TensorFlow creates a classifier and tensorflow implements classification.

TensorFlow creates a classifier and tensorflow implements classification.

The examples in this article share the code used to create a classifier in TensorFlow for your reference. The details are as follows:

Create a classifier for the iris dataset.

Load the sample data set and implement a simple binary classifier to predict whether a flower is an iris. There are three types of iris data sets, but here we only predict whether the iris is a mountain. Import the iris dataset and tool library to convert the original dataset accordingly.

# Combining Everything Together # ---------------------------------- # This file will perform binary classification on the # iris dataset. we will only predict if a flower is # I. setosa or not. # We will create a simple binary classifier by creating a line # and running everything through a sigmoid to get a binary predictor. # The two features we will use are pedal length and pedal width. # We will use ba Tch training, but this can be easily # adapted to stochastic training. import matplotlib. pyplot as pltimport numpy as npfrom sklearn import datasetsimport tensorflow as tffrom tensorflow. python. framework import opsops. reset_default_graph () # import the iris dataset # convert the target data to 1 or 0 based on whether the target data is an iris. # Because the iris dataset marks the iris as 0, we set it from 0 to 1 and Mark other species as 0. # This training only uses two features: the petal length and the petal width. These two features are in the third and fourth columns of x-value # iris.tar get = {0, 1, 2 }, where '0' is setosa # iris. data ~ [Sepal. width, sepal. length, pedal. width, pedal. length] iris = datasets. load_iris () binary_target = np. array ([1. if x = 0 else 0. for x in iris.tar get]) iris_2d = np. array ([[x [2], x [3] for x in iris. data]) # declare the batch training size batch_size = 20 # initialize the computing image sess = tf. session () # declare the data placeholder xforwardata = tf. placeholder (shape = [None, 1], dtype = tf. float32) x2_data = tf. placeholder (shape = [None, 1], dtype = tf. float32) y_target = tf. placeh Older (shape = [None, 1], dtype = tf. float32) # declare the model variable # Create variables A and B (0 = x1-A * x2 + B) A = tf. variable (tf. random_normal (shape = [1, 1]) B = tf. variable (tf. random_normal (shape = [1, 1]) # defines a linear model: # If the data point found is above a straight line, then the data point is substituted into the x2-x1 * A-B to calculate the result is greater than 0; # Similarly, the data point found below the line, then the data point is substituted into the x2-x1 * A-B and the calculation result is less than 0. # X1-A * x2 + bmy_mult = tf. matmul (x2_data, A) my_add = tf. add (my_mult, B) my_output = tf. subtract (xforwardeddata, my_add) # added the sigmoid cross entropy loss function (cross entropy) of TensorFlow xentropy = tf. nn. sigmoid_cross_entropy_with_logits (logits = my_output, labels = y_target) # declare the optimizer method my_opt = tf. train. gradientDescentOptimizer (0.05) train_step = my_opt.minimize (xentropy) # create a variable initialization operation init = tf. global_variables_initializer () sess. run (init) #1000 runtime iterations for I in range (1000): rand_index = np. random. choice (len (iris_2d), size = batch_size) # rand_x = np. transpose ([iris_2d [rand_index]) # input three types of data: petal length, petal width, and target variable rand_x = iris_2d [rand_index] rand_x1 = np. array ([[x [0] for x in rand_x]) rand_x2 = np. array ([[x [1] for x in rand_x]) # rand_y = np. transpose ([binary_target [rand_index]) rand_y = np. array ([[y] for y in binary_target [rand_index]) sess. run (train_step, feed_dict = {x1_data: rand_x1, x2_data: rand_x2, y_target: rand_y}) if (I + 1) % 200 = 0: print ('step # '+ str (I + 1) + 'a =' + str (sess. run (A) + ', B =' + str (sess. run (B) # plot # obtain the slope/intercept # Pull out slope/intercept [[slope] = sess. run (A) [[intercept] = sess. run (B) # create a fitting line x = np. linspace (0, 3, num = 50) ablineValues = [] for I in x: ablineValues. append (slope * I + intercept) # plot the fitting curve set1__x = [a [1] for I, a in enumerate (iris_2d) if binary_target [I] = 1] setosa_y = [a [0] for I, a in enumerate (iris_2d) if binary_target [I] = 1] non_set1__x = [a [1] for I, a in enumerate (iris_2d) if binary_target [I] = 0] non_set1__y = [a [0] for I, a in enumerate (iris_2d) if binary_target [I] = 0] plt. plot (set0000_x, set0000_y, 'rx ', MS = 10, mew = 2, label = 'setosa') plt. plot (non_set1__x, non_set1__y, 'ro', label = 'non-setosa ') plt. plot (x, ablineValues, 'B-') plt. xlim ([0.0, 2.7]) plt. ylim ([0.0, 7.1]) plt. suptitle ('linear Separator For I. setosa ', fontsize = 20) plt. xlabel ('petal length') plt. ylabel ('petal width') plt. legend (loc = 'lower right') plt. show ()

Output:

Step #200 A = [[ 8.70572948]], b = [[-3.46638322]]Step #400 A = [[ 10.21302414]], b = [[-4.720438]]Step #600 A = [[ 11.11844635]], b = [[-5.53361702]]Step #800 A = [[ 11.86427212]], b = [[-6.0110755]]Step #1000 A = [[ 12.49524498]], b = [[-6.29990339]]

The above is all the content of this article. I hope it will be helpful for your learning and support for helping customers.