[ML] API Integration tests: adds test for Data Frame Analytics evalua…

…te endpoint (elastic#97856) * wip: add api test for evaluate endpoint * Add api test for evaluate endpoint * add tests for view only and unauthorized user Co-authored-by: Kibana Machine <[email protected]>
alvarezmelissa87 · Apr 23, 2021 · 755b529 · 755b529
1 parent a36606d
commit 755b529
Show file tree

Hide file tree

Showing 2 changed files with 189 additions and 0 deletions.
diff --git a/x-pack/test/api_integration/apis/ml/data_frame_analytics/evaluate.ts b/x-pack/test/api_integration/apis/ml/data_frame_analytics/evaluate.ts
@@ -0,0 +1,188 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0; you may not use this file except in compliance with the Elastic License
+ * 2.0.
+ */
+
+import expect from '@kbn/expect';
+import { FtrProviderContext } from '../../../ftr_provider_context';
+import { USER } from '../../../../functional/services/ml/security_common';
+import { DataFrameAnalyticsConfig } from '../../../../../plugins/ml/public/application/data_frame_analytics/common';
+import { DeepPartial } from '../../../../../plugins/ml/common/types/common';
+import { COMMON_REQUEST_HEADERS } from '../../../../functional/services/ml/common_api';
+
+export default ({ getService }: FtrProviderContext) => {
+  const esArchiver = getService('esArchiver');
+  const supertest = getService('supertestWithoutAuth');
+  const ml = getService('ml');
+
+  const currentTime = `${Date.now()}`;
+  const generateDestinationIndex = (analyticsId: string) => `user-${analyticsId}`;
+  const jobEval: any = {
+    regression: {
+      index: generateDestinationIndex(`regression_${currentTime}`),
+      evaluation: {
+        regression: {
+          actual_field: 'stab',
+          predicted_field: 'ml.stab_prediction',
+          metrics: {
+            r_squared: {},
+            mse: {},
+            msle: {},
+            huber: {},
+          },
+        },
+      },
+    },
+    classification: {
+      index: generateDestinationIndex(`classification_${currentTime}`),
+      evaluation: {
+        classification: {
+          actual_field: 'y',
+          predicted_field: 'ml.y_prediction',
+          metrics: { multiclass_confusion_matrix: {}, accuracy: {}, recall: {} },
+        },
+      },
+    },
+  };
+  const jobAnalysis: any = {
+    classification: {
+      source: {
+        index: ['ft_bank_marketing'],
+        query: {
+          match_all: {},
+        },
+      },
+      analysis: {
+        classification: {
+          dependent_variable: 'y',
+          training_percent: 20,
+        },
+      },
+    },
+    regression: {
+      source: {
+        index: ['ft_egs_regression'],
+        query: {
+          match_all: {},
+        },
+      },
+      analysis: {
+        regression: {
+          dependent_variable: 'stab',
+          training_percent: 20,
+        },
+      },
+    },
+  };
+
+  interface TestConfig {
+    jobType: string;
+    config: DeepPartial<DataFrameAnalyticsConfig>;
+    eval: any;
+  }
+
+  const testJobConfigs: TestConfig[] = ['regression', 'classification'].map((jobType, idx) => {
+    const analyticsId = `${jobType}_${currentTime}`;
+    return {
+      jobType,
+      config: {
+        id: analyticsId,
+        description: `Testing ${jobType} evaluation`,
+        dest: {
+          index: generateDestinationIndex(analyticsId),
+          results_field: 'ml',
+        },
+        analyzed_fields: {
+          includes: [],
+          excludes: [],
+        },
+        model_memory_limit: '60mb',
+        ...jobAnalysis[jobType],
+      },
+      eval: jobEval[jobType],
+    };
+  });
+
+  async function createJobs(mockJobConfigs: TestConfig[]) {
+    for (const jobConfig of mockJobConfigs) {
+      await ml.api.createAndRunDFAJob(jobConfig.config as DataFrameAnalyticsConfig);
+    }
+  }
+
+  describe('POST data_frame/_evaluate', () => {
+    before(async () => {
+      await esArchiver.loadIfNeeded('ml/bm_classification');
+      await esArchiver.loadIfNeeded('ml/egs_regression');
+      await ml.testResources.setKibanaTimeZoneToUTC();
+      await createJobs(testJobConfigs);
+    });
+
+    after(async () => {
+      await ml.api.cleanMlIndices();
+    });
+
+    testJobConfigs.forEach((testConfig) => {
+      describe(`EvaluateDataFrameAnalytics ${testConfig.jobType}`, async () => {
+        it(`should evaluate ${testConfig.jobType} analytics job`, async () => {
+          const { body } = await supertest
+            .post(`/api/ml/data_frame/_evaluate`)
+            .auth(USER.ML_POWERUSER, ml.securityCommon.getPasswordForUser(USER.ML_POWERUSER))
+            .set(COMMON_REQUEST_HEADERS)
+            .send(testConfig.eval)
+            .expect(200);
+
+          if (testConfig.jobType === 'classification') {
+            const { classification } = body;
+            expect(body).to.have.property('classification');
+            expect(classification).to.have.property('recall');
+            expect(classification).to.have.property('accuracy');
+            expect(classification).to.have.property('multiclass_confusion_matrix');
+          } else {
+            const { regression } = body;
+            expect(body).to.have.property('regression');
+            expect(regression).to.have.property('mse');
+            expect(regression).to.have.property('msle');
+            expect(regression).to.have.property('r_squared');
+          }
+        });
+
+        it(`should evaluate ${testConfig.jobType} job for the user with only view permission`, async () => {
+          const { body } = await supertest
+            .post(`/api/ml/data_frame/_evaluate`)
+            .auth(USER.ML_VIEWER, ml.securityCommon.getPasswordForUser(USER.ML_VIEWER))
+            .set(COMMON_REQUEST_HEADERS)
+            .send(testConfig.eval)
+            .expect(200);
+
+          if (testConfig.jobType === 'classification') {
+            const { classification } = body;
+            expect(body).to.have.property('classification');
+            expect(classification).to.have.property('recall');
+            expect(classification).to.have.property('accuracy');
+            expect(classification).to.have.property('multiclass_confusion_matrix');
+          } else {
+            const { regression } = body;
+            expect(body).to.have.property('regression');
+            expect(regression).to.have.property('mse');
+            expect(regression).to.have.property('msle');
+            expect(regression).to.have.property('r_squared');
+          }
+        });
+
+        it(`should not allow unauthorized user to evaluate ${testConfig.jobType} job`, async () => {
+          const { body } = await supertest
+            .post(`/api/ml/data_frame/_evaluate`)
+            .auth(USER.ML_UNAUTHORIZED, ml.securityCommon.getPasswordForUser(USER.ML_UNAUTHORIZED))
+            .set(COMMON_REQUEST_HEADERS)
+            .send(testConfig.eval)
+            .expect(403);
+
+          expect(body.error).to.eql('Forbidden');
+          expect(body.message).to.eql('Forbidden');
+        });
+      });
+    });
+  });
+};
diff --git a/x-pack/test/api_integration/apis/ml/data_frame_analytics/index.ts b/x-pack/test/api_integration/apis/ml/data_frame_analytics/index.ts
@@ -20,6 +20,7 @@ export default function ({ loadTestFile }: FtrProviderContext) {
     loadTestFile(require.resolve('./get_spaces'));
     loadTestFile(require.resolve('./update_spaces'));
     loadTestFile(require.resolve('./delete_spaces'));
+    loadTestFile(require.resolve('./evaluate'));
     loadTestFile(require.resolve('./explain'));
   });
 }