---
schema_version: "1"
id: fd60d6e1ad12199743eead51754cb8662e097ca31ebe6076ddc8635e62abd205
title: Prerequisites
collection: admin
version: current
locale: en-us
portal_url: https://docs.salesforce.com/admin/current/en-us/ind/264-0-0/ai_accelerator_configure_usecases_prerequisites.md
canonical_url: https://help.salesforce.com/s/articleView?id=ind.ai_accelerator_configure_usecases_prerequisites.htm&release=264.0.0&type=5
source_site: help.salesforce.com
content_type: documentation
content_hash: 62c55f1a2ad115751b68642eb4531126d3bfc3fb62c7cf6df8bf96a448ce4547
description: Before you configure a use case, create a Einstein Discovery model and deploy the generated predictive model. If your use-case requires features to be computed in real time, configure an Apex class with the feature extraction logic. Determine the objects and fields that you want to store the computed features and prediction results in.
products:
  - Industry Solutions
  - Cross Cloud
release: "264"
---
# Prerequisites

Before you configure a use case, create a Einstein Discovery model and deploy the generated predictive model. If your use-case requires features to be computed in real time, configure an Apex class with the feature extraction logic. Determine the objects and fields that you want to store the computed features and prediction results in.

### Required Editions

<table class="slds-table slds-table_bordered slds-max-medium-table_stacked slds-m-bottom_small edition"><colgroup><col></colgroup><tbody><tr class=""><td class="slds-cell-wrap" style="vertical-align:top;">Available in: Lightning Experience</td></tr></tbody></table>

![Note](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-ind-ai_accelerator_scoring_framework-264-0-0-production-enus/579cbc22-5fe0-45e4-b0a0-608ab8690ae6/images/icon_note.png)

Note You can also configure predictive use cases for machine learning models built in Amazon SageMaker. These models have their prediction platform set as Data Cloud, and predictions are processed in batches. Before configuring a use case for a SageMaker model, [set up your data in SageMaker](https://help.salesforce.com/s/articleView?id=data.c360_a_set_up_your_data.htm&type=5&language=en_US) and [connect the SageMaker model with Data Cloud](https://help.salesforce.com/s/articleView?id=data.c360_a_connect_your_external_model.htm&type=5&language=en_US) to define the prediction criteria for your use case.

1. Create an Insights & Predictions model for your use case. See [Create a Model](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_create.htm&type=5&language=en_US).
2. Deploy the predictive model to a prediction definition. See [Deploy Models](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_model_deploy.htm&type=5&language=en_US).
3. If you want to compute features in batches, add a supplemental dataset when mapping the model’s variables. For information about mapping model variables and adding a supplemental dataset, see [Map Model Variables](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_model_deploy_map.htm&type=5&language=en_US).
    
    When the feature input type is Batch Input, features from the supplemental dataset are used.
4. If you want to compute features in real time, configure an Apex class to implement the interface for a feature extractor. For information about configuring an Apex class, see [Adding an Apex Class](https://developer.salesforce.com/docs/atlas.en-us.apexcode.meta/apexcode/apex_qs_class.htm).
5. Identify the objects and fields in which you want to store the computed features and prediction results.
6. If you don’t want to save computed features and predictions results in the required fields of standard objects, create a custom object without any required fields.

![Note](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-ind-ai_accelerator_scoring_framework-264-0-0-production-enus/579cbc22-5fe0-45e4-b0a0-608ab8690ae6/images/icon_note.png)

Note The AI Accelerator facilitates the reuse of deployed predictive models across various use cases, streamlining the application and broadening the impact of existing models. Einstein Discovery (ED) supports the deployment and configuration of unused models for new use cases. Einstein Data Cloud (EDC) uses specified sorting criteria to sort results. The UI provides tooltips and examples to guide users through the sorting process. See [Create an Einstein Discovery Model](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_build.htm&type=5&language=en_US).

![Example](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-ind-ai_accelerator_scoring_framework-264-0-0-production-enus/579cbc22-5fe0-45e4-b0a0-608ab8690ae6/images/icon_example.png)

Example A mobile service provider wants to predict the likelihood of a customer switching to another service provider. The service provider’s Salesforce admin:

- Creates an Einstein Discovery model for the use case with the analysis type as Insights & Predictions.
- Deploys the generated predictive model to a new predictive definition.
- Configures an Apex class for computing features in real time.