3.14 Takeaways

This chapter explains linear regressions in detail. Here is a list of key points:

• A linear regression is a supervised Machine Learning algorithm used to solve regression problems.

• A linear regression is built from a dataset available in BigML and used to make an evaluation, a single prediction, or a batch prediction.

• You can create a linear regression with just one click or configure it as you wish. BigML provides several configuration options before creating your linear regression.

• To create a linear regression you need a dataset containing at least one numeric field.

• Categorical fields must be converted to numeric values in order to train a linear regression model.

• If you do not specify any Objective Field, BigML will take the last numeric field in your dataset.

• BigML allows you to include your numeric fields’ missing values as valid values to train your linear regression model.

• The chart view provides a visual way to analyze a field impact on predictions given certain values for the rest of the fields.

• You get a prediction interval along with the predicted value.

• BigML displays all your linear regression coefficients in a table view which you can also download as a CSV file.

• You need to evaluate your linear regression model’s performance using data that the model has not seen before.

• The ultimate goal in building a linear regression is being able to make predictions with it.

• BigML allows you to quickly make predictions for single instances by providing a form containing the fields used by the linear regression, so you can easily set the input data and get an immediate response.

• BigML batch predictions allow you to make simultaneous predictions for multiple instances. All you need is the linear regression you want to use to make predictions and a dataset containing the instances for which you want to obtain predictions.

• You can configure your batch predictions output file settings.

• You can download your linear regression to perform local predictions.

• You can add descriptive information to your linear regressions (name, description, tags, and category).

• You can move your linear regressions between projects.

• You can share your linear regressions with other people using the secret link.

• You can stop your linear regression creation by deleting them.

• You can permanently delete an existing linear regression.