1. Aprende
    2. /
  2. Cursos
    3. /
  3. Ensemble Methods in Python
  • 1

    Combining Multiple Models

    Gratuito

    Do you struggle to determine which of the models you built is the best for your problem? You should give up on that, and use them all instead! In this chapter, you'll learn how to combine multiple models into one using "Voting" and "Averaging". You'll use these to predict the ratings of apps on the Google Play Store, whether or not a Pokémon is legendary, and which characters are going to die in Game of Thrones!

  • 2

    Bagging

    Bagging is the ensemble method behind powerful machine learning algorithms such as random forests. In this chapter you'll learn the theory behind this technique and build your own bagging models using scikit-learn.

  • 3

    Boosting

    Boosting is class of ensemble learning algorithms that includes award-winning models such as AdaBoost. In this chapter, you'll learn about this award-winning model, and use it to predict the revenue of award-winning movies! You'll also learn about gradient boosting algorithms such as CatBoost and XGBoost.

  • 4

    Stacking

    Get ready to see how things stack up! In this final chapter you'll learn about the stacking ensemble method. You'll learn how to implement it using scikit-learn as well as with the mlxtend library! You'll apply stacking to predict the edibility of North American mushrooms, and revisit the ratings of Google apps with this more advanced approach.

Initializing

Ejercicio

Boosting contest: Light vs Extreme

While the performance of the CatBoost model is relatively good, let's try two other flavors of boosting and see which performs better: the "Light" or the "Extreme" approach.

CatBoost is highly recommended when there are categorical features. In this case, all features are numeric, therefore one of the other approaches might produce better results.

As we are building regressors, we'll use an additional parameter, objective, which specifies the learning function to be used. To apply a squared error, we'll set objective to 'reg:squarederror' for XGBoost and 'mean_squared_error' for LightGBM.

In addition, we'll specify the parameter n_jobs for XGBoost to improve its computation runtime.

OBS: be careful not to use classifiers, or your session might expire!

Instrucciones

100 XP
  • Build an XGBRegressor using the parameters: max_depth = 3, learning_rate = 0.1, n_estimators = 100, and n_jobs=2.
  • Build an LGBMRegressor using the parameters: max_depth = 3, learning_rate = 0.1, and n_estimators = 100.