Classification and Regression with the BigML Dashboard

2.3 Creating Ensembles with 1-Click

To create an ensemble in BigML you have two options: either the BigML unique 1-click feature which provides a convenient way to quickly train an ensemble from a dataset, or configuring a number of options that give you fine control on how the ensemble is created. This section will guide you through the process of creating an ensemble with just 1-click.

By far, the easiest and quickest way to create an ensemble is using the 1-click model option that is available in the 1-click action menu. (See Figure 2.7 .)

\includegraphics[]{images/ensembles/one-click-ensemble}
Figure 2.7 1-click ensemble from dataset detail view

Alternatively, you can select the 1-click model option in the 1-click action menu from the dataset list view. (See Figure 2.8 .)

\includegraphics[]{images/ensembles/one-click-ensemble-list}
Figure 2.8 1-click ensemble from dataset list view

Both options build a new ensemble using default values for all available configuration options (see section 2.4 ). This can give you a quick starting point to understand how your ensembles behaves and how it can be improved.

Creating an ensemble may take a variable time, depending on how big your dataset is, your subscription plan, etc. Once your ensemble is ready, it will be automatically displayed on your BigML Dashboard and you will be able to explore it through BigML visualizations (see section 2.5 ), and using it for evaluations (see Chapter 7 ) or predictions (see section 2.6 .)