Merge pull request #262 from marlenezw/eval_image

adding eval tables to gh action.
Azure-Samples · Nov 13, 2024 · f20795d · f20795d
2 parents 9302f02 + 5e5ef90
commit f20795d
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 17 deletions.
diff --git a/.github/workflows/evaluate.yml b/.github/workflows/evaluate.yml
@@ -75,13 +75,20 @@ jobs:
         with:
           name: eval_result
           path: ./src/api/evaluate/eval_results.jsonl
+
+      - name: Upload image eval results as build artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: eval_image_result
+          path: ./src/api/evaluate/image_eval_results.jsonl
 
       - name: GitHub Summary Step
         if: ${{ success() }}
         working-directory: ./src/api
         run: |
           echo "" >> $GITHUB_STEP_SUMMARY
 
-          echo "📊 Promptflow Evaluation Results" >> $GITHUB_STEP_SUMMARY
+          echo "📊 Evaluation Results" >> $GITHUB_STEP_SUMMARY
           cat evaluate/eval_results.md >> $GITHUB_STEP_SUMMARY
+          cat evaluate/image_eval_results.md >> $GITHUB_STEP_SUMMARY
         
diff --git a/src/api/evaluate/evaluators.py b/src/api/evaluate/evaluators.py
@@ -138,22 +138,6 @@ def __init__(self, project_scope):
                 credential=DefaultAzureCredential(), 
                 azure_ai_project=project_scope,
             ),
-            # "violence":ViolenceMultimodalEvaluator(
-            #     credential=DefaultAzureCredential(), 
-            #     azure_ai_project=project_scope,
-            # ), 
-            # "self_harm":SelfHarmMultimodalEvaluator(
-            #     credential=DefaultAzureCredential(), 
-            #     azure_ai_project=project_scope,
-            # ), 
-            # "hate_unfairness":HateUnfairnessMultimodalEvaluator(
-            #     credential=DefaultAzureCredential(), 
-            #     azure_ai_project=project_scope,
-            # ), 
-            # "sexual":SexualMultimodalEvaluator(
-            #     credential=DefaultAzureCredential(), 
-            #     azure_ai_project=project_scope,
-            # ),
             "protected_material": ProtectedMaterialMultimodalEvaluator(
                 credential=DefaultAzureCredential(),
                 azure_ai_project=project_scope,
@@ -188,6 +172,7 @@ def __call__(self, *, messages, **kwargs):
         print("\n===== Calling Evaluate API - Content Safety & Protected Material Evaluator for multi-modal =======")
         output = {}
         if runningonGH:
+            print('Running in actions')
             for message in messages:
                 conversation = {"conversation": { "messages" : message}}