Dashboard mutations - Arize AX Docs

Create and copy dashboards, and add, update, duplicate or delete their widgets. For task-oriented walkthroughs of these operations, see the Dashboard guide. Every mutation below is sent as a POST to https://app.arize.com/graphql with an x-api-key header; see Forming calls.

Mutations in this page

createDashboard: Create dashboard.
createDefaultDashboard: Create default dashboard.
createDashboardFromTemplate: Create dashboard from template.
copyDashboard: Copy dashboard.
updateDashboardName: Update dashboard name.
updateDashboardStatus: Update dashboard status.
duplicateWidget: Duplicate widget.
createLineChartWidget: Update a single line chart widget on a dashboard
updateLineChartWidget: Update line chart widget.
deleteLineChartWidget: Delete line chart widget.
createBarChartWidget: Create bar chart widget.
updateBarChartWidget: Update bar chart widget.
deleteBarChartWidget: Delete bar chart widget.
createStatisticWidget: Create statistic widget.
updateStatisticWidget: Update a single widget
deleteStatisticWidget: Delete statistic widget.
createTextWidget: Create text widget.
updateTextWidget: Update a single widget
deleteTextWidget: Delete text widget.
createPivotTableWidget: Create pivot table widget.
updatePivotTableWidget: Update a single pivot table widget
deletePivotTableWidget: Delete pivot table widget.
createExperimentChartWidget: Create an experiment chart widget
updateExperimentChartWidget: Update experiment chart widget.
deleteExperimentChartWidget: Delete experiment chart widget.

Reference

createDashboard

createDashboard(input: CreateDashboardMutationInput!): CreateDashboardMutationPayload

Arguments

CreateDashboardMutationInput!

required

Show CreateDashboardMutationInput fields

String!

required

ID!

required

String

Returns

Dashboard

Show Dashboard fields

Scalar fields: id, name, createdAt, status, timezone, startTime, endTime, timeSeriesDataGranularity, timeRangeKey, isRestricted.Object fields (select subfields): creator, space, models, statisticWidgets, performanceSlices, barChartWidgets, lineChartWidgets, experimentChartWidgets, driftLineChartWidgets, monitorLineChartWidgets, scatterChartWidgets, textWidgets, pivotTableWidgets, tags, dashboardRoleBindings, _access.

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateDashboard($input: CreateDashboardMutationInput!) {
  createDashboard(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "name": "<string>",
    "spaceId": "<ID>"
  }
}

createDefaultDashboard

createDefaultDashboard(input: CreateDefaultDashboardMutationInput!): CreateDefaultDashboardMutationPayload

Arguments

CreateDefaultDashboardMutationInput!

required

Show CreateDefaultDashboardMutationInput fields

ID!

required

ID!

required

ModelEnvironmentName!

required

One of: production, validation, training, tracing.

[DimensionValueInput!]

Show DimensionValueInput fields

ID!

required

String!

required

The value of this dimension value, e.g. ‘Wells Fargo’ or ‘60’

DimensionValueSimilarityReference

The similarity references for this dimension value

Show DimensionValueSimilarityReference fields

Float!

required

The threshold for this similarity reference, e.g. ‘Wells Fargo’ or ‘60’

[SimilarityReferenceInput!]!

required

The list of references for this similarity reference Nested SimilarityReferenceInput (same shape as above).

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateDefaultDashboard($input: CreateDefaultDashboardMutationInput!) {
  createDefaultDashboard(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "spaceId": "<ID>",
    "modelId": "<ID>",
    "modelEnvironmentName": "production"
  }
}

createDashboardFromTemplate

createDashboardFromTemplate(input: CreateDashboardFromTemplateMutationInput!): CreateDashboardFromTemplateMutationPayload

Arguments

CreateDashboardFromTemplateMutationInput!

required

Show CreateDashboardFromTemplateMutationInput fields

String!

required

ID!

required

ModelEnvironmentName

One of: production, validation, training, tracing.

[String!]

ModelEnvironmentName

One of: production, validation, training, tracing.

[String!]

DashboardTemplates!

required

The template type of the dashboard One of: regression, scoredModel, rankingModel, multiClassModel, topFeatures, featureAnalysis, featurePerformanceHeatmap, featureSlices, compareModels, compareProdVsTrain, generativeLlmModel, generativeLlmModelV2, featureDrift, modelMonitor, regressionPreProduction, scoredModelPreProduction, classificationPreProduction, training, validation.

[String!]

The features for a given model to use for this dashboard

String

The class of the classification model

PositiveInt

The @K value for the performance metric. Model must be ranking

[String!]

The value of this feature value, e.g. ‘Wells Fargo’ or ‘60’

Metric

The performance metric used for the dashboard One of: accuracy, precision, recall, f_1, sensitivity, specificity, falsePositiveRate, falseNegativeRate, falseNegativeDensity, mse, rmse, mae, mape, predictionAverage, actualsAverage, auc, logLoss, rSquared, prAuc, meanError, calibration, ndcg, recallParity, falsePositiveRateParity, disparateImpact, wape, smape, mase, udf, groupAuc, mrr, rankingMap, rankingRecall, rankingPrecision, avg, count, sum, percentEmpty, cardinality, psi, kl, js, ks, newValues, missingValues, averageStringListLength, microAveragedPrecision, microAveragedRecall, multiClassPrecision, multiClassRecall, macroAveragedPrecision, macroAveragedRecall, standardDeviation, p50, p95, p99, p99_9.

Int

The number of slices to create for histogram charts

A comparison dataset pointing to the plot’s model baseline that will be referenced by this plot for drift and comparison calculations.

[ID!]

List of monitor ids to create alert widgets

[String!]

