[ML] DFAnalytics results: ensure ml result fields are shown in data g…

…rid (#68305)

* wip: ensure top classes and influencer result col show up correctly

* ensure ml subFields columns are populated

x-pack/plugins/ml/public/application/components/data_grid/common.ts

@@ -29,6 +29,7 @@ import {
   FEATURE_IMPORTANCE,
   FEATURE_INFLUENCE,
   OUTLIER_SCORE,
+  TOP_CLASSES,
 } from '../../data_frame_analytics/common/constants';
 import { formatHumanReadableDateTimeSeconds } from '../../util/date_utils';
 import { getNestedProperty } from '../../util/object_utils';
@@ -110,7 +111,10 @@ export const getDataGridSchemasFromFieldTypes = (fieldTypes: FieldTypes, results
       schema = 'numeric';
     }
 
-    if (field.includes(`${resultsField}.${FEATURE_IMPORTANCE}`)) {
+    if (
+      field.includes(`${resultsField}.${FEATURE_IMPORTANCE}`) ||
+      field.includes(`${resultsField}.${TOP_CLASSES}`)
+    ) {
       schema = 'json';
     }
 

x-pack/plugins/ml/public/application/data_frame_analytics/common/analytics.ts

@@ -228,6 +228,16 @@ export const getPredictionFieldName = (
   return predictionFieldName;
 };
 
+export const getNumTopClasses = (
+  analysis: AnalysisConfig
+): ClassificationAnalysis['classification']['num_top_classes'] => {
+  let numTopClasses;
+  if (isClassificationAnalysis(analysis) && analysis.classification.num_top_classes !== undefined) {
+    numTopClasses = analysis.classification.num_top_classes;
+  }
+  return numTopClasses;
+};
+
 export const getNumTopFeatureImportanceValues = (
   analysis: AnalysisConfig
 ):

x-pack/plugins/ml/public/application/data_frame_analytics/common/constants.ts

@@ -7,4 +7,5 @@
 export const DEFAULT_RESULTS_FIELD = 'ml';
 export const FEATURE_IMPORTANCE = 'feature_importance';
 export const FEATURE_INFLUENCE = 'feature_influence';
+export const TOP_CLASSES = 'top_classes';
 export const OUTLIER_SCORE = 'outlier_score';

x-pack/plugins/ml/public/application/data_frame_analytics/common/fields.ts

@@ -5,6 +5,7 @@
  */
 
 import {
+  getNumTopClasses,
   getNumTopFeatureImportanceValues,
   getPredictedFieldName,
   getDependentVar,
@@ -18,7 +19,7 @@ import { Field } from '../../../../common/types/fields';
 import { ES_FIELD_TYPES, KBN_FIELD_TYPES } from '../../../../../../../src/plugins/data/public';
 import { newJobCapsService } from '../../services/new_job_capabilities_service';
 
-import { FEATURE_IMPORTANCE, FEATURE_INFLUENCE, OUTLIER_SCORE } from './constants';
+import { FEATURE_IMPORTANCE, FEATURE_INFLUENCE, OUTLIER_SCORE, TOP_CLASSES } from './constants';
 
 export type EsId = string;
 export type EsDocSource = Record<string, any>;
@@ -177,6 +178,7 @@ export const getDefaultFieldsFromJobCaps = (
 
   const featureImportanceFields = [];
   const featureInfluenceFields = [];
+  const topClassesFields = [];
   const allFields: any = [];
   let type: ES_FIELD_TYPES | undefined;
   let predictedField: string | undefined;
@@ -207,20 +209,29 @@ export const getDefaultFieldsFromJobCaps = (
     type = newJobCapsService.getFieldById(dependentVariable)?.type;
     const predictionFieldName = getPredictionFieldName(jobConfig.analysis);
     const numTopFeatureImportanceValues = getNumTopFeatureImportanceValues(jobConfig.analysis);
+    const numTopClasses = getNumTopClasses(jobConfig.analysis);
 
     const defaultPredictionField = `${dependentVariable}_prediction`;
     predictedField = `${resultsField}.${
       predictionFieldName ? predictionFieldName : defaultPredictionField
     }`;
 
-    if ((numTopFeatureImportanceValues ?? 0) > 0 && needsDestIndexFields === true) {
+    if ((numTopFeatureImportanceValues ?? 0) > 0) {
       featureImportanceFields.push({
         id: `${resultsField}.${FEATURE_IMPORTANCE}`,
         name: `${resultsField}.${FEATURE_IMPORTANCE}`,
         type: KBN_FIELD_TYPES.UNKNOWN,
       });
     }
 
+    if ((numTopClasses ?? 0) > 0) {
+      topClassesFields.push({
+        id: `${resultsField}.${TOP_CLASSES}`,
+        name: `${resultsField}.${TOP_CLASSES}`,
+        type: KBN_FIELD_TYPES.UNKNOWN,
+      });
+    }
+
     // Only need to add these fields if we didn't use dest index pattern to get the fields
     if (needsDestIndexFields === true) {
       allFields.push(
@@ -234,7 +245,12 @@ export const getDefaultFieldsFromJobCaps = (
     }
   }
 
-  allFields.push(...fields, ...featureImportanceFields, ...featureInfluenceFields);
+  allFields.push(
+    ...fields,
+    ...featureImportanceFields,
+    ...featureInfluenceFields,
+    ...topClassesFields
+  );
   allFields.sort(({ name: a }: { name: string }, { name: b }: { name: string }) =>
     sortExplorationResultsFields(a, b, jobConfig)
   );

x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/exploration_results_table/use_exploration_results.ts

@@ -19,7 +19,11 @@ import {
 import { SavedSearchQuery } from '../../../../../contexts/ml';
 
 import { getIndexData, getIndexFields, DataFrameAnalyticsConfig } from '../../../../common';
-import { DEFAULT_RESULTS_FIELD, FEATURE_IMPORTANCE } from '../../../../common/constants';
+import {
+  DEFAULT_RESULTS_FIELD,
+  FEATURE_IMPORTANCE,
+  TOP_CLASSES,
+} from '../../../../common/constants';
 import { sortExplorationResultsFields, ML__ID_COPY } from '../../../../common/fields';
 
 export const useExplorationResults = (
@@ -47,8 +51,9 @@ export const useExplorationResults = (
     25,
     // reduce default selected rows from 20 to 8 for performance reasons.
     8,
-    // by default, hide feature-importance columns and the doc id copy
-    (d) => !d.includes(`.${FEATURE_IMPORTANCE}.`) && d !== ML__ID_COPY
+    // by default, hide feature-importance and top-classes columns and the doc id copy
+    (d) =>
+      !d.includes(`.${FEATURE_IMPORTANCE}.`) && !d.includes(`.${TOP_CLASSES}.`) && d !== ML__ID_COPY
   );
 
   useEffect(() => {