# March 2024

> March 2024 - Read about DataRobot's new preview and generally available features released on March
> 27, 2024.

This Markdown file sits beside the HTML page at the same path (with a `.md` suffix). It summarizes the topic and lists links for tools and LLM context.

Companion generated at `2026-05-06T18:17:10.005256+00:00` (UTC).

## Primary page

- [March 2024](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html): Full documentation for this topic (HTML).

## Sections on this page

- [In the spotlight](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#in-the-spotlight): In-page section heading.
- [Multiclass modeling and confusion matrix](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#multiclass-modeling-and-confusion-matrix): In-page section heading.
- [March features](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#march-features): In-page section heading.
- [GA](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#ga): In-page section heading.
- [Support for Parquet file ingestion](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#support-for-parquet-file-ingestion): In-page section heading.
- [Publish wrangling recipes with smart downsampling](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#publish-wrangling-recipes-with-smart-downsampling): In-page section heading.
- [Uploading, modeling, and generating insights on 10GB for OTV experiments](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#uploading-modeling-and-generating-insights-on-10gb-for-OTV-experiments): In-page section heading.
- [Tokenization improvements for Japanese text feature drift](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#tokenization-improvements-japanese-text-feature-drift): In-page section heading.
- [Custom model training data assignment update](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#custom-model-training-data-assignment-update): In-page section heading.
- [Share custom applications](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#share-custom-applications): In-page section heading.
- [Manage notebook file systems with codespaces](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#manage-notebook-file-systems-with-codespaces): In-page section heading.
- [Preview](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#preview): In-page section heading.
- [Generative AI with NeMo Guardrails on NVIDIA GPUs](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#generative-ai-with-nemo-guardrails-on-nvidia-gpus): In-page section heading.
- [Real-time predictions on DataRobot serverless prediction environments](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#realtime-predictions-on-datarobot-serverless-prediction-environments): In-page section heading.
- [Create hosted custom metrics from custom jobs](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#create-hosted-custom-metrics-from-custom-jobs): In-page section heading.
- [Disable column filtering for prediction requests](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#disable-column-filtering-for-prediction-requests): In-page section heading.
- [Manage custom applications in Registry](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#manage-custom-applications-in-registry): In-page section heading.
- [API](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#api): In-page section heading.
- [New import path for the datarobot-mlops package](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/march2024-announce.html#new-import-path-for-the-datarobot-mlops-package): In-page section heading.

## Related documentation

- [AI Platform releases](https://docs.datarobot.com/en/docs/release/index.html): Linked from this page.
- [Managed SaaS releases](https://docs.datarobot.com/en/docs/release/cloud-history/index.html): Linked from this page.
- [2024 AI Platform releases](https://docs.datarobot.com/en/docs/release/cloud-history/2024-announce/index.html): Linked from this page.
- [Self-Managed AI Platform release notes](https://docs.datarobot.com/en/docs/release/archive-release-notes/index.html): Linked from this page.
- [documentation](https://docs.datarobot.com/en/docs/workbench/nxt-workbench/experiments/create-experiments/create-predictive/ml-basic-experiment.html#classification-targets): Linked from this page.
- [enable smart downsampling in the publishing settings](https://docs.datarobot.com/en/docs/workbench/nxt-workbench/dataprep/wrangle-data/pub-recipe.html#configure-smart-downsampling): Linked from this page.
- [March 2023 release](https://docs.datarobot.com/en/docs/release/cloud-history/2023-announce/march2023-announce.html#assign-training-data-to-a-custom-model-version): Linked from this page.
- [Add training data to a custom model documentation](https://docs.datarobot.com/en/docs/classic-ui/mlops/deployment/custom-models/custom-model-workshop/custom-model-training-data.html): Linked from this page.
- [Custom model training data assignment update](https://docs.datarobot.com/en/docs/release/deprecations-and-migrations/cus-model-training-data.html): Linked from this page.
- [share custom apps](https://docs.datarobot.com/en/docs/wb-apps/custom-apps/manage-custom-app.html): Linked from this page.
- [codespaces](https://docs.datarobot.com/en/docs/workbench/wb-notebook/codespaces/index.html): Linked from this page.
- [documentation](https://docs.datarobot.com/en/docs/classic-ui/mlops/deployment/prediction-env/pred-env.html#predictions-on-datarobot-serverless-environments): Linked from this page.
- [Add generic custom job](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-jobs-workshop/nxt-create-jobs/nxt-create-custom-job.html#create-a-generic-custom-job): Linked from this page.
- [testing](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-model-workshop/nxt-test-custom-model.html): Linked from this page.
- [documentation](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-model-workshop/nxt-create-custom-model.html#disable-column-filtering-for-prediction-requests): Linked from this page.