The names of the monitors

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateDashboardFromTemplate($input: CreateDashboardFromTemplateMutationInput!) {
  createDashboardFromTemplate(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "name": "<string>",
    "modelId": "<ID>",
    "template": "regression"
  }
}

copyDashboard

copyDashboard(input: CopyDashboardMutationInput!): CopyDashboardMutationPayload

Arguments

CopyDashboardMutationInput!

required

Show CopyDashboardMutationInput fields

ID!

required

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CopyDashboard($input: CopyDashboardMutationInput!) {
  copyDashboard(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "dashboardId": "<ID>"
  }
}

updateDashboardName

updateDashboardName(input: UpdateDashboardNameMutationInput!): UpdateDashboardNameMutationPayload

Arguments

UpdateDashboardNameMutationInput!

required

Show UpdateDashboardNameMutationInput fields

String!

required

ID!

required

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateDashboardName($input: UpdateDashboardNameMutationInput!) {
  updateDashboardName(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "name": "<string>",
    "dashboardId": "<ID>"
  }
}

updateDashboardStatus

updateDashboardStatus(input: UpdateDashboardStatusMutationInput!): UpdateDashboardStatusMutationPayload

Arguments

UpdateDashboardStatusMutationInput!

required

Show UpdateDashboardStatusMutationInput fields

String!

required

ID!

required

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateDashboardStatus($input: UpdateDashboardStatusMutationInput!) {
  updateDashboardStatus(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "status": "<string>",
    "dashboardId": "<ID>"
  }
}

duplicateWidget

duplicateWidget(input: DuplicateWidgetMutationInput!): DuplicateWidgetMutationPayload

Arguments

DuplicateWidgetMutationInput!

required

Show DuplicateWidgetMutationInput fields

ID!

required

[Int!]!

required

The grid position the duplicated widget should be placed at.

String!

required

The title of the duplicate widget.

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

String

The label for the y-axis

BarChartSortOrder

The sort order of the bars One of: label_asc, label_desc, vol_asc, vol_desc.

Float

The top N bars to display

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

String

The timezone of the widget

DateTime

The start time of the widget

DateTime

The end time of the widget

Boolean

Whether to overwrite the global time

DataGranularity

The granularity of the time series data One of: oneMinute, fiveMinute, hour, day, week, month.

String

The time range key of the widget

String

The custom statistic text of the widget

String

Returns

Show BarChartWidget fields

Scalar fields: id, dashboardId, title, gridPosition, creationStatus, sortOrder, yMin, yMax, yAxisLabel, topN, isNormalized, isStacked, binOption, numBins, customBins, quantiles, performanceMetric, fontSize, showLegend, legendPlacement, xAxisLabel, xAxisTilt, timezone, startTime, endTime, timeSeriesDataGranularity, overwriteGlobalTime, timeRangeKey.Object fields (select subfields): plots, config, bars, stackedBars.

Show StatisticWidget fields

Scalar fields: dashboardId, title, modelId, modelVersionIds, dimensionCategory, performanceMetric, aggregation, rankingAtK, modelEnvironmentName, gridPosition, timeSeriesMetricType, id, creationStatus, customStatisticText, statisticValue.Object fields (select subfields): predictionValueClass, modelVersionEnvironmentBatches, filters, dimension, model, modelVersions, customMetric.

Show LineChartWidget fields

Scalar fields: id, dashboardId, title, gridPosition, creationStatus, timeSeriesMetricType, yMin, yMax, yAxisLabel, fontSize, showLegend, legendPlacement, xAxisLabel, xAxisTilt, timezone, startTime, endTime, timeSeriesDataGranularity, overwriteGlobalTime, timeRangeKey, widgetType.Object fields (select subfields): driftDistribution, monitor, config, plots, lines, scatterData.

Show TextWidget fields

Scalar fields: id, title, content, gridPosition, creationStatus, dashboardId.

Show ExperimentChartWidget fields

Scalar fields: id, dashboardId, title, gridPosition, creationStatus, fontSize, showLegend, legendPlacement, xAxisLabel, xAxisTilt, yMin, yMax.Object fields (select subfields): plots, lines.

Show PivotTableWidget fields

Scalar fields: id, dashboardId, title, modelId, modelVersionIds, gridPosition, creationStatus, groupedDimension, groupedDimensionTitle, groupedDimensionCategory, modelEnvironmentName, overwriteGlobalTime, timezone, startTime, endTime, timeRangeKey, queryFilter.Object fields (select subfields): modelVersionEnvironmentBatches, model, modelVersions, columns, filters, pivotData.

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DuplicateWidget($input: DuplicateWidgetMutationInput!) {
  duplicateWidget(input: $input) {
    barChartWidget { id }
    statisticWidget { id }
    lineChartWidget { id }
  }
}

{
  "input": {
    "widgetId": "<ID>",
    "dashboardId": "<ID>",
    "gridPosition": [
      1
    ],
    "title": "<string>"
  }
}

createLineChartWidget

Update a single line chart widget on a dashboard createLineChartWidget(input: CreateLineChartWidgetMutationInput!): CreateLineChartWidgetMutationPayload

Arguments

CreateLineChartWidgetMutationInput!

required

Show CreateLineChartWidgetMutationInput fields

String!

required

WidgetCreationStatus

One of: needsInit, pending, created, published, unpublished.

ID!

required

TimeSeriesMetricCategory!

required

One of: modelDataMetric, evaluationMetric.

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

[Int!]

The grid position of the widget, defaults to next available slot if not specified

[LineChartPlotInputInput!]

Show LineChartPlotInputInput fields

ID!

required

The model ID on the widget.

[ID!]!

required

The model version IDs on the widget. If empty, use all versions.

ModelEnvironmentName!

required

The model environment name on the plot. One of: production, validation, training, tracing.

[ID!]

The model version environment batch IDs on the plot. If empty, use all batches

DimensionCategory

The dimension category of the widget plot. ex. featureLabel, prediction, actuals One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

CustomMetricInput

The custom metric function of the widget, if specified.

Show CustomMetricInput fields

ID!

required

String!

required

The name of the custom metric

String

The description of the custom metric

String!

required

The custom metric, ex. ‘AVG(predictionScore)’, etc.

Boolean!

required

Whether the custom metric requires a positive class to calculate

Metric!

required

The metric function of the plot i.e. accuracy, count, etc One of: accuracy, precision, recall, f_1, sensitivity, specificity, falsePositiveRate, falseNegativeRate, falseNegativeDensity, mse, rmse, mae, mape, predictionAverage, actualsAverage, auc, logLoss, rSquared, prAuc, meanError, calibration, ndcg, recallParity, falsePositiveRateParity, disparateImpact, wape, smape, mase, udf, groupAuc, mrr, rankingMap, rankingRecall, rankingPrecision, avg, count, sum, percentEmpty, cardinality, psi, kl, js, ks, newValues, missingValues, averageStringListLength, microAveragedPrecision, microAveragedRecall, multiClassPrecision, multiClassRecall, macroAveragedPrecision, macroAveragedRecall, standardDeviation, p50, p95, p99, p99_9.

