绘制具有3个特征的Binary Logistic回归

Question

I have built a logistic regression model for binary classification. I am trying to develop a visualisation but unable to. I can do it with two features but my model uses 3 features and hence struggling to convert it into a 3d graph. Below is the code i have tried but doesnt work. I would be grateful if someone could guide me in this please. I want to be able to visualise my results. i am still new to visualising graph.

from matplotlib.colors import ListedColormap
X_set, y_set = X_train, y_train
X1, X2, X3 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:, 0].max() + 1, step = 0.01),
                     np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:, 1].max() + 1, step = 0.01),
                     np.arange(start = X_set[:, 2].min() - 1, stop = X_set[:, 2].max() + 1, step = 0.01))
plt.contourf(X1, X2, X3, classifier.predict(np.array([X1.ravel(), X2.ravel(), X3.ravel()]).T).reshape(X1.shape),
             alpha = 0.75, cmap = ListedColormap(('red', 'green')))
plt.xlim(X1.min(), X1.max())
plt.ylim(X2.min(), X2.max())
plt.ylim(X3.min(), X3.max())
for i, j in enumerate(np.unique(y_set)):
    plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
                c = ListedColormap(('red', 'green'))(i), label = j)
plt.title('Logistic Regression (Training set)')
plt.xlabel('Recency')
plt.ylabel('Frequency')
plt.zlabel('Monetary')
plt.legend()
plt.show()

Answer 1

plt.contourf(x, y, z) would work fine for visualising two features (x, y) and a prediction function, which depends on these two features. The predictions z must be a 2D array. With three features you could fix one of them and then make a 2D contour plots using plt.contourf(). To get a more complete picture, repeat this with the feature fixed to another value and/or fixing one of the other features.

Or plot your training set in 3D with a scatter plot: https://matplotlib.org/2.1.1/gallery/mplot3d/scatter3d.html

True and predicted labels could be visualised using different markers for the four possible outcomes: (true=1, pred=1), (true=1, pred=0), (true=0, pred=1) and (true=0, pred=0).