Basic Neural Network
in Tensorflow

Code courtesy of MorvanZhou

1. https://github.com/MorvanZhou/tutorials/tree/master/tensorflowTUT/tf12_plot_result
2. Youtube video tutorial: https://www.youtube.com/channel/UCdyjiB5H8Pu7aDTNVXTTpcg
3. Youku video tutorial: http://i.youku.com/pythontutorial

Load packages

# These are css/html style for good looking ipython notebooks
from IPython.core.display import HTML
css = open('c:/ml/style-notebook.css').read()
HTML('<style>{}</style>'.format(css))

from __future__ import print_function
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline

print ("PACKAGES LOADED")

PACKAGES LOADED

def add_layer(inputs, in_size, out_size, activation_function=None):
    Weights = tf.Variable(tf.random_normal([in_size, out_size]))
    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)
    Wx_plus_b = tf.matmul(inputs, Weights) + biases
    if activation_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activation_function(Wx_plus_b)
    return outputs

Create some simple data

x_data = np.linspace(-1, 1, 300)
x_data = x_data.reshape(x_data.shape + (1,))
noise = np.random.normal(0, 0.05, x_data.shape)
y_data = np.square(x_data) - 0.5 + noise

plt.scatter(x_data, y_data)

<matplotlib.collections.PathCollection at 0x25348d304e0>

Define placeholders for network inputs

xs = tf.placeholder(tf.float32, [None, 1])
ys = tf.placeholder(tf.float32, [None, 1])

Add a hidden layer

Input size is 1, output size is 10. xs is inputs vector.

lay_1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)

Add an output layer

prediction = add_layer(lay_1, 10, 1, activation_function=None)

Error between prediciton and real data

loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction), reduction_indices=[1]))
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

Session run

sess = tf.Session()
init = tf.global_variables_initializer()
sess.run(init)

Online Plotting

import time

def draw_lines():
    fig = plt.figure()
    ax = fig.add_subplot(1,1,1)
    ax.scatter(x_data, y_data)
    return ax

for i in range(1000):
    # training
    sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
    if i % 100 == 0:
        # to visualize the result and improvement
        print("train_step:", i)
        ax = draw_lines()
        try:
            ax.lines.remove(lines[0])
        except Exception:
            pass
        prediction_value = sess.run(prediction, feed_dict={xs: x_data})
        # plot the prediction
        lines = ax.plot(x_data, prediction_value, 'r-', lw=5)
        plt.show(block=False)
        time.sleep(1)

train_step: 0

train_step: 400

train_step: 800

train_step: 1200

train_step: 1600