DimensionInput

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

The category of the dimension, e.g. ‘tag’ or ‘featureLabel’, etc. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

String

The class of the classification model (categorical prediction values) on the widget.Provided only if timeseriesMetricType = ‘evaluationMetrics’ and prediction value type is ‘categorical’.

PositiveInt

The @K value for the performance metric. Model must be ranking.

String!

required

The title of the plot.

Int!

required

The position of the plot on the widget.Used to maintain order of widget and edit panel

[LineChartFilterItemInputInput!]!

required

The filters on this plot.

Show LineChartFilterItemInputInput fields

FilterRowType!

required

The type of filter. One of: featureLabel, tagLabel, predictionValue, actuals, modelVersion, batchId, predictionClass, predictionScore, actualClass, actualScore, topKPercentile, spanProperty, llmEval, annotation, userAnnotation.

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age. Nested DimensionInput (same shape as above).

[DimensionValueInput!]

The dimension values of the filter. Nested DimensionValueInput (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

A comparison dataset pointing to the plot’s model baseline that will be referenced by this plot for drift and comparison calculations.

FilteredBaselineInput

A comparison dataset pointing to a stream of data that will be referenced by this plot for drift and comparison calculations. For example: last month’s production data.

Show FilteredBaselineInput fields

ModelEnvironmentName!

required

The model environment on the plot, ex. production, training, validation. Defaulted to production. One of: production, validation, training, tracing.

Float

The evaluation window length of the moving window.

Float

The evaluation window delay length of the moving window.

DateTime

The start bookend of the baseline’s time range. If set, this time should be used for the baseline

DateTime

The end bookend of the baseline’s time range. If set, this time should be used for the baseline

The batchId that you are tagging as the baseline

SetModelDataSetBaselineInput

A comparison dataset pointing to a specific batch that will be referenced by this plot. For example: Batch: Test, Environment: Validation

Show SetModelDataSetBaselineInput fields

ID!

required

The batchId that you are tagging as the baseline

DateTime

The override start bookend of the baseline’s time range.

DateTime

The override end bookend of the baseline’s time range

Boolean

Whether or not split by is enabled on the plot.

String

The dimension to split by

DimensionCategory

The dimension category to split by One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

Boolean

If splitBy is enabled, this determines whether to show the overall metric alongside the splits.

[String]

The cohorts to filter by

Boolean

When true, split-by cohorts are re-resolved from live data on every fetch instead of using the frozen cohorts list.

[String!]

A list of colors to use when plotting. Use one or two colors, two will be a gradient.

MetricAggregationType

The aggregation type of the metric. One of: span, trace, session.

String

The X-axis dimension name for scatter plots.

DimensionCategory

The X-axis dimension category for scatter plots. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

String

The Y-axis dimension name for scatter plots.

DimensionCategory

The Y-axis dimension category for scatter plots. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

String

The color-by dimension name for scatter plots (optional, categorical).

DimensionCategory

The color-by dimension category for scatter plots. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

One of: lineChartWidget, barChartWidget, statisticWidget, driftLineChartWidget, textWidget, monitorLineChartWidget, experimentChartWidget, scatterChartWidget, pivotTableWidget.

The monitor id when creating a monitorLineChartWidget

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

String

The timezone of the widget

DateTime

The start time of the widget

DateTime

The end time of the widget

Boolean

Whether to overwrite the global time

DataGranularity

The granularity of the time series data One of: oneMinute, fiveMinute, hour, day, week, month.

String

The time range key of the widget

String

Returns

Show LineChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateLineChartWidget($input: CreateLineChartWidgetMutationInput!) {
  createLineChartWidget(input: $input) {
    lineChartWidget { id }
  }
}

{
  "input": {
    "title": "<string>",
    "dashboardId": "<ID>",
    "timeSeriesMetricType": "modelDataMetric"
  }
}

updateLineChartWidget

updateLineChartWidget(input: UpdateLineChartWidgetMutationInput!): UpdateLineChartWidgetMutationPayload

Arguments

UpdateLineChartWidgetMutationInput!

required

Show UpdateLineChartWidgetMutationInput fields

ID!

required

ID!

required

LineChartWidgetUpdateDataInput!

required

Show LineChartWidgetUpdateDataInput fields

String!

required

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

TimeSeriesMetricCategory!

required

One of: modelDataMetric, evaluationMetric.

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

String

The label for the y-axis

[Int!]!

required

[LineChartPlotInputInput!]!

required

Show LineChartPlotInputInput fields

ID!

required

The model ID on the widget.

[ID!]!

required

The model version IDs on the widget. If empty, use all versions.

ModelEnvironmentName!

required

The model environment name on the plot. One of: production, validation, training, tracing.

[ID!]

The model version environment batch IDs on the plot. If empty, use all batches

DimensionCategory

CustomMetricInput

The custom metric function of the widget, if specified. Nested CustomMetricInput (same shape as above).

Metric!

required

DimensionInput

Nested DimensionInput (same shape as above).

String

The class of the classification model (categorical prediction values) on the widget.Provided only if timeseriesMetricType = ‘evaluationMetrics’ and prediction value type is ‘categorical’.

PositiveInt

The @K value for the performance metric. Model must be ranking.

String!

required

The title of the plot.

Int!

required

The position of the plot on the widget.Used to maintain order of widget and edit panel

[LineChartFilterItemInputInput!]!

required

The filters on this plot. Nested LineChartFilterItemInputInput (same shape as above).

A comparison dataset pointing to the plot’s model baseline that will be referenced by this plot for drift and comparison calculations.

FilteredBaselineInput

A comparison dataset pointing to a stream of data that will be referenced by this plot for drift and comparison calculations. For example: last month’s production data. Nested FilteredBaselineInput (same shape as above).

SetModelDataSetBaselineInput

A comparison dataset pointing to a specific batch that will be referenced by this plot. For example: Batch: Test, Environment: Validation Nested SetModelDataSetBaselineInput (same shape as above).

Boolean

Whether or not split by is enabled on the plot.

String

The dimension to split by

DimensionCategory

Boolean

If splitBy is enabled, this determines whether to show the overall metric alongside the splits.

