averaged_perceptron_example.py

from matplotlib import pyplot as plt
from matplotlib import animation
from shogun import csv_file, features, labels, machine, parameter_observer

f_feats_train = csv_file("classifier_binary_2d_linear_features_train.dat")
f_feats_test = csv_file("classifier_binary_2d_linear_features_test.dat")
f_labels_train = csv_file("classifier_binary_2d_linear_labels_train.dat")
f_labels_test = csv_file("classifier_binary_2d_linear_labels_test.dat")

features_train = features(f_feats_train)
features_test = features(f_feats_test)
labels_train = labels(f_labels_train)
labels_test = labels(f_labels_test)

perceptron = machine("AveragedPerceptron", labels=labels_train, learn_rate=1.0, max_iterations=1000)

observer = parameter_observer("ParameterObserverLogger")
perceptron.subscribe(observer)

perceptron.train(features_train)
labels_predict = perceptron.apply(features_test)

fig = plt.figure()
ax = plt.axes()
line, = ax.plot([], [])

# Plot the background for each frame
def init():

    # Plot all the data points every time
    for i in range(f_size):
        plt.plot(features_train[i][0], features_train[i][1], 'ro')

    line.set_data([], [])
    return line,

# Print the observation line on screen
def animate(i):
    obs = observer.get_observation(i)
    vect = obs.get("weights")
    x = np.linspace(-10, 10, 1000)
    line.set_data(x, vect*x)
    return line,

anim = animation.FuncAnimation(fig, animate, init_func=init, frames=f_size, interval=10, blit=True)
anim.save('basic_animation.mp4', fps=30, extra_args=['-vcodec', 'libx264'])