utils.py

import numpy as np
from matplotlib import pyplot

# Some functions to plot our points and draw the lines
def plot_points(features, labels, fix_margins=True):
    X = np.array(features)
    y = np.array(labels)
    spam = X[np.argwhere(y==1)]
    ham = X[np.argwhere(y==0)]
    if fix_margins:
        pyplot.xlim(0, 11)
        pyplot.ylim(0, 11)
    pyplot.scatter([s[0][0] for s in spam],
                [s[0][1] for s in spam],
                s = 100,
                color = 'cyan',
                edgecolor = 'k',
                marker = '^')
    pyplot.scatter([s[0][0] for s in ham],
                [s[0][1] for s in ham],
                s = 100,
                color = 'red',
                edgecolor = 'k',
                marker = 's')
    pyplot.xlabel('Lottery')
    pyplot.ylabel('Sale')
    pyplot.legend(['Spam','Ham'])

def plot_model(X, y, model, fix_margins=True):
    X = np.array(X)
    y = np.array(y)
    plot_points(X, y)
    plot_step = 0.01
    x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
    if fix_margins:
        x_min=0
        y_min=0
        x_max=12
        y_max=12
    xx, yy = np.meshgrid(np.arange(x_min, x_max, plot_step),
                         np.arange(y_min, y_max, plot_step))
    Z = model.predict(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)
    pyplot.contourf(xx, yy, Z, colors=['red', 'blue'], alpha=0.2, levels=range(-1,2))
    pyplot.contour(xx, yy, Z,colors = 'k',linewidths = 3)
    pyplot.show()

def display_tree(dt):
    from sklearn.externals.six import StringIO  
    from IPython.display import Image  
    from sklearn.tree import export_graphviz
    import pydotplus
    dot_data = StringIO()
    export_graphviz(dt, out_file=dot_data,  
                    filled=True, rounded=True,
                    special_characters=True)
    graph = pydotplus.graph_from_dot_data(dot_data.getvalue())  
    return Image(graph.create_png())
    
def plot_trees(model):
    estimators = gradient_boosting_model.estimators_
    for i in range(len(estimators)):
        tree.plot_tree(estimators[i][0])
        pyplot.show()
        #plot_model(new_X, new_y, estimators[i][0])
        
def plot_regressor(model, features, labels):
    x = np.linspace(0,85,1000)
    pyplot.scatter(features, labels)
    pyplot.plot(x, model.predict(x.reshape([-1,1])))
    pyplot.xlabel("Age")
    pyplot.ylabel("Days per week")
    pyplot.show()