[String]

The cohorts to filter by

Boolean

When true, split-by cohorts are re-resolved from live data on every fetch instead of using the frozen cohorts list.

[String!]

A list of colors to use when plotting. Use one or two colors, two will be a gradient.

MetricAggregationType

The aggregation type of the metric. One of: span, trace, session.

String

The X-axis dimension name for scatter plots.

DimensionCategory

String

The Y-axis dimension name for scatter plots.

DimensionCategory

String

The color-by dimension name for scatter plots (optional, categorical).

DimensionCategory

Associated monitor

One of: lineChartWidget, barChartWidget, statisticWidget, driftLineChartWidget, textWidget, monitorLineChartWidget, experimentChartWidget, scatterChartWidget, pivotTableWidget.

DataGranularity

The granularity of the time series data One of: oneMinute, fiveMinute, hour, day, week, month.

Boolean

Whether to overwrite the global time

String

The timezone of the widget

DateTime

The start time of the widget

DateTime

The end time of the widget

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

String

The time range key of the widget

String

Returns

Show LineChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateLineChartWidget($input: UpdateLineChartWidgetMutationInput!) {
  updateLineChartWidget(input: $input) {
    lineChartWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "creationStatus": "needsInit",
      "timeSeriesMetricType": "modelDataMetric",
      "gridPosition": [
        1
      ],
      "plots": [
        {
          "modelId": "<ID>",
          "modelVersionIds": [
            "<ID>"
          ],
          "modelEnvironmentName": "production",
          "metric": "accuracy",
          "title": "<string>",
          "position": 1,
          "filters": [
            {
              "filterType": "featureLabel",
              "operator": "greaterThan"
            }
          ]
        }
      ]
    }
  }
}

deleteLineChartWidget

deleteLineChartWidget(input: DeleteLineChartWidgetMutationInput!): DeleteLineChartWidgetMutationPayload

Arguments

DeleteLineChartWidgetMutationInput!

required

Show DeleteLineChartWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeleteLineChartWidget($input: DeleteLineChartWidgetMutationInput!) {
  deleteLineChartWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "lineChartWidgetId": "<ID>"
  }
}

createBarChartWidget

createBarChartWidget(input: CreateBarChartWidgetMutationInput!): CreateBarChartWidgetMutationPayload

Arguments

CreateBarChartWidgetMutationInput!

required

Show CreateBarChartWidgetMutationInput fields

String!

required

WidgetCreationStatus

One of: needsInit, pending, created, published, unpublished.

ID!

required

[Int!]

The grid position of the widget, defaults to next available slot if not specified

Boolean

PerformanceMetric

One of: accuracy, precision, recall, f_1, sensitivity, specificity, falsePositiveRate, falseNegativeRate, falseNegativeDensity, mse, rmse, mae, mape, predictionAverage, actualsAverage, auc, logLoss, rSquared, prAuc, meanError, calibration, ndcg, recallParity, falsePositiveRateParity, disparateImpact, wape, smape, mase, udf, groupAuc, mrr, rankingMap, rankingRecall, rankingPrecision, microAveragedPrecision, microAveragedRecall, multiClassPrecision, multiClassRecall, macroAveragedPrecision, macroAveragedRecall.

CustomBinOption

The distribution bin option of the widget, if specified One of: numBins, customBins, medianCentered, discreteBins, decileBins, quantiles, discreteTopN.

Int

The number of bins on the distribution.

[Float!]

The bin edges on the distribution.

[Float!]

The array of percentiles of the distribution.

[BarChartPlotInputInput!]!

required

Show BarChartPlotInputInput fields

String!

required

The title of the plot.

Int!

required

The position of the plot on the widget.Used to maintain order of widget and edit panel

The external model ID on the plot. Required for project-backed plots; omit when datasetSource is set.

[ID!]!

required

The model version IDs on the plot. If empty, use all versions.

ModelEnvironmentName

The model environment ID on the plot. One of: production, validation, training, tracing.

[ID!]

The batch IDs on the plot. If empty, use all batches

DimensionCategory!

required

The metric function of the plot. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

DataQualityMetric

The data quality metric of the plot, ex. count, avg, etc. One of: avg, count, sum, percentEmpty, cardinality, standardDeviation, p50, p95, p99, p99_9, newValues, missingValues, averageStringListLength.

DimensionInput

The dimension of the plot, should be non-null when dimensionCategory = FEATURE_LABEL

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

String

The class of the classification model (categorical prediction values) on the widget.Provided only if model is categorical and evaluation type is specified

PositiveInt

The @K value for the performance metric. Model must be ranking.

[BarChartFilterItemInputInput!]!

required

The filters on this plot.

Show BarChartFilterItemInputInput fields

FilterRowType!

required

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age. Nested DimensionInput (same shape as above).

[DimensionValueInput!]

The dimension values of the filter. Nested DimensionValueInput (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

[String!]

A list of colors to use when plotting. Use one or two colors, two will be a gradient.

String

The dimension category to stack bars by (e.g., ‘featureLabel’, ‘tag’).

String

The dimension name to stack bars by (e.g., ‘_grade’).

WidgetDatasetSourceInput

Dataset or experiment source for this plot. Required for dataset-backed plots; omit when modelId is set.

Show WidgetDatasetSourceInput fields

ID!

required

The dataset backing this series.

Optional dataset version. For dataset-only series, pins the version to read. When experimentId is set, the experiment’s version is used instead.

Optional experiment on the dataset. When set, the series reads experiment run data.

[DatasetSourceFieldInput!]!

required

Field descriptors for this series. Distribution bar charts use exactly one entry. Nested DatasetSourceFieldInput (same shape as above).

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

BarChartSortOrder

The sort order of the bars One of: label_asc, label_desc, vol_asc, vol_desc.

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

String

The timezone of the widget

DateTime

The start time of the widget

DateTime

The end time of the widget

Boolean

Whether to overwrite the global time

DataGranularity

The granularity of the time series data One of: oneMinute, fiveMinute, hour, day, week, month.

