---
title: Add/retrain models
description: Describes how to retrain Leaderboard models or add new models from the blueprint repository.
section_name: Time series
maturity: public-preview
---

# Add/retrain models {: #add-retrain-models }

There are two methods for adding new models to your experiment:

- [Retrain](#train-on-new-settings) existing Leaderboard models using new settings.
- Add new models from the blueprint [repository](#blueprint-repository).

This page describes adding and retraining for [date/time-partitioned experiments](ts-experiment-create#simple-date-time-partitioning){ target=_blank }. See also information on adding or retraining [random- or stratified-partitioned](ml-experiment-add){ target=_blank } experiments.


## Train on new settings {: #train-on-new-settings }

Once the Leaderboard is populated, you can retrain any existing model, which will create a new Leaderboard model. To retrain, select a model from the **Leaderboard** by clicking on it.

Change a model characteristic by clicking the change icon (![](images/icon-change-white.png)) next to the component in **Training settings**:

![](images/ts-exp-add-1.png)

### Change feature list (post-modeling) {: #change-feature-list-post-modeling }

To change the feature list:

1. Click the icon next to the current feature list to open the feature list selection modal. The current list is greyed out and unavailable for selection.

2. Select a new feature list. Note that you cannot change the feature list for the model prepared for deployment because it is a ["frozen" run](frozen-run){ target=_blank }.

![](images/ts-exp-add-2.png)

### Change training period {: #change-training-period }

To change the training period:

Click the icon to change the training period size and optionally [enforce a frozen run](frozen-run#start-a-frozen-run){ target=_blank }. While you can change the training range and sampling rate, you cannot change the duration of the validation partition once models are built.

!!! note
	Consider [retraining your model on the most recent data](otv#retrain-before-deployment){ target=_blank } before final deployment.

The **New Training Period** modal has multiple selectors, described below:

![](images/wb-exp-eval-16.png)

|   | Selection | Description |
|---|---|---|
| <div class="table-label">1</div> | Frozen run toggle  | [Freeze the run](frozen-run) ("freeze" parameter settings from a model’s early, smaller-sized run).|
| <div class="table-label">2</div>  | Training mode   | Rerun the model using a different training period. Before setting this value, see [the details](ts-customization#duration-and-row-count) of row count vs. duration and how they apply to different folds. |
| <div class="table-label">3</div>  | Snap to  | "Snap to" predefined points to facilitate entering values and avoid manually scrolling or calculation. |
| <div class="table-label">4</div>  | [Enable time window sampling](ts-leaderboard#time-window-sampling) | Train on a subset of data within a time window for a duration or [start/end](ts-leaderboard#setting-the-start-and-end-dates) training mode. Check to enable and specify a percentage. |
| <div class="table-label">5</div>  | [Sampling method](ts-leaderboard#set-rows-or-duration)   | Select the sampling method used to assign rows from the dataset. |
| <div class="table-label">6</div>   | Summary graphic | View a summary of the observations and testing partitions used to build the model. |
| <div class="table-label">7</div> | Final Model  | View an image that changes as you adjust the dates, reflecting the data to be used in the model you will make predictions with (see the [note](ts-predictions#about-final-models) about final models). |

Once you have set a new value, click **Train new models**. DataRobot builds the new model and displays it on the Leaderboard.

### Change monotonic feature lists {: #change-monotonic-feautre-lists }

To change the feature lists applied for monotonic modeling:

Click the icon next to **Monotonic constraints** and select at least one new feature list in the resulting modal. You must [create feature lists](wb-data-tab#create-a-feature-list){ target=_blank } to use for applying monotonic constraints prior to modeling. Note that if the model does not support monotonic constraints the label and icon are not displayed.

## Blueprint repository {: #blueprint-repository }

{% include 'includes/blueprint-repo.md' %}


### Add models {: #add-models }

From the blueprint repository, you can add one or more blueprints to your experiment by selecting the checkbox to the left of the blueprint name. Note the badges under the blueprint name, which in some cases indicate support for special modeling flows. For example, the MONO badge identifies blueprints that support monotonic constraints.

![](images/ts-exp-add-3.png)

1. Click the blueprint name to see a graphical representation of the tasks that comprise that blueprint.

2. Choose settings for the new model build; settings differ slightly depending on the partitioning method applied.


3. Set the feature list and [training period](#train-on-new-settings) to apply to all selected blueprints. For date/time models, DataRobot recommends a feature list. Use the recommended list or select a new list from the dropdown:

	![](images/ts-exp-add-4.png)

4. Once the configuration is set, click **Train models** to start building.

### Search models {: #search-models }

There are three ways to filter the Leaderboard display to show only those blueprints matching the selected criteria:

- Use the search bar to return all blueprints with matching strings in the name or description:

	![](images/ts-exp-add-5.png)

- Click a [badge](leaderboard-ref#tags-and-indicators){ target=_blank } to return all blueprints with that badge:

	![](images/ts-exp-add-6.png)

	Click again on the badge to remove it as a filter.

- Click **Edit filters** to choose blueprints by model family and/or property. Available fields and their settings are dependent on the project and/or model type.

	![](images/ts-exp-add-7.png)
