---
title: Create experiments
description: Describes how to create and manage experiments in the DataRobot Workbench interface.

---

# Create experiments {: #create-experiments }

There are two AI experimentation "types" available in Workbench:

* _Predictive modeling_, described on this page, makes row-by-row predictions based on your data.


* _Time-aware modeling_, described [here](ts-experiment-create){ target=_blank }, models using _time-relevant data_ to make row-by-row predictions, time series forecasts, or current value predictions ["nowcasts"](nowcasting).


Experiments are the individual "projects" within a [Use Case](wb-build-usecase). 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](ml-experiment-evaluate#insights), as well as [experiment summary information](ml-experiment-evaluate#view-experiment-info).

See the associated [FAQ](wb-experiment-ref) for important additional information.

{% include 'includes/wb-experiment-create-basic.md' %}


{% include 'includes/wb-add-experiment-add-data.md' %}

{% include 'includes/wb-select-target.md' %}

After the 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 settings.

![](images/wb-exp-10.png)

Click **Next** to view **Additional settings**. Either build models with the default settings or [modify the default settings](#customize-settings) and then begin. If using the default settings, click **Start modeling** to begin the [Quick mode](model-data#modeling-modes-explained){ target=_blank } Autopilot modeling process.  

## Customize basic settings {: #customize-basic-settings }

{% include 'includes/wb-customize-basic-experiment-1.md' %}

After changing any or all of the settings described, click **Next** and:

*  Click **Start modeling** to begin the [Quick mode](model-data#modeling-modes-explained){ target=_blank } predictive modeling process. 
* Customize more [advanced settings](#configure-additional-settings).

### Change modeling mode {: #change-modeling-mode }

By default, DataRobot builds experiments using Quick Autopilot. However, you can change the modeling mode to train specific blueprints or all applicable repository blueprints.

![](images/wb-exp-32.png)

The following table describes each of the modeling modes:

|  Modeling mode | Description |
|-------------|-------------|
|  Quick (default)  | Using a sample size of 64%, Quick Autopilot runs a [subset](model-ref#specifics-of-quick-autopilot) of models, based on the specified target feature and performance metric, to provide a base set of models that build and provide insights quickly.|
|  Manual | Manual mode gives you full control over which blueprints to execute. After [EDA2](eda-explained) completes, DataRobot redirects you to the [blueprint repository](ml-experiment-evaluate.html#blueprints-repository-tab) where you can select one or more blueprints for training. |
|  Comprehensive    | [Comprehensive Autopilot mode](more-accuracy) runs all Repository blueprints on the maximum Autopilot sample size to ensure more accuracy for models. This mode results in extended build times.

{% include 'includes/wb-customize-basic-experiment-2.md' %}

You can also [change the selected list](ml-experiment-add#train-on-new-settings) on a per-model basis once the experiment finishes building.

## Customize advanced settings {: #customize-advanced-settings }

To apply more advanced modeling criteria before training, you can optionally:

* [Modify partitioning](#data-partitioning-tab).
* [Configure additional settings](#configure-additional-settings).
* [Change configuration settings](#change-the-configuration).

### Data partitioning tab {: #data-partitioning-tab }

Partitioning describes the method DataRobot uses to “clump” observations (or rows) together for evaluation and model building. Workbench defaults to [five-fold](data-partitioning){ target=_blank } cross-validation with [stratified sampling](partitioning#ratio-preserved-partitioning-stratified){ target=_blank } (for binary classification experiments) or random (for regression experiments) and a 20% holdout fold.

![](images/wb-exp-13.png)

!!! note
	If there is a date feature available, your experiment is eligible for **Date/time** partitioning, which assigns rows to backtests chronologically instead of, for example, randomly. This is the only valid partitioning method for time-aware projects. See the [time-aware modeling](ts-experiment-create#data-partitioning-tab){ target=_blank } documentation for more information.

Change the partitioning method or validation type from **Additional settings** or by clicking the **Partitioning** field in the summary:

![](images/wb-exp-12.png)

#### Set the partitioning method {: #set-the-partitioning-method }

The partitioning method instructs DataRobot in how to assign rows when training models. Note that the choice of partitioning method and validation type is dependent on the target feature and/or partition column. In other words, not all selections will always display as available. The following table briefly describes each method; see also [this section](partitioning#partitioning-details){ target=_blank } for more partitioning details.

![](images/wb-exp-13-part-ml.png)

Method	| Description
--------|-------------
Stratified | Rows are randomly assigned 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. This is the default method for binary classification problems. 
Random | DataRobot randomly assigns rows to the training, validation, and holdout sets. This is the default method for regression problems. 
User-defined grouping | Creates a 1:1 mapping between values of this feature and validation partitions. Each unique value receives its own partition, and all rows with that value are placed in that partition. This method is recommended for partition features with low cardinality. See [partition by grouping](#partition-by-grouping), below.
Automated grouping | All rows with the same single value for the selected feature are guaranteed to be in the same training or test set. Each partition can contain more than one value for the feature, but each individual value will be automatically grouped together. This method is recommended for partition features with high cardinality. See [partition by grouping](#partition-by-grouping), below.
Date/time | See [time-aware experiments](ts-experiment-create#enable-time-aware-modeling).

#### Set the validation type {: #set-the-validation-type }

Validation type sets the method used on data to validate models. Choose a method and set the associated fields. A graphic below the configuration fields illustrates the settings. See the description of validation type when using [user-defined or automated group partitioning](#partition-by-grouping). 

![](images/wb-exp-13-ml.png)

| Field	| Description |
|-------|-------------|
| [Cross-validation](data-partitioning#k-fold-cross-validation-cv){ target=_blank }: 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. | :~~:|
| [Cross-validation folds](data-partitioning#k-fold-cross-validation-cv){ target=_blank } | Sets the number of folds used with the cross-validation method. A higher number increases the size of training data available in each fold; consequently increasing the total training time.|
| 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.|
|[Training-validation-holdout](data-partitioning#training-validation-and-holdout-tvh){ target=_blank }: For larger datasets, partitions data into three distinct sections&mdash;training, validation, and holdout&mdash; with predictions based on a single pass over the data. | :~~:|
| Validation percentage | Sets the percentage of data that Workbench uses for validation of the trained model.
| 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.

!!! note
	 If the dataset exceeds 800MB, training-validation-holdout is the only available validation type for all partitioning methods.

#### Partition by grouping {: #partition-by-grouping }

While less common, user-defined and automated group partitioning provide a method for partitioning by _partition feature_&mdash;a feature from the dataset that is the basis of grouping. 

* With _user-defined grouping_, one partition is created for each unique value of the selected partition feature. That is, rows are assigned to partitions using the values of the selected partition feature, one partition for each unique value. When this method is selected, DataRobot recommends specifying a feature that has fewer than 10 unique values of the partition feature.

* With _automated grouping_, all rows with the same single (specified) value of the partition feature are assigned to the same partition. Each partition can contain multiple values of that feature. When this method is selected, DataRobot recommends specifying a feature that has six or more unique values. 

Once either of these methods are selected, you are prompted to enter the partition feature. Help text provides information on the number of values the partition feature must contain; click in the dropdown to view features with a unique value count.

![](images/wb-exp-36-ml.png)

After choosing a partition feature, set the the validation type. The applicability of validation type is dependent on the unique values for the partition features, as illustrated in the following chart.

![](images/wb-exp-37-ml.png)

Automated grouping uses the same [validation settings](#set-the-validation-type) as described above. User-defined grouping, however, prompts for values specific to the partition feature. For _cross-validation_, setting holdout is optional. If you do set it, you select a value of the partition feature instead of a percentage. For _training-validation-holdout_, select a value of the partition feature for each section, again instead of a percentage.

![](images/wb-exp-38-ml.png)




### Configure additional settings {: #configure-additional-settings }

Choose the **Additional settings** tab to set more advanced modeling capabilities. Note that the **Time series modeling** tab will be available or greyed out depending on whether DataRobot found any date/time features in the dataset.

![](images/wb-exp-33.png)

Configure the following, as required by your business use case.

* [Monotonic feature constraints](#monotonic-feature-constraints)
* [Weight](#weight)
* [Insurance-specific settings](#insurance-specific-settings)

{% include 'includes/wb-config-additional-settings.md' %}

#### Insurance-specific settings {: #insurance-specific-settings }

Several features are available that address frequent weighting needs of the insurance industry. The table below describes each briefly, but more detailed information can be found [here](additional#additional-weighting-details).

![](images/wb-exp-35.png)

Setting | Description
------- | -----------
[Exposure](additional#set-exposure) | In regression problems, sets a feature to be treated with strict proportionality in target predictions, adding a measure of exposure when modeling insurance rates. DataRobot handles a feature selected for Exposure as a special column, adding it to raw predictions when building or scoring a model; the selected column(s) must be present in any dataset later uploaded for predictions.
[Count of Events](additional#set-count-of-events) | Improves modeling of a zero-inflated target by adding information on the frequency of non-zero events.
[Offset](additional#set-offset) | Adjusts the model intercept (linear model) or margin (tree-based model) for each sample; it accepts multiple features.

### Change the configuration {: #change-the-configuration }

You can make changes to the project's target or feature list before you begin modeling by returning to the **Target** page. To return, click the target icon, the **Back** button, or the **Target** field in the summary:

![](images/wb-exp-14.png)

## What's next? {: #whats-next }

After you start modeling, DataRobot populates the Leaderboard with models as they complete. You can:

* Begin [model evaluation](ml-experiment-evaluate) on any available model.
* Use the [**View experiment info**](ml-experiment-evaluate#view-experiment-info){ target=_blank } option to view a variety of information about the experiment.