String

The time range key of the widget

Int

The top N bars to display

Boolean!

required

Whether to render the widget as a stacked bar chart

String

Returns

Show BarChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateBarChartWidget($input: CreateBarChartWidgetMutationInput!) {
  createBarChartWidget(input: $input) {
    barChartWidget { id }
  }
}

{
  "input": {
    "title": "<string>",
    "dashboardId": "<ID>",
    "plots": [
      {
        "title": "<string>",
        "position": 1,
        "modelVersionIds": [
          "<ID>"
        ],
        "dimensionCategory": "featureLabel",
        "filters": [
          {
            "filterType": "featureLabel",
            "operator": "greaterThan"
          }
        ]
      }
    ],
    "isStacked": true
  }
}

updateBarChartWidget

updateBarChartWidget(input: UpdateBarChartWidgetMutationInput!): UpdateBarChartWidgetMutationPayload

Arguments

UpdateBarChartWidgetMutationInput!

required

Show UpdateBarChartWidgetMutationInput fields

ID!

required

ID!

required

BarChartWidgetUpdateDataInput!

required

Show BarChartWidgetUpdateDataInput fields

String!

required

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

[Int!]!

required

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

String

The label for the y-axis

BarChartSortOrder

The sort order for the bar chart widget One of: label_asc, label_desc, vol_asc, vol_desc.

Float

The top N bars to display

Boolean!

required

Boolean

If true, the bar chart is rendered as a stacked bar chart.

PerformanceMetric

CustomBinOption

The distribution bin option of the widget, if specified One of: numBins, customBins, medianCentered, discreteBins, decileBins, quantiles, discreteTopN.

Int

The number of bins on the distribution.

[Float!]

The bin edges on the distribution.

[Float!]

The array of percentiles of the distribution.

[BarChartPlotInputInput!]!

required

Show BarChartPlotInputInput fields

String!

required

The title of the plot.

Int!

required

The position of the plot on the widget.Used to maintain order of widget and edit panel

The external model ID on the plot. Required for project-backed plots; omit when datasetSource is set.

[ID!]!

required

The model version IDs on the plot. If empty, use all versions.

ModelEnvironmentName

The model environment ID on the plot. One of: production, validation, training, tracing.

[ID!]

The batch IDs on the plot. If empty, use all batches

DimensionCategory!

required

DataQualityMetric

DimensionInput

The dimension of the plot, should be non-null when dimensionCategory = FEATURE_LABEL Nested DimensionInput (same shape as above).

String

The class of the classification model (categorical prediction values) on the widget.Provided only if model is categorical and evaluation type is specified

PositiveInt

The @K value for the performance metric. Model must be ranking.

[BarChartFilterItemInputInput!]!

required

The filters on this plot. Nested BarChartFilterItemInputInput (same shape as above).

[String!]

A list of colors to use when plotting. Use one or two colors, two will be a gradient.

String

The dimension category to stack bars by (e.g., ‘featureLabel’, ‘tag’).

String

The dimension name to stack bars by (e.g., ‘_grade’).

WidgetDatasetSourceInput

Dataset or experiment source for this plot. Required for dataset-backed plots; omit when modelId is set. Nested WidgetDatasetSourceInput (same shape as above).

DataGranularity

The granularity of the time series data One of: oneMinute, fiveMinute, hour, day, week, month.

Boolean

Whether to overwrite the global time

String

The timezone of the widget

DateTime

The start time of the widget

DateTime

The end time of the widget

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

String

The time range key of the widget

String

Returns

Show BarChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateBarChartWidget($input: UpdateBarChartWidgetMutationInput!) {
  updateBarChartWidget(input: $input) {
    barChartWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "creationStatus": "needsInit",
      "gridPosition": [
        1
      ],
      "isNormalized": true,
      "plots": [
        {
          "title": "<string>",
          "position": 1,
          "modelVersionIds": [
            "<ID>"
          ],
          "dimensionCategory": "featureLabel",
          "filters": [
            {
              "filterType": "featureLabel",
              "operator": "greaterThan"
            }
          ]
        }
      ]
    }
  }
}

deleteBarChartWidget

deleteBarChartWidget(input: DeleteBarChartWidgetMutationInput!): DeleteBarChartWidgetMutationPayload

Arguments

DeleteBarChartWidgetMutationInput!

required

Show DeleteBarChartWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeleteBarChartWidget($input: DeleteBarChartWidgetMutationInput!) {
  deleteBarChartWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "barChartWidgetId": "<ID>"
  }
}

createStatisticWidget

createStatisticWidget(input: CreateStatisticWidgetMutationInput!): CreateStatisticWidgetMutationPayload

Arguments

CreateStatisticWidgetMutationInput!

required

Show CreateStatisticWidgetMutationInput fields

ID!

required

The dashboard of the widget.

String!

required

The title on the widget.

[Int!]

The grid position of the widget.

TimeSeriesMetricCategory!

required

The type of timeseries metric on the widget. One of: modelDataMetric, evaluationMetric.

WidgetCreationStatus

One of: needsInit, pending, created, published, unpublished.

DimensionInput

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

ID!

required

The model ID on the widget.

ModelEnvironmentName

The model environment name on the widget. One of: production, validation, training, tracing.

[ID!]

The environment batch IDs on the widget. If empty, use all batches

[ID!]

The model version IDs on the widget. If empty, use all versions.

DimensionCategory

The dimension category of the widget. ex. featureLabel, prediction, actuals. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

PerformanceMetric

The performance metric function of the widget. One of: accuracy, precision, recall, f_1, sensitivity, specificity, falsePositiveRate, falseNegativeRate, falseNegativeDensity, mse, rmse, mae, mape, predictionAverage, actualsAverage, auc, logLoss, rSquared, prAuc, meanError, calibration, ndcg, recallParity, falsePositiveRateParity, disparateImpact, wape, smape, mase, udf, groupAuc, mrr, rankingMap, rankingRecall, rankingPrecision, microAveragedPrecision, microAveragedRecall, multiClassPrecision, multiClassRecall, macroAveragedPrecision, macroAveragedRecall.

DataQualityMetric

The data quality metric of the widget, ex. count, avg, etc. One of: avg, count, sum, percentEmpty, cardinality, standardDeviation, p50, p95, p99, p99_9, newValues, missingValues, averageStringListLength.

String

The class of the classification model (categorical prediction values) on the widget.Provided only if timeseriesMetricType = ‘evaluationMetrics’ and prediction value type is ‘categorical’.

PositiveInt

The @K value for the performance metric. Model must be ranking.

[StatisticWidgetFilterItemInput!]

The filters on this plot.

Show StatisticWidgetFilterItemInput fields

FilterRowType!

required

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age.

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

[DimensionValueInput!]

The dimension values of the filter.

Show DimensionValueInput fields

ID!

required

String!

required

The value of this dimension value, e.g. ‘Wells Fargo’ or ‘60’

DimensionValueSimilarityReference

The similarity references for this dimension value Nested DimensionValueSimilarityReference (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

CustomMetricInput

The custom metric function of the widget, if specified.

Show CustomMetricInput fields

ID!

required

String!

required

The name of the custom metric

String

The description of the custom metric

String!

required

The custom metric, ex. ‘AVG(predictionScore)’, etc.

Boolean!

required

Whether the custom metric requires a positive class to calculate

String

The text of the metric on the widget.

String

Returns

Show StatisticWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateStatisticWidget($input: CreateStatisticWidgetMutationInput!) {
  createStatisticWidget(input: $input) {
    statisticWidget { id }
  }
}

{
  "input": {
    "dashboardId": "<ID>",
    "title": "<string>",
    "timeSeriesMetricType": "modelDataMetric",
    "modelId": "<ID>"
  }
}

updateStatisticWidget

Update a single widget updateStatisticWidget(input: UpdateStatisticWidgetMutationInput!): UpdateStatisticWidgetMutationPayload

Arguments

UpdateStatisticWidgetMutationInput!

required

Show UpdateStatisticWidgetMutationInput fields

ID!

required

ID!

required

The dashboard of the widget.

StatisticWidgetUpdateDataInput!

required

Show StatisticWidgetUpdateDataInput fields

String!

required

The title on the widget.

[Int!]!

required

The grid position of the widget, defaults to next available slot if not specified

ID!

required

The model ID on the widget.

[ID!]!

required

The model version IDs on the widget. If empty, use all versions.

ModelEnvironmentName

The model environment name of the widget. Defaulted to production. One of: production, validation, training, tracing.

[ID!]!

required

The environment batch id of the widget.

DimensionCategory

The dimension category of the widget. ex. featureLabel, prediction, actuals One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

PerformanceMetric

DataQualityMetric

The aggregation type of the widget, ex. count, avg, etc. One of: avg, count, sum, percentEmpty, cardinality, standardDeviation, p50, p95, p99, p99_9, newValues, missingValues, averageStringListLength.

String

The class of the classification model (categorical prediction values) on the widget. Provided only if timeseriesMetricType = ‘evaluationMetrics’ and prediction value type is categorical.

TimeSeriesMetricCategory!

required

The type of timeseries metric on the widget. One of: modelDataMetric, evaluationMetric.

DimensionInput

The dimension of the widget, non-null if timeSeriesMetricType

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

WidgetCreationStatus!

required

The status of the widget, to support various dashboard edit states. One of: needsInit, pending, created, published, unpublished.

[StatisticWidgetFilterItemInput!]!

required

The filters on this widget.

Show StatisticWidgetFilterItemInput fields

FilterRowType!

required

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age. Nested DimensionInput (same shape as above).

[DimensionValueInput!]

The dimension values of the filter. Nested DimensionValueInput (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

CustomMetricInput

The custom metric function of the widget, if specified.

Show CustomMetricInput fields

ID!

required

String!

required

The name of the custom metric

String

The description of the custom metric

String!

required

The custom metric, ex. ‘AVG(predictionScore)’, etc.

Boolean!

required

Whether the custom metric requires a positive class to calculate

PositiveInt

The @K value for the performance metric. Model must be ranking.

String

The text of the metric on the widget.

String

Returns

Show StatisticWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateStatisticWidget($input: UpdateStatisticWidgetMutationInput!) {
  updateStatisticWidget(input: $input) {
    statisticWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "gridPosition": [
        1
      ],
      "modelId": "<ID>",
      "modelVersionIds": [
        "<ID>"
      ],
      "modelVersionEnvironmentMetadataIds": [
        "<ID>"
      ],
      "timeSeriesMetricType": "modelDataMetric",
      "creationStatus": "needsInit",
      "filters": [
        {
          "filterType": "featureLabel",
          "operator": "greaterThan"
        }
      ]
    }
  }
}

deleteStatisticWidget

deleteStatisticWidget(input: DeleteStatisticWidgetMutationInput!): DeleteStatisticWidgetMutationPayload

Arguments

DeleteStatisticWidgetMutationInput!

required

Show DeleteStatisticWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeleteStatisticWidget($input: DeleteStatisticWidgetMutationInput!) {
  deleteStatisticWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "statWidgetId": "<ID>"
  }
}

createTextWidget

createTextWidget(input: CreateTextWidgetMutationInput!): CreateTextWidgetMutationPayload

Arguments

CreateTextWidgetMutationInput!

required

Show CreateTextWidgetMutationInput fields

String!

required

The title on the widget.

String!

required

The text content on the widget.

[Int!]!

required

The grid position of the widget.

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

ID!

required

The dashboard of the widget.

String

Returns

Show TextWidget fields

Scalar fields: id, title, content, gridPosition, creationStatus, dashboardId.

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateTextWidget($input: CreateTextWidgetMutationInput!) {
  createTextWidget(input: $input) {
    textWidget { id }
  }
}

{
  "input": {
    "title": "<string>",
    "content": "<string>",
    "gridPosition": [
      1
    ],
    "creationStatus": "needsInit",
    "dashboardId": "<ID>"
  }
}

updateTextWidget

Update a single widget updateTextWidget(input: UpdateTextWidgetMutationInput!): UpdateTextWidgetMutationPayload

Arguments

UpdateTextWidgetMutationInput!

required

Show UpdateTextWidgetMutationInput fields

ID!

required

ID!

required

The dashboard of the widget.

TextWidgetUpdateDataInput!

required

Show TextWidgetUpdateDataInput fields

String!

required

The title on the widget.

String!

required

The text content on the widget.

[Int!]!

required

The grid position of the widget.

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

String

Returns

Show TextWidget fields

Scalar fields: id, title, content, gridPosition, creationStatus, dashboardId.

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateTextWidget($input: UpdateTextWidgetMutationInput!) {
  updateTextWidget(input: $input) {
    textWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "content": "<string>",
      "gridPosition": [
        1
      ],
      "creationStatus": "needsInit"
    }
  }
}

deleteTextWidget

deleteTextWidget(input: DeleteTextWidgetMutationInput!): DeleteTextWidgetMutationPayload

Arguments

DeleteTextWidgetMutationInput!

required

Show DeleteTextWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeleteTextWidget($input: DeleteTextWidgetMutationInput!) {
  deleteTextWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "textWidgetId": "<ID>"
  }
}

createPivotTableWidget

createPivotTableWidget(input: CreatePivotTableWidgetMutationInput!): CreatePivotTableWidgetMutationPayload

Arguments

CreatePivotTableWidgetMutationInput!

required

Show CreatePivotTableWidgetMutationInput fields

ID!

required

The dashboard of the widget.

String!

required

The title on the widget.

[Int!]

The grid position of the widget.

WidgetCreationStatus

One of: needsInit, pending, created, published, unpublished.

ID!

required

The model ID on the widget.

ModelEnvironmentName

The model environment name on the widget. One of: production, validation, training, tracing.

[ID!]

The environment batch IDs on the widget. If empty, use all batches

[ID!]

The model version IDs on the widget. If empty, use all versions.

String!

required

The categorical dimension to group by (e.g., model name).

String

Optional custom title for the grouped dimension column.

DimensionCategory

The dimension category for the grouped dimension. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

[PivotTableColumnInput!]

The columns (numeric dimensions with aggregations) on the pivot table.

Show PivotTableColumnInput fields

String!

required

The dimension to aggregate. Numeric dimensions support sum/avg; all dimensions support count/cardinality.

PivotTableAggregation!

required

The aggregation type (avg, sum, count, cardinality). One of: avg, sum, count, cardinality.

Int!

required

The display order of the column.

String

Optional custom title for the column.

DimensionCategory

The dimension category for this column. One of: featureLabel, prediction, actuals, actualScore, actualClass, predictionClass, predictionScore, tag, spanProperty, llmEval, annotation, userAnnotation, modelVersion, batchId.

[PivotTableWidgetFilterItemInput!]

The filters on this widget.

Show PivotTableWidgetFilterItemInput fields

FilterRowType!

required

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age.

Show DimensionInput fields

ID!

required

String!

required

The label of this dimension, e.g. ‘Bank Name’ or ‘age’

DimensionDataType!

required

The data type of the values in dimensionName, ex. ‘FLOAT’, ‘STRING’, etc. One of: STRING, LONG, FLOAT, DOUBLE, EMBEDDING, STRING_LIST, DICTIONARY.

DimensionCategory

[DimensionValueInput!]

The dimension values of the filter.

Show DimensionValueInput fields

ID!

required

String!

required

The value of this dimension value, e.g. ‘Wells Fargo’ or ‘60’

DimensionValueSimilarityReference

The similarity references for this dimension value Nested DimensionValueSimilarityReference (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

Boolean

Whether to override the dashboard time range.

String

Timezone for time range.

DateTime

Custom start time.

DateTime

Custom end time.

String

Preset time range key.

String

Query filter string.

String

Returns

Show PivotTableWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreatePivotTableWidget($input: CreatePivotTableWidgetMutationInput!) {
  createPivotTableWidget(input: $input) {
    pivotTableWidget { id }
  }
}

{
  "input": {
    "dashboardId": "<ID>",
    "title": "<string>",
    "modelId": "<ID>",
    "groupedDimension": "<string>"
  }
}

updatePivotTableWidget

Update a single pivot table widget updatePivotTableWidget(input: UpdatePivotTableWidgetMutationInput!): UpdatePivotTableWidgetMutationPayload

Arguments

UpdatePivotTableWidgetMutationInput!

required

Show UpdatePivotTableWidgetMutationInput fields

ID!

required

ID!

required

The dashboard of the widget.

PivotTableWidgetUpdateDataInput!

required

Show PivotTableWidgetUpdateDataInput fields

String!

required

The title on the widget.

[Int!]!

required

The grid position of the widget.

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

ID!

required

The model ID on the widget.

ModelEnvironmentName

The model environment name on the widget. One of: production, validation, training, tracing.

[ID!]!

required

The environment batch IDs on the widget. If empty, use all batches

[ID!]!

required

The model version IDs on the widget. If empty, use all versions.

String!

required

The categorical dimension to group by (e.g., model name).

String

Optional custom title for the grouped dimension column.

DimensionCategory

[PivotTableColumnInput!]!

required

The columns (numeric dimensions with aggregations) on the pivot table.

Show PivotTableColumnInput fields

String!

required

The dimension to aggregate. Numeric dimensions support sum/avg; all dimensions support count/cardinality.

PivotTableAggregation!

required

The aggregation type (avg, sum, count, cardinality). One of: avg, sum, count, cardinality.

Int!

required

The display order of the column.

String

Optional custom title for the column.

DimensionCategory

[PivotTableWidgetFilterItemInput!]!

required

The filters on this widget.

Show PivotTableWidgetFilterItemInput fields

FilterRowType!

required

ComparisonOperator!

required

The operator of the filter. One of: greaterThan, lessThan, equals, notEquals, greaterThanOrEqual, lessThanOrEqual, topN, contains, containsString, similarTo.

DimensionInput

The model dimension to filter, e.g. Age. Nested DimensionInput (same shape as above).

[DimensionValueInput!]

The dimension values of the filter. Nested DimensionValueInput (same shape as above).

[String!]

The binary values of the prediction.

[Float!]

The numeric values of the prediction.

[String!]

The categorical values of the prediction.

Boolean

Whether to override the dashboard time range.

String

Timezone for time range.

DateTime

Custom start time.

DateTime

Custom end time.

String

Preset time range key.

String

Query filter string.

