Create and manage Experiments¶
Experiments are the individual "projects" within a Use Case. They allow you to vary data, targets, and modeling settings to find the optimal models to solve your business problem. Within each experiment, you have access to its Leaderboard and model insights, as well as experiment summary information. After selecting a model, you can, from within the experiment:
See the associated FAQ for important additional information.
Follow the steps below to create a new experiment from within a Use Case.
You can also start modeling directly from a dataset by clicking the Start modeling button. The Set up new experiment page opens. From there, the instructions follow the flow described below.
From within a Use Case, click Add new and select Add experiment. The Set up new experiment page opens, which lists all data previously loaded to the Use Case.
Add data to the experiment, either by adding new data (1) or selecting a dataset that has already been loaded to the Use Case (2).
Once the data is loaded to the Use Case (as in option 2 above), click to select the dataset you want to use in the experiment. Workbench opens a preview of the data:
|Click to return to the data listing and choose a different dataset.|
|Click on the icon to proceed and set the target.|
|Click Next to proceed and set the target.|
Once you have proceeded, Workbench prepares the dataset for modeling (EDA 1). When the process finishes, set the target either by hovering over the feature name and clicking or entering the name in the entry box:
Scroll through the list of features to find your target. If it is not showing, expand the display:
Once located, click the entry in the table to use the feature as the target.
Type the name of the target feature you would like to predict in the entry box. 特徴量名の文字を入力するに従って、一致する特徴量がリスト表示されます。
Once a target is entered, Workbench displays a histogram providing information about the target feature's distribution and, in the right pane, a summary of the experiment parameters.
From here, you are ready to build models with the default settings. Or, you can modify the default settings and then begin. If using the default settings, click Start modeling to begin the Quick mode Autopilot modeling process.
Changing experiment parameters is a good way to iterate on a Use Case. Before starting to model, you can:
Once you have reset any or all of the above, click Start modeling to begin the Quick mode modeling process.
特徴量セットは、DataRobotでモデルの構築に使用する特徴量のサブセットを制御します。 Workbench defaults to the Informative Features list, but you can modify that prior to model building. To change, click on the Feature list dropdown and select a different list:
You can also modify the list on a per-model basis once the experiment finishes building.
パーティショニングは、評価とモデル構築のためにDataRobotが観測値（または行）をまとめて「集中させる」方法を示します。 Workbench defaults to five-fold, stratified partitioning with a 20% holdout fold. You can change both the partitioning method and the validation type:
Click the icon for Additional settings, Next, or the Partitioning field in the summary:
Set the fields that you want to change:
フィールド 説明 パーティション Sets the partitioning method:
- Random. Randomly assigns observations (rows) to the training, validation, and holdout sets.
- Stratified. Randomly assigns rows to training, validation, and holdout sets, preserving (as close as possible to) the same ratio of values for the prediction target as in the original data.
検定タイプ Sets the method used on data for to validate models.
- Cross-validation. Separates the data into two or more “folds” and creates one model per fold, with the data assigned to that fold used for validation and the rest of the data used for training.
- 教師-検定-ホールドアウト. For larger datasets, partitions data into three distinct sections—training, validation, and holdout— with predictions based on a single pass over the data.
交差検定の分割 Sets the number of folds used with the cross-validation method to maximize the data available for each partition. Holdout percentage Sets the percentage of data that Workbench “hides” when training. The Leaderboard shows a Holdout value, which is calculated using the trained model's predictions on the holdout partition. When models are built, Workbench unlocks holdout so that you can compare validation and Holdout scores.
To change the target after selection, click the target icon or the Back button: