github-mirrors/kibana
1
0
Fork 0
mirror of https://github.com/elastic/kibana.git synced 2025-04-23 17:28:26 -04:00
Code Issues Projects Releases Packages Wiki Activity

handle runtime fields in validation step (#96340)

Browse source
This commit is contained in:
Melissa Alvarez 2021-04-07 15:43:06 -04:00 • committed by GitHub
parent ad06d16beb
commit 71c326c8bf
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 22 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

3
x-pack/plugins/ml/common/types/data_frame_analytics.ts
View file

@ -6,6 +6,7 @@
*/
import Boom from '@hapi/boom';
import type { estypes } from '@elastic/elasticsearch';
import { RuntimeMappings } from './fields';
import { EsErrorBody } from '../util/errors';
@ -75,7 +76,7 @@ export interface DataFrameAnalyticsConfig {
};
source: {
index: IndexName | IndexName[];
query?: any;
query?: estypes.QueryContainer;
runtime_mappings?: RuntimeMappings;
};
analysis: AnalysisConfig;

38
x-pack/plugins/ml/server/models/data_frame_analytics/validation.ts
View file

@ -195,12 +195,13 @@ function getTrainingPercentMessage(trainingDocs: number) {
async function getValidationCheckMessages(
asCurrentUser: IScopedClusterClient['asCurrentUser'],
analyzedFields: string[],
index: string | string[],
analysisConfig: AnalysisConfig,
query: estypes.QueryContainer = defaultQuery
source: DataFrameAnalyticsConfig['source']
) {
const analysisType = getAnalysisType(analysisConfig);
const depVar = getDependentVar(analysisConfig);
const index = source.index;
const query = source.query || defaultQuery;
const messages = [];
const emptyFields: string[] = [];
const percentEmptyLimit = FRACTION_EMPTY_LIMIT * 100;
@ -236,6 +237,7 @@ async function getValidationCheckMessages(
size: 0,
track_total_hits: true,
body: {
...(source.runtime_mappings ? { runtime_mappings: source.runtime_mappings } : {}),
query,
aggs,
},
@ -247,21 +249,22 @@ async function getValidationCheckMessages(
if (body.aggregations) {
// @ts-expect-error
Object.entries(body.aggregations).forEach(([aggName, { doc_count: docCount, value }]) => {
const empty = docCount / totalDocs;
if (docCount !== undefined) {
const empty = docCount / totalDocs;
if (docCount > 0 && empty > FRACTION_EMPTY_LIMIT) {
emptyFields.push(aggName);
if (docCount > 0 && empty > FRACTION_EMPTY_LIMIT) {
emptyFields.push(aggName);
if (aggName === depVar) {
depVarValid = false;
dependentVarWarningMessage.text = i18n.translate(
'xpack.ml.models.dfaValidation.messages.depVarEmptyWarning',
{
defaultMessage:
'The dependent variable has at least {percentEmpty}% empty values. It may be unsuitable for analysis.',
values: { percentEmpty: percentEmptyLimit },
}
);
if (aggName === depVar) {
depVarValid = false;
dependentVarWarningMessage.text = i18n.translate(
'xpack.ml.models.dfaValidation.messages.depVarEmptyWarning',
{
defaultMessage:
'The dependent variable has at least {percentEmpty}% empty values. It may be unsuitable for analysis.',
values: { percentEmpty: percentEmptyLimit },
}
);
}
}
}
@ -374,9 +377,8 @@ export async function validateAnalyticsJob(
const messages = await getValidationCheckMessages(
client.asCurrentUser,
job.analyzed_fields.includes,
job.source.index,
job.analysis,
job.source.query
job.source
);
return messages;
}