Import TensorFlow for the learning algorithm. We'll need NumPy to set up the initial data. And we'll use matplotlib to visualize our data.
In [1]:
%matplotlib inline
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
Define some constants used by the learning algorithm. There are called hyper-parameters.
In [2]:
learning_rate = 0.01
training_epochs = 100
Set up fake data that we will use to to find a best fit line
In [3]:
x_train = np.linspace(-1, 1, 101)
y_train = 2 * x_train + np.random.randn(*x_train.shape) * 0.33
Plot the raw data
In [4]:
plt.scatter(x_train, y_train)
Out[4]:
Set up the input and output nodes as placeholders since the value will be injected by x_train and y_train.
In [5]:
X = tf.placeholder("float")
Y = tf.placeholder("float")
Define the model as y = w'*x
In [6]:
def model(X, w):
return tf.multiply(X, w)
Set up the weights variable
In [7]:
w = tf.Variable(0.0, name="weights")
Define the cost function as the mean squared error
In [8]:
y_model = model(X, w)
cost = tf.reduce_mean(tf.square(Y-y_model))
Define the operation that will be called on each iteration of the learning algorithm
In [9]:
train_op = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)
Initialize all variables
In [10]:
sess = tf.Session()
init = tf.global_variables_initializer()
sess.run(init)
Train on each (x, y) pair multiple times
In [11]:
for epoch in range(training_epochs):
for (x, y) in zip(x_train, y_train):
sess.run(train_op, feed_dict={X: x, Y: y})
Fetch the value of the learned parameter
In [12]:
w_val = sess.run(w)
In [13]:
sess.close()
Visualize the best fit curve
In [14]:
plt.scatter(x_train, y_train)
y_learned = x_train*w_val
plt.plot(x_train, y_learned, 'r')
plt.show()