LaTeX : https://drive.google.com/file/d/1Ncid4LFcY5dfM-lLtrK3_CEiwcfCt6J_/view?usp=sharing

https://colab.research.google.com/drive/1v7S2z85R42Xa5_XdwzL2lFTZ3eyuyUEq?usp=sharing

L1 Stats : https://drive.google.com/open?id=1gLsanCnO5CE7D5t1rF9yVtojX6oit1wz

php :

//Our SQL statement, which will select a list of tables from the current MySQL database.

$sql = "SHOW TABLES";

//Prepare our SQL statement,

$statement = $pdo->prepare($sql);

//Execute the statement.

$statement->execute();

//Fetch the rows from our statement.

$tables = $statement->fetchAll(PDO::FETCH_NUM);

//Loop through our table names.

foreach($tables as $table){

//Print the table name out onto the page.

echo $table[0], '<br>';

}

master m2 :

!wget --output-document=vgg19.npy https://media.githubusercontent.com/media/tensorlayer/pretrained-models/master/models/vgg19.npy

class Vgg19:

def __init__(self, vgg19_npy_path="./vgg19.npy", reuse=False, name="vgg19"):

self.data_dict = np.load(vgg19_npy_path, encoding='latin1', allow_pickle=True).item()

self.reuse = reuse

self.name = name

self.couches ={}

def build(self, couche_in, withFC):

self.couches["IN"] = couche_in

couche_in = couche_in - tf.constant(np.array([103.939, 116.779, 123.68]).reshape((1, 1, 1, 3)), dtype='float32')

with tf.variable_scope(self.name, reuse=self.reuse):

struct = ["conv1_1", "conv1_2", "pool1","conv2_1", "conv2_2", 'pool2',

"conv3_1","conv3_2","conv3_3","conv3_4","pool3","conv4_1",

"conv4_2","conv4_3","conv4_4","pool4","conv5_1","conv5_2",

"conv5_3","conv5_4", "pool5", "fc6", "fc7"]

tmp = couche_in

for line in struct:

if line[:4]=="conv":

tmp = self.conv_layer(tmp, line)

elif line[:4]=="pool":

tmp = self.max_pool(tmp, line)

elif withFC:

tmp = self.fc_layer(tmp, line, relu=True)

self.data_dict = None

def get(self, name):

return self.couches[name]

def max_pool(self, bottom, name):

return tf.nn.max_pool(bottom, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', name=name)

def conv_layer(self, bottom, name):

with tf.variable_scope(name):

filt = tf.constant(self.data_dict[name][0], name="filter")

conv = tf.nn.conv2d(bottom, filt, [1, 1, 1, 1], padding='SAME')

conv_biases = tf.constant(self.data_dict[name][1], name="biases")

bias = tf.nn.bias_add(conv, conv_biases)

relu = tf.nn.relu(bias)

self.couches[name] = relu

return relu

def fc_layer(self, bottom, name, relu):

with tf.variable_scope(name):

x = tf.contrib.layers.flatten(bottom)

weights = tf.constant(self.data_dict[name][0], name="weights")

biases = tf.constant(self.data_dict[name][1], name="biases")

fc = tf.nn.bias_add(tf.matmul(x, weights), biases)

if relu:

fc = tf.nn.relu(fc)

self.couches[name] = fc

return fc

AES L3

Databases :

https://www.data.gouv.fr/fr/

http://data.montpellier3m.fr/

CR : https://drive.google.com/open?id=1nTE5U1gdQeYw7zjPTDsasM3OSF2AaeZP