In the previous article (broken link) I explained the installation and showed a "hello world" from TensorFlow.

This article explains the use of constants and variables, first with a 1-dimensional array of 4 elements, how a numpy array is converted to a TensorFlow object and how to show its elements; then it shows how to handle the equation y = x+1 with x = 9.

Handling a one-dimensional array with tensorflow In [1]:

#Se importa numpy y tensorflow
import numpy as np
import tensorflow as tf

In [2]:

#Se crea un arreglo de una dimensión
tensor_1d = np.array([5.7, 2, 8.0, 25.99])

In [3]:

#Se imprime el arreglo
print (tensor_1d)
[  5.7    2.     8.    25.99]

In [4]:

#Se imprime el último elemento del arreglo -> 25.99
print (tensor_1d[-1])
25.99

In [5]:

#Dimensiones del arreglo -> 1 dimensión
tensor_1d.ndim

Out[5]:

1

In [6]:

#Cantidad de elementos del arreglo -> 4 elementos
tensor_1d.shape

Out[6]:

(4,)

In [7]:

#Tipo de los elementos que contiene el arreglo ->float64
tensor_1d.dtype

Out[7]:

dtype('float64')

In [8]:

#Convierte el arreglo a un objeto tensorflow
tf_tensor=tf.convert_to_tensor(tensor_1d,dtype=tf.float64)

In [9]:

# tipo tensor, constante, 4 elementos y de tipo float64
tf_tensor

Out[9]:

<tf.Tensor 'Const:0' shape=(4,) dtype=float64>

In [10]:

# Se inicia sesión tensorflow y se corre la la impresión de los elementos
with tf.Session() as sess:
    print (sess.run(tf_tensor))
    print (sess.run(tf_tensor[-1]))
[  5.7    2.     8.    25.99]
25.99

Now an equation will be calculated with some variables:

x = 1  
y = x + 9  
print(y)  
import tensorflow as tf  
x = tf.constant(1,name='x')  
y = tf.Variable(x+9,name='y')  
print(y)

In [13]:

#Se define el valor de la constante x = 1
x = tf.constant(1,name='x')

In [14]:

#Se define el valor de la variable y = x+9
y = tf.Variable(x+9,name='y')

In [16]:

#Se imprime el valor de y, que devuelve el tipo de objeto tensor flow, la variable y, no tiene elementos
#y es de tipo int32.
print(y)
<tf.Variable 'y:0' shape=() dtype=int32_ref>

For it to work you have to create the session and execute run

In [20]:

# Inicializa todas las variables
model = tf.global_variables_initializer()

In [22]:

#Se inicia la sesión, se corre model y luego y. El resultado es y -> 10
with tf.Session() as session:
    session.run(model)
    print(session.run(y))
10

In [ ]:

undefined

Make a donation! If you liked the article you can make a donation with Bitcoin (BTC) using the digital wallet of your choice to the following address: 17MtNybhdkA9GV3UNS6BTwPcuhjXoPrSzV

Or by scanning the QR code from your wallet:

17MtNybhdkA9GV3UNS6BTwPcuhjXoPrSzV