String

Returns

Show PivotTableWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdatePivotTableWidget($input: UpdatePivotTableWidgetMutationInput!) {
  updatePivotTableWidget(input: $input) {
    pivotTableWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "gridPosition": [
        1
      ],
      "creationStatus": "needsInit",
      "modelId": "<ID>",
      "modelVersionEnvironmentMetadataIds": [
        "<ID>"
      ],
      "modelVersionIds": [
        "<ID>"
      ],
      "groupedDimension": "<string>",
      "columns": [
        {
          "dimension": "<string>",
          "aggregation": "avg",
          "position": 1
        }
      ],
      "filters": [
        {
          "filterType": "featureLabel",
          "operator": "greaterThan"
        }
      ]
    }
  }
}

deletePivotTableWidget

deletePivotTableWidget(input: DeletePivotTableWidgetMutationInput!): DeletePivotTableWidgetMutationPayload

Arguments

DeletePivotTableWidgetMutationInput!

required

Show DeletePivotTableWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeletePivotTableWidget($input: DeletePivotTableWidgetMutationInput!) {
  deletePivotTableWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "pivotTableWidgetId": "<ID>"
  }
}

createExperimentChartWidget

Create an experiment chart widget

createExperimentChartWidget(input: CreateExperimentChartWidgetMutationInput!): CreateExperimentChartWidgetMutationPayload

Arguments

CreateExperimentChartWidgetMutationInput!

required

Show CreateExperimentChartWidgetMutationInput fields

String!

required

WidgetCreationStatus

One of: needsInit, pending, created, published, unpublished.

ID!

required

[Int!]

The grid position the widget should be placed at.

[ExperimentChartPlotInputInput!]

Show ExperimentChartPlotInputInput fields

String!

required

The title on the plot.

Int!

required

The position of the plot.

ID!

required

The dataset of the plot.

String!

required

The evaluation metric of the dataset.

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

String

Returns

Show ExperimentChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation CreateExperimentChartWidget($input: CreateExperimentChartWidgetMutationInput!) {
  createExperimentChartWidget(input: $input) {
    experimentChartWidget { id }
  }
}

{
  "input": {
    "title": "<string>",
    "dashboardId": "<ID>"
  }
}

updateExperimentChartWidget

updateExperimentChartWidget(input: UpdateExperimentChartWidgetMutationInput!): UpdateExperimentChartWidgetMutationPayload

Arguments

UpdateExperimentChartWidgetMutationInput!

required

Show UpdateExperimentChartWidgetMutationInput fields

ID!

required

ID!

required

ExperimentChartWidgetUpdateDataInput!

required

Show ExperimentChartWidgetUpdateDataInput fields

String!

required

WidgetCreationStatus!

required

One of: needsInit, pending, created, published, unpublished.

[Int!]!

required

[ExperimentChartPlotInputInput!]!

required

Show ExperimentChartPlotInputInput fields

String!

required

The title on the plot.

Int!

required

The position of the plot.

ID!

required

The dataset of the plot.

String!

required

The evaluation metric of the dataset.

String

The font size of the widget

Boolean

Whether to show the legend on the widget

String

The placement of the legend on the widget

String

The label for the x-axis

String

The tilt of the x-axis

Float

The minimum domain on the y-axis

Float

The maximum domain on the y-axis

String

Returns

Show ExperimentChartWidget fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation UpdateExperimentChartWidget($input: UpdateExperimentChartWidgetMutationInput!) {
  updateExperimentChartWidget(input: $input) {
    experimentChartWidget { id }
  }
}

{
  "input": {
    "id": "<ID>",
    "dashboardId": "<ID>",
    "updateData": {
      "title": "<string>",
      "creationStatus": "needsInit",
      "gridPosition": [
        1
      ],
      "plots": [
        {
          "title": "<string>",
          "position": 1,
          "datasetId": "<ID>",
          "evaluationMetric": "<string>"
        }
      ]
    }
  }
}

deleteExperimentChartWidget

deleteExperimentChartWidget(input: DeleteExperimentChartWidgetMutationInput!): DeleteExperimentChartWidgetMutationPayload

Arguments

DeleteExperimentChartWidgetMutationInput!

required

Show DeleteExperimentChartWidgetMutationInput fields

String

Returns

Dashboard

Show Dashboard fields

Example

Only required input fields are shown. Replace <ID> and <string> placeholders with real values; the object graph page shows how to look IDs up.

mutation DeleteExperimentChartWidget($input: DeleteExperimentChartWidgetMutationInput!) {
  deleteExperimentChartWidget(input: $input) {
    dashboard { id name status }
  }
}

{
  "input": {
    "experimentChartWidgetId": "<ID>"
  }
}

​Mutations in this page

​Reference

​createDashboard

​Arguments

​Returns

​Example

​createDefaultDashboard

​Arguments

​Returns

​Example

​createDashboardFromTemplate

​Arguments

​Returns

​Example

​copyDashboard

​Arguments

​Returns

​Example

​updateDashboardName

​Arguments

​Returns

​Example

​updateDashboardStatus

​Arguments

​Returns

​Example

​duplicateWidget

​Arguments

​Returns

​Example

​createLineChartWidget

​Arguments

​Returns

​Example

​updateLineChartWidget

​Arguments

​Returns

​Example

​deleteLineChartWidget

​Arguments

​Returns

​Example

​createBarChartWidget

​Arguments

​Returns

​Example

​updateBarChartWidget

​Arguments

​Returns

​Example

​deleteBarChartWidget

​Arguments

​Returns

​Example

​createStatisticWidget

​Arguments

​Returns

​Example

​updateStatisticWidget

​Arguments

​Returns

​Example

​deleteStatisticWidget

​Arguments

​Returns

​Example

​createTextWidget

​Arguments

​Returns

​Example

​updateTextWidget

​Arguments

​Returns

​Example

​deleteTextWidget

​Arguments

​Returns

​Example

​createPivotTableWidget

​Arguments

Mutations in this page

Reference

createDashboard

Arguments

Returns

Example

createDefaultDashboard

Arguments

Returns

Example

createDashboardFromTemplate

Arguments

Returns

Example

copyDashboard

Arguments

Returns

Example

updateDashboardName

Arguments

Returns

Example

updateDashboardStatus

Arguments

Returns

Example

duplicateWidget

Arguments

Returns

Example

createLineChartWidget

Arguments

Returns

Example

updateLineChartWidget

Arguments

Returns

Example

deleteLineChartWidget

Arguments

Returns

Example

createBarChartWidget

Arguments

Returns

Example

updateBarChartWidget

Arguments

Returns

Example

deleteBarChartWidget

Arguments

Returns

Example

createStatisticWidget

Arguments

Returns

Example

updateStatisticWidget

Arguments

Returns

Example

deleteStatisticWidget

Arguments

Returns

Example

createTextWidget

Arguments

Returns

Example

updateTextWidget

Arguments

Returns

Example

deleteTextWidget

Arguments

Returns

Example

createPivotTableWidget

Arguments