Hands-On Data Analysis with Pandas

Making predictions

Now that we have an idea of the effect each of our regressors has, let's use our model to make predictions of year length for the planets in the test set:

>>> preds = lm.predict(X_test)

Let's visualize how well we did by plotting the actual and predicted values:

>>> fig, axes = plt.subplots(1, 1, figsize=(5, 3))>>> axes.plot(...     X_test.semimajoraxis, y_test, 'bo', label='actuals', ...     alpha=0.5... )>>> axes.plot(...     X_test.semimajoraxis, preds, 'ro', label='predictions', ...     alpha=0.5... )>>> plt.xlabel('semimajoraxis')>>> plt.ylabel('period')>>> plt.legend()>>> plt.suptitle('Linear Regression')

The predicted values seem pretty close to the actual values and follow a similar pattern:

We can check their correlation to see ...