Scoring and prediction methods

DataRobot provides multiple methods to score datasets using models that have been created. One of the easiest methods is batch scoring via the DataRobot user interface (UI). For this, we need to follow these steps:

1. Create a file with the dataset to be scored. Given that we are using a public dataset, we will simply use the same dataset to score. In a real project, you will have access to a new dataset for which you want to create predictions. For our purposes, we simply created a copy of our imports-85-data.xlsx dataset file and named it imports-85-data-score.xlsx.
2. Now, let's select the Predict tab and then the Test Predictions tab for the XGBoost (XGB) models, as shown in the following screenshot:

   Figure 8.1 – Batch scoring

   In the preceding screenshot, you will see that you have an option to drag and drop a new dataset to add the scoring file to the model.

3. Let's select our imports-85-data-score.xlsx scoring file and...