Merge pull request #2047 from Agenta-AI/feat/implement-human-evaluations

Feat: implemented human evaluations
Agenta-AI · Sep 3, 2024 · 60d237e · 60d237e
2 parents e91e37b + 538b852
commit 60d237e
Show file tree

Hide file tree

Showing 16 changed files with 280 additions and 1,427 deletions.
diff --git a/agenta-web/cypress/e2e/ab-testing-evaluation.cy.ts b/agenta-web/cypress/e2e/ab-testing-evaluation.cy.ts
@@ -43,9 +43,9 @@ describe("A/B Testing Evaluation workflow", () => {
 
     context("When executing the evaluation", () => {
         it("Should successfully execute the evaluation process", () => {
-            cy.visit(`/apps/${app_id}/annotations/human_a_b_testing`)
-            cy.url().should("include", "/annotations/human_a_b_testing")
-            cy.clickLinkAndWait('[data-cy="new-annotation-modal-button"]')
+            cy.visit(`/apps/${app_id}/evaluations?selectedEvaluation=ab_testing_evaluation`)
+            cy.url().should("include", "/evaluations?selectedEvaluation=ab_testing_evaluation")
+            cy.clickLinkAndWait('[data-cy="new-human-eval-modal-button"]')
 
             cy.get(".ant-modal-content").should("exist")
             cy.get('[data-cy="variants-dropdown-0"]').trigger("mouseover")

diff --git a/agenta-web/cypress/e2e/app-navigation.cy.ts b/agenta-web/cypress/e2e/app-navigation.cy.ts
@@ -28,28 +28,19 @@ describe("App Navigation without errors", () => {
         cy.get('[data-cy="app-testset-list"]').should("exist")
     })
 
-    it("should navigate successfully to Automatic Evaluation results evaluators page", () => {
-        cy.clickLinkAndWait('[data-cy="app-auto-evaluations-link"]')
-        cy.clickLinkAndWait('[data-cy="app-evaluators-link"]')
-        cy.url().should("include", "/evaluations/new-evaluator")
-    })
+    it("should navigate successfully to Evaluations page", () => {
+        cy.clickLinkAndWait('[data-cy="app-evaluations-link"]')
+        cy.url().should("include", "/evaluations")
+        cy.contains(/evaluations/i)
 
-    it("should navigate successfully to Automatic Evaluation results page", () => {
-        cy.clickLinkAndWait('[data-cy="app-auto-evaluations-link"]')
-        cy.clickLinkAndWait('[data-cy="app-evaluations-results-link"]')
-        cy.url().should("include", "/evaluations/results")
-    })
+        cy.get(".ant-tabs-tab").eq(1).click()
+        cy.url().should("include", "/evaluations?selectedEvaluation=ab_testing_evaluation")
 
-    it("should navigate successfully to A/B Test page", () => {
-        cy.clickLinkAndWait('[data-cy="app-human-evaluations-link"]')
-        cy.clickLinkAndWait('[data-cy="app-human-ab-testing-link"]')
-        cy.location("pathname").should("include", "/annotations/human_a_b_testing")
-    })
+        cy.get(".ant-tabs-tab").eq(2).click()
+        cy.url().should("include", "/evaluations?selectedEvaluation=single_model_evaluation")
 
-    it("should navigate successfully to Single Model Test page", () => {
-        cy.clickLinkAndWait('[data-cy="app-human-evaluations-link"]')
-        cy.clickLinkAndWait('[data-cy="app-single-model-test-link"]')
-        cy.location("pathname").should("include", "/annotations/single_model_test")
+        cy.get(".ant-tabs-tab").eq(0).click()
+        cy.url().should("include", "/evaluations?selectedEvaluation=auto_evaluation")
     })
 
     if (isDemo()) {

diff --git a/agenta-web/cypress/e2e/single-model-test-evaluation.cy.ts b/agenta-web/cypress/e2e/single-model-test-evaluation.cy.ts
@@ -16,9 +16,9 @@ describe("Single Model Test workflow", () => {
 
     context("When executing the evaluation", () => {
         it("Should successfully execute the evaluation process", () => {
-            cy.visit(`/apps/${app_id}/annotations/single_model_test`)
-            cy.url().should("include", "/annotations/single_model_test")
-            cy.clickLinkAndWait('[data-cy="new-annotation-modal-button"]')
+            cy.visit(`/apps/${app_id}/evaluations?selectedEvaluation=single_model_evaluation`)
+            cy.url().should("include", "/evaluations?selectedEvaluation=single_model_evaluation")
+            cy.clickLinkAndWait('[data-cy="new-human-eval-modal-button"]')
 
             cy.get(".ant-modal-content").should("exist")
 
@@ -49,10 +49,10 @@ describe("Single Model Test workflow", () => {
         })
 
         it("Should modify the evaluation vote scores", () => {
-            cy.visit(`/apps/${app_id}/annotations/single_model_test`)
-            cy.url().should("include", "/annotations/single_model_test")
+            cy.visit(`/apps/${app_id}/evaluations?selectedEvaluation=single_model_evaluation`)
+            cy.url().should("include", "/evaluations?selectedEvaluation=single_model_evaluation")
             cy.wait(1000)
-            cy.clickLinkAndWait('[data-cy="single-model-view-evaluation-button"]')
+            cy.clickLinkAndWait(".ant-table-row").eq(0)
             cy.get('[data-cy="evalInstructionsShown-ok-btn"]').click()
             cy.get('[data-cy="evaluation-vote-panel-numeric-vote-input"]').clear()
             cy.get('[data-cy="evaluation-vote-panel-numeric-vote-input"]').type("85")

diff --git a/agenta-web/src/components/Evaluations/AutomaticEvaluationResult.tsx b/agenta-web/src/components/Evaluations/AutomaticEvaluationResult.tsx