## Documentation content

March 27, 2024

This page provides announcements of newly released features available in DataRobot's SaaS single- and multi-tenant AI Platform, with links to additional resources. From the release center, you can also access:

- Monthly deployment announcement history
- Self-Managed AI Platform release notes

## In the spotlight

### Multiclass modeling and confusion matrix

Multiclass experiments are now available in Workbench as a preview feature.

**Video: Multiclass modeling**

When building experiments, DataRobot determines the type based on the number of values for a given target feature. If more than two, the experiment is handled as either multiclass or regression (for numeric targets). Special handling by DataRobot allows:

- Changing a regression experiment to multiclass.
- Using aggregation, with configurable settings, to support more than 1000 classes.

Once a model is built, a multiclass confusion matrix helps to visualize where the model is, perhaps, mislabeling one class as another.

Preview [documentation](https://docs.datarobot.com/en/docs/workbench/nxt-workbench/experiments/create-experiments/create-predictive/ml-basic-experiment.html#classification-targets)

Feature flag ON by default: Unlimited multiclass

## March features

The following table lists each new feature:

**Features grouped by capability**

| Name | GA | Preview |
| --- | --- | --- |
| Data |  |  |
| Support for Parquet file ingestion | ✔ |  |
| Publish wrangling recipes with smart downsampling | ✔ |  |
| Modeling |  |  |
| Multiclass modeling and confusion matrix |  | ✔ |
| Uploading, modeling, and generating insights on 10GB for OTV experiments | ✔ |  |
| Predictions and MLOps |  |  |
| Tokenization improvements for Japanese text feature drift | ✔ |  |
| Custom model training data assignment update | ✔ |  |
| Generative AI with NeMo Guardrails on NVIDIA GPUs |  | ✔* |
| Real-time predictions on DataRobot serverless prediction environments |  | ✔ |
| Create hosted custom metrics from custom jobs |  | ✔ |
| Disable column filtering for prediction requests |  | ✔ |
| Notebooks |  |  |
| Manage notebook file systems with codespaces | ✔ |  |
| Apps |  |  |
| Share custom applications | ✔ |  |
| Manage custom applications in Registry |  | ✔ |
| API |  |  |
| New import path for the datarobot-mlops package | ✔ |  |
| * Premium feature |  |  |

### GA

#### Support for Parquet file ingestion

Ingestion of Parquet files is now GA for the AI Catalog, training datasets, and predictions datasets. The following Parquet file types are supported:

- Single Parquet files
- Single zipped Parquet files
- Multiple Parquet files (registered as separate datasets)
- Zipped multi-Parquet file (merged to create a single dataset in DataRobot)

#### Publish wrangling recipes with smart downsampling

After building a wrangling recipe in Workbench, [enable smart downsampling in the publishing settings](https://docs.datarobot.com/en/docs/workbench/nxt-workbench/dataprep/wrangle-data/pub-recipe.html#configure-smart-downsampling) to reduce the size of your output dataset and optimize model training. Smart downsampling is a data science technique that reduces the time it takes to fit a model without sacrificing accuracy, as well as account for class imbalance by stratifying the sample by class.

#### Uploading, modeling, and generating insights on 10GB for OTV experiments

To improve the scalability and user experience for OTV (out-of-time validation) experiments, this deployment introduces scaling for larger datasets. When using out-of-time validation as the partitioning method, DataRobot no longer needs to downsample for datasets as large as 10GB. Instead, a multistage Autopilot process supports the greatly expanded input allowance.

#### Tokenization improvements for Japanese text feature drift

Text tokenization for the Feature Details chart on the Data Drift tab is improved for Japanese text features, implementing word-gram-based data drift analysis with MeCab tokenization. In addition, default stop-word filtering is improved for Japanese text features.

#### Custom model training data assignment update

In April 2024, the assignment of training data to a custom model version—announced in the [March 2023 release](https://docs.datarobot.com/en/docs/release/cloud-history/2023-announce/march2023-announce.html#assign-training-data-to-a-custom-model-version) —replaces the deprecated method of assigning training data at the custom model level. This means that the “per custom model version” method becomes the default, and the “per custom model” method is removed. In preparation, you can review the manual conversion instructions in the [Add training data to a custom model documentation](https://docs.datarobot.com/en/docs/classic-ui/mlops/deployment/custom-models/custom-model-workshop/custom-model-training-data.html):

The automatic conversion of any remaining custom models using the “per custom model” method will occur automatically when the deprecation period ends, assigning the training data at the custom model version level. For most users, no action is required; however, if you have any remaining automation relying on unconverted custom models using the “per custom model” assignment method, you should update them to support the “per custom model version” method to avoid any gaps in functionality.

For a summary of the assignment method changes, you can view the [Custom model training data assignment update](https://docs.datarobot.com/en/docs/release/deprecations-and-migrations/cus-model-training-data.html) documentation.

#### Share custom applications

You can [share custom apps](https://docs.datarobot.com/en/docs/wb-apps/custom-apps/manage-custom-app.html) with both DataRobot and non-DataRobot users, expanding the reach of custom applications you create. Access sharing functionality from the actions menu on the Applications page. To share a custom app with non-DataRobot users, you must toggle on Enable external sharing and specify the email domains and addresses that are permitted access to the app. DataRobot provides a link to share with these users after configuring the sharing settings.

#### Manage notebook file systems with codespaces

Now generally available, DataRobot Workbench includes [codespaces](https://docs.datarobot.com/en/docs/workbench/wb-notebook/codespaces/index.html) to enhance the code-first experience, especially when working with DataRobot Notebooks. A codespace, similar to a repository or folder file tree structure, can contain any number of files and nested folders. Within a codespace, you can open, view, and edit multiple notebook and non-notebook files at the same time. You can also execute multiple notebooks in the same container session (with each notebook running on its own kernel).

For Managed AI Platform users, DataRobot provides backup functionality and retention policies for codespaces. DataRobot takes snapshots of the codespace volume on session shutdown and on codespace deletion and will retain the contents for 30 days in the event you want to restore the codespace data.

### Preview

#### Generative AI with NeMo Guardrails on NVIDIA GPUs

Use NVIDIA with DataRobot to quickly build out end-to-end generative AI (GenAI) capabilities by unlocking accelerated performance and leveraging the very best of open-source models and guardrails. The DataRobot integration with NVIDIA creates an inference software stack that provides full, end-to-end Generative AI capability, ensuring performance, governance, and safety through significant functionality out of the box.

When you create a custom Generative AI model in the NextGen DataRobot Model workshop, you can select an [NVIDIA] Triton Inference Server (vLLM backend) base environment. DataRobot has natively built in the [NVIDIA Triton Inference Server](https://developer.nvidia.com/triton-inference-server) to provide extra acceleration for all of your GPU-based models as you build and deploy them onto NVIDIA devices.

Then, navigating to the custom model's resources settings, you can select a resource bundle from the range of NVIDIA devices available as build environments in DataRobot.

In addition, DataRobot also provides a powerful interface to create custom metrics through an integration with NeMo Guardrails. The integration with NeMo provides powerful rails to ensure your model stays on topic, using interventions to block prompts and completions if they violate the "on topic" principles provided by NeMo.

