--- schema_version: "1" id: ef563202008bdc6b0180d1cb637794c9282626d85b4f82b1f7a7c7c3da13a3d1 title: Configure Number Variables collection: admin version: current locale: en-us portal_url: https://docs.salesforce.com/admin/current/en-us/analytics/264-0-0/bi_edd_story_number.md canonical_url: https://help.salesforce.com/s/articleView?id=analytics.bi_edd_story_number.htm&release=264.0.0&type=5 source_site: help.salesforce.com content_type: documentation content_hash: 408ade6e962f3bf36e0c239d95784707e6e26b80ccbaf4bcbd94fc25cb998e4c description: Configure settings for individual number variables in your model. products: - CRM Analytics - Cross Cloud release: "264" --- # Configure Number Variables Configure settings for individual number variables in your model. ### Required Editions ![Note](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/images/icon_note.png) Note Einstein Discovery stories are now models. We wish we could snap our fingers to update the name everywhere, but you can expect to see the previous name in a few places until we replace it.
Available in Salesforce Classic and Lightning Experience.
Available with CRM Analytics, which is available for an extra cost in Enterprise, Performance, and Unlimited Editions. Also available in Developer Edition.
User Permissions Needed
To configure the number fields in a model: Create and Update Einstein Discovery Models
1. On the Model Settings page, click a numeric variable. 2. On the Alert tab, respond to any suggestions regarding data issues for this field. For more information, see [Handle Quality Alerts](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_quality_alert.htm&release=264.0.0&type=5 "Einstein Discovery displays quality alerts when it detects issues in your training data, your model’s quality metrics, or during cross-validation."). 3. On the Performance tab, see how well your model works for each value within the selected variable. Performance is shown as a decimal score, where 1 is perfect accuracy. For example, in the variable DailyQuantity, values “143 to 193” have a performance score of 0.73, and values “501 to 791” have a score of 0.67. You can then conclude that the model performs better when the daily quantity is from 143 through 193 than when it’s from 501 through 791. ![Performance tab for a number variable](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_story_number_performance.png) 1. Optionally, if using a binary classification or regression model, click **Row Count Analysis** to see a detailed comparison of the values by performance and row count. ![Row Count Analysis chart for a number variable in a binary classification or regression model](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_story_number_performance_rowCountAnalysis.png) 2. Optionally, if using a multiclass classification model, see the performance of each value by outcome. - To expand a value to see performance by outcome, click ![Actions button](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/resources-global/analytics/images/actions_button.png). - To see a graphic comparison of values and outcomes by performance and row count, click **Row Count Analysis**. - Use the dropdown to select different outcomes (1). The chart updates according to the selected outcome. - Toggle between Actual Class and Predicted Class (2). The **Actual Class** groups data based on the observed value. The **Predicted Class** groups data based on the predicted value. ![Row Count Analysis chart for a number variable in a multiclass model](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_story_number_performance_multiclass_rowCountAnalysis.png) - To see a graph of performance by value and outcome, click **Detailed Analysis**. The darker the circle, the better the performance. The bigger the circle, the higher the row count. ![Detailed Analysis chart for a number variable in a multiclass model](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_story_number_performance_detailed_analysis.png) 4. In the Settings tab, configure the following settings. ![Edit Variable window, showing the Settings tab for a number field](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_story_number_settings.png)
Setting Description
Analyze for bias

Select this option to exclude a variable from the model so that it doesn’t influence predictions and recommendations. If selected, Einstein Discovery shows a shield icon next to the title of the insight to remind you it’s a sensitive variable.

Shield icon for sensitive fields

This enables you to evaluate and assess the field’s impact in the model. Einstein Discovery still notifies you if it shows a 50% or higher correlation to the model’s outcome variable.
Transform

Select a transformation to apply data improvements.

  • Replace Missing Values enables statistical imputation and fills in missing values. Use this transformation to improve model performance. See Replace Missing Numerical Values.
    Note
    Note Replace Missing Values influences the model generated during analysis but has no effect on the generated insights.
  • Projected Predictions allows you to provide trending data that factor into your predictions to make them more accurate. See Configure Projected Predictions.
    Note
    Note Projected Predictions influences the model generated during analysis but has no effect on the generated insights.
Bucket Values By

Select how to bucket data.

Note
Note A numeric variable with low cardinality (10 or fewer observations) is converted to a text variable. Einstein Discovery displays this value along the x-axis rather than in a range (bucket).
Number of Buckets Specify the number of buckets to show in charts. Einstein Discovery automatically aggregates numbers into groups based on behavior.
Include Only Create a numeric range by adding minimum and maximum values to the Starting Value and Ending Value fields.
Histogram A bar chart graph shows the number of values that occur across the range of numbers.
Changes take effect after you create the model. - **[Replace Missing Numerical Values](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_story_impute.htm&release=264.0.0&type=5)** Einstein Discovery allows you to replace missing numerical values in your model. Use the Replace Missing Variables transformation to enable statistical imputation on a numeric variable. Einstein Discovery automatically replaces missing values with data derived from another subset of your data. - **[Configure Projected Predictions](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_story_number_projected_predictions.htm&release=264.0.0&type=5)** Einstein Discovery allows you to provide trending data for numeric variables that factor into your predictions to make them more accurate. For example, you predict opportunity win/loss based on expected price fluctuations for raw materials. To factor price into your predicted outcome, use the Projected Predictions transformation and provide a dataset that projects the expected price for raw materials on the close date.