

In [1]:

    
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf

Using activation function

You can use activation function which prepared by tensorflow. It is in tf.nn library!



In [2]:

    
sess = tf.Session()



In [3]:

    
# Relu
print(sess.run(tf.nn.relu([-3., 3., 10.])))









    



[  0.   3.  10.]



In [4]:

    
# Relu6
# Set upper limit as 6
print(sess.run(tf.nn.relu6([-3., 3., 10.])))









    



[ 0.  3.  6.]



In [5]:

    
# sigmoid
print(sess.run(tf.nn.sigmoid([-1., 0., 1.])))









    



[ 0.26894143  0.5         0.7310586 ]



In [6]:

    
# Hyperbolic tangent
print(sess.run(tf.nn.tanh([-1., 0., 1.])))









    



[-0.76159418  0.          0.76159418]



In [7]:

    
# Softsign function
print(sess.run(tf.nn.softsign([-1., 0., 1.])))









    



[-0.5  0.   0.5]



In [8]:

    
# Softplus function
print(sess.run(tf.nn.softplus([-1., 0., 1.])))









    



[ 0.31326166  0.69314718  1.31326163]



In [9]:

    
# ELU(Exponential Linear Unit)
print(sess.run(tf.nn.elu([-1., 0., 1.])))









    



[-0.63212055  0.          1.        ]



In [ ]: