1'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(256, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block3_conv2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(256, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block3_conv3'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(256, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block3_conv4'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2), name='block3_pool'))
# Block 4
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block4_conv1'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block4_conv2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block4_conv3'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block4_conv4'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2), name='block4_pool'))
# Block 5
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block5_conv1'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block5_conv2'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block5_conv3'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(Conv2D(512, (3, 3), padding='same', kernel_regularizer=keras.regularizers.l2(weight_decay), kernel_initializer=he_normal(), name='block5_conv4'))
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool'))
return model
def fct_dense(input_, out_dim, two_layers):
with tf.variable_scope("dense_layers"):
fc_1 = fully_connected(input_, out_dim, act=tf.nn.relu, variable_scope="fc1")
if two_layers:
fc_2 = fully_connected(fc_1, out_dim, act=tf.nn.relu, variable_scope="fc2")
else:
fc_2 = fc_1
out = fc_2
return out
if __name__ == '__main__':
# todo special treat for each type of execution
with tf.Graph().as_default():
arguments = docopt.docopt(__doc__)
NUM_EPOCH = int(arguments["--num-epoch"])
BATCH_SIZE = int(arguments["--batch-size"])
SEED_TRAIN_VALIDATION = 0
SEED = int(arguments["--seed"])
OUT_DIM = int(arguments["--out-dim"]) if arguments["--out-dim"] is not None else None
VALIDATION_SIZE = int(arguments["--validation-size"])
data = dataset.Cifar10Dataset(validation_size=VALIDATION_SIZE, seed=SEED_TRAIN_VALIDATION)
data.load()
data.normalize()
data.data_astype(np.float32)
data.labels_astype(np.float32)
data.to_image()
data.to_one_hot()
logger.debug("Start benchmark with parameters: {}".format(" ".join(sys.argv[1:])))
logger.debug("Using dataset {} with validation size {} and seed for spliting set {}.".format(data.s_name, data.validation_size, data.seed))
logger.debug("Shape of train set data: {}; shape of train set labels: {}".format(data.train[0].shape, data.train[1].shape))
logger.debug("Shape of validation set data: {}; shape of validation set labels: {}".format(data.validation[0].shape, data.validation[1].shape))
logger.debug("Shape of test set data: {}; shape of test set labels: {}".format(data.test[0].shape, data.test[1].shape))
logger.debug("Sample of label: {}".format(data.train[1][0]))
# todo separated function for parameters parsing
input_dim, output_dim = data.train[0].shape[1:], data.train[1].shape[1]
x = tf.placeholder(tf.float32, shape=[None, *input_dim], name="x")
y = tf.placeholder(tf.float32, shape=[None, output_dim], name="label")
vgg_conv_model = VGG19(input_dim)
conv_x = vgg_conv_model(x)
input_classif = fct_dense(conv_x, OUT_DIM, False)
classif, keep_prob = classification_cifar(input_classif, output_dim)
# calcul de la loss
with tf.name_scope("xent"):
cross_entropy = tf.reduce_mean(
tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=classif, name="xentropy"),
name="xentropy_mean")
tf.summary.scalar('loss-xent', cross_entropy)
# todo learning rate as hyperparameter
# calcul du gradient
with tf.name_scope("train"):
global_step = tf.Variable(0, name="global_step", trainable=False)
train_optimizer = tf.train.AdamOptimizer(learning_rate=1e-4).minimize(cross_entropy,
global_step=global_step)
# calcul de l'accuracy
with tf.name_scope("accuracy"):
predictions = tf.argmax(classif, 1)
correct_prediction = tf.equal(predictions, tf.argmax(y, 1))
accuracy_op = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
tf.summary.scalar("accuracy", accuracy_op)
merged_summary = tf.summary.merge_all()
init = tf.global_variables_initializer()
# Create a session for running Ops on the Graph.
# Instantiate a SummaryWriter to output summaries and the Graph.
# summary_writer = tf.summary.FileWriter("debug_benchmark_vgg")
# Initialize all Variable objects
# actual learning
summary_writer = tf.summary.FileWriter("tb_dense_end_to_end")
with tf.Session() as sess:
summary_writer.add_graph(sess.graph)
# summary_writer.add_graph(sess.graph)
# Initialize all Variable objects
sess.run(init)
# actual learning
# feed_dict_val = {x: data.validation[0], y: data.validation[1], keep_prob: 1.0}
global_start = t.time()
feed_dict = {}
feed_dict_val = {}
feed_dict_test = {}
for i in range(NUM_EPOCH):
j = 0
start = t.time()
for X_batch, Y_batch in batch_generator(data.train[0], data.train[1], BATCH_SIZE, True):
feed_dict.update({x: X_batch, y: Y_batch, keep_prob: 0.5})
_, loss, acc = sess.run([train_optimizer, cross_entropy, accuracy_op], feed_dict=feed_dict)
if j % 100 == 0:
logger.debug("epoch: {}/{}; batch: {}/{}; loss: {}; acc: {}".format(i, NUM_EPOCH, j, int(data.train[0].shape[0]/BATCH_SIZE), loss, acc))
# summary_str = sess.run(merged_summary, feed_dict=feed_dict)
# summary_writer.add_summary(summary_str, j)
j += 1
training_time = t.time() - global_start
accuracies_val = []
i = 0
for X_batch, Y_batch in batch_generator(data.validation[0], data.validation[1], 1000, False):
feed_dict_val.update({x: X_batch, y: Y_batch, keep_prob: 1.0})
accuracy = sess.run([accuracy_op], feed_dict=feed_dict_val)
accuracies_val.append(accuracy[0])
i += 1
accuracies_test = []
i = 0
for X_batch, Y_batch in batch_generator(data.test[0], data.test[1], 1000, False):
feed_dict_test.update({x: X_batch, y: Y_batch, keep_prob: 1.0})
accuracy = sess.run([accuracy_op], feed_dict=feed_dict_test)
accuracies_test.append(accuracy[0])
i += 1
global_acc_val = sum(accuracies_val) / i
global_acc_test = sum(accuracies_test) / i
printed_r_list = [str(global_acc_val),
str(global_acc_test),
str(training_time),
str(NUM_EPOCH),
str(BATCH_SIZE),
str(OUT_DIM),
str(VALIDATION_SIZE),
str(SEED),
]
print(",".join(printed_r_list))