Feature flags OFF by default: The NVIDIA and NeMo Guardrails integrations require access to premium features for GenAI experimentation, LLM assessment, and GPU inference. Contact your DataRobot representative or administrator for information on enabling the required features.

#### Real-time predictions on DataRobot serverless prediction environments

Create DataRobot serverless prediction environments to make scalable real-time predictions on Kubernetes, with configurable compute instance settings. To create a DataRobot serverless prediction environment for real-time predictions in Kubernetes, when you add a prediction environment, set the Platform to DataRobot Serverless:

When you deploy a model to a DataRobot serverless prediction environment, you can configure the compute instance settings in Advanced Predictions Configuration:

Preview [documentation](https://docs.datarobot.com/en/docs/classic-ui/mlops/deployment/prediction-env/pred-env.html#predictions-on-datarobot-serverless-environments).

Feature flag OFF by default: Enable Serverless Predictions on K8s

#### Create hosted custom metrics from custom jobs

When you create a new custom job in the Registry, you can create new hosted custom metrics or add metrics from the gallery. In the NextGen interface, click Registry > Jobs and then click + Add new (or the button when the custom job panel is open) to select one of the following custom metric types:

| Custom job type | Description |
| --- | --- |
| Add generic custom job | Add a custom job to implement automation (for example, custom tests) for models and deployments. |
| Create new | Add a custom job for a new hosted custom metric, defining the custom metric settings and associating the metric with a deployment. |
| Create new from template | Add a custom job for a custom metric from a template provided by DataRobot, associating the metric with a deployment and setting a baseline. |

After you create a custom metric job on the Registry > Jobs tab, on the custom metric job's Assemble tab, you can access the Connected deployments panel. Click + Connect to deployment, define a custom metric name, and then select a deployment ID to link the metric to a deployment:

Preview [documentation](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-jobs-workshop/nxt-create-jobs/nxt-create-custom-job.html).

Feature flags OFF by default: Enable Hosted Custom Metrics, Enable Notebooks Custom Environments

#### Disable column filtering for prediction requests

When you assemble a custom model, you can enable or disable column filtering for custom model predictions. The filtering setting you select is applied in the same way during custom model [testing](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-model-workshop/nxt-test-custom-model.html) and deployment. By default, the target column is filtered out of prediction requests and, if training data is assigned, any additional columns not present in the training dataset are filtered out of any scoring requests sent to the model. Alternatively, if the prediction dataset is missing columns, an error message appears to notify you of the missing features.

You can disable this column filtering when you assemble a custom model. In the Model workshop, open a custom model to the Assemble tab, and, in the Settings section, under Column filtering, clear Exclude target column from requests (or, if training data is assigned, clear Exclude target and extra columns not in training data):

Preview [documentation](https://docs.datarobot.com/en/docs/workbench/nxt-registry/nxt-model-workshop/nxt-create-custom-model.html#disable-column-filtering-for-prediction-requests).

Feature flag OFF by default: Enable Feature Filtering for Custom Model Predictions

#### Manage custom applications in Registry

Now available for preview, the Applications page in the NextGen Registry is home to all custom applications and application sources available to you. You can now create application sources, which contain the files, environment, and runtime parameters for custom applications, and build directly from these sources. You can also use the Applications page to manage applications by sharing or deleting them.

Preview [documentation](https://docs.datarobot.com/en/docs/wb-apps/custom-apps/manage-custom-app.html).

Feature flag OFF by default: Enable Custom Applications Workshop

### API

#### New import path for the datarobot-mlops package

In the [datarobot-mlops and datarobot-mlops-connected-client](https://pypi.org/project/datarobot-mlops/10.0.3a2/) Python packages, the import path is changing from `import datarobot.mlops.mlops` to `import datarobot_mlops.mlops`. This fixes an issue where the DataRobot package and the MLOps packages conflict with each other when installed to the same Python environment. You must manually update this import. The example below shows how you can update your code to be compatible by attempting both import paths:

```
try:
    from datarobot_mlops.mlops import MLOps
except ImportError:
    from datarobot.mlops.mlops import MLOps
```

All product and company names are trademarks™ or registered® trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them.