changes for inter-annotator-difference

.github/workflows/linters.yml

@@ -5,10 +5,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v1
-      - name: Set up Python 3.9.20
+      - name: Set up Python 3.9.21
         uses: actions/setup-python@v1
         with:
-          python-version: 3.9.20
+          python-version: 3.9.21
       - name: Debug Message - Check Github branch
         run: echo "Current Git branch is ${GITHUB_REF##*/}" 
       - name: Install Black

backend/dataset/views.py

@@ -325,9 +325,9 @@ def download(self, request, pk):
             json_data = json.dumps(exported_items, default=str)
             # Create a StreamingHttpResponse with the JSON data
             response = StreamingHttpResponse(json_data, content_type="application/json")
-            response[
-                "Content-Disposition"
-            ] = f'attachment; filename="{dataset_instance}.json"'
+            response["Content-Disposition"] = (
+                f'attachment; filename="{dataset_instance}.json"'
+            )
             return response
         return StreamingHttpResponse(
             exported_items, status=status.HTTP_200_OK, content_type=content_type

backend/functions/tasks.py

@@ -452,25 +452,25 @@ def get_modified_stats_result(
         for key, value in result_rev_anno_stats.items():
             result[f"Reviewer - {key.replace('_', ' ').title()} Annotations"] = value
         for key, value in result_sup_anno_stats.items():
-            result[
-                f"Superchecker - {key.replace('_', ' ').title()} Annotations"
-            ] = value
+            result[f"Superchecker - {key.replace('_', ' ').title()} Annotations"] = (
+                value
+            )
     if meta_stats or complete_stats:
         for key, value in result_ann_meta_stats.items():
             for sub_key in value.keys():
-                result[
-                    f"Annotator - {key.replace('_', ' ').title()} {sub_key}"
-                ] = value[sub_key]
+                result[f"Annotator - {key.replace('_', ' ').title()} {sub_key}"] = (
+                    value[sub_key]
+                )
         for key, value in result_rev_meta_stats.items():
             for sub_key in value.keys():
                 result[f"Reviewer - {key.replace('_', ' ').title()} {sub_key}"] = value[
                     sub_key
                 ]
         for key, value in result_sup_meta_stats.items():
             for sub_key in value.keys():
-                result[
-                    f"Superchecker - {key.replace('_', ' ').title()} {sub_key}"
-                ] = value[sub_key]
+                result[f"Superchecker - {key.replace('_', ' ').title()} {sub_key}"] = (
+                    value[sub_key]
+                )
 
     # adding unassigned tasks count
     result["Annotator - Unassigned Tasks"] = get_task_count_unassigned(proj_id, user)
@@ -770,9 +770,9 @@ def schedule_mail_to_download_all_projects(
             query_params = QueryDict(mutable=True)
             query_params["include_input_data_metadata_json"] = "true"
             query_params["export_type"] = "CSV"
-            query_params[
-                "task_status"
-            ] = "incomplete,annotated,reviewed,super_checked,exported"
+            query_params["task_status"] = (
+                "incomplete,annotated,reviewed,super_checked,exported"
+            )
             custom_request = Request(factory.get(url, data=query_params, timeout=15))
             custom_request.user = user
             try:

backend/notifications/tasks.py

@@ -38,9 +38,7 @@ def create_notification_handler(
             else (
                 f"/projects/{project_id}"
                 if project_id
-                else f"/task/{task_id}"
-                if task_id
-                else None
+                else f"/task/{task_id}" if task_id else None
             )
         )
         new_notif = Notification(

backend/organizations/tasks.py

@@ -47,9 +47,7 @@ def get_all_annotation_reports(
         else (
             "Part Time"
             if participation_type == 2
-            else "Contract Basis"
-            if participation_type == 4
-            else "N/A"
+            else "Contract Basis" if participation_type == 4 else "N/A"
         )
     )
     role = get_role_name(user.role)
@@ -130,9 +128,7 @@ def get_all_review_reports(
         else (
             "Part Time"
             if participation_type == 2
-            else "Contract Basis"
-            if participation_type == 4
-            else "N/A"
+            else "Contract Basis" if participation_type == 4 else "N/A"
         )
     )
     role = get_role_name(user.role)
@@ -246,9 +242,7 @@ def get_all_supercheck_reports(
         else (
             "Part Time"
             if user.participation_type == 2
-            else "Contract Basis"
-            if user.participation_type == 4
-            else "N/A"
+            else "Contract Basis" if user.participation_type == 4 else "N/A"
         )
     )
     role = get_role_name(user.role)
@@ -464,9 +458,7 @@ def send_user_reports_mail_org(
             else (
                 "Part Time"
                 if participation_type == 2
-                else "Contract Basis"
-                if participation_type == 4
-                else "N/A"
+                else "Contract Basis" if participation_type == 4 else "N/A"
             )
         )
         for participation_type in participation_types
@@ -1199,9 +1191,7 @@ def send_user_analytics_mail_org(
                 else (
                     "Part Time"
                     if participation_type == 2
-                    else "Contract Basis"
-                    if participation_type == 4
-                    else "N/A"
+                    else "Contract Basis" if participation_type == 4 else "N/A"
                 )
             )
             role = get_role_name(annotator.role)

backend/organizations/views.py

@@ -456,9 +456,7 @@ def quality_reports(self, request, pk=None):
                 else (
                     "Part Time"
                     if participation_type == 2
-                    else "Contract Basis"
-                    if participation_type == 4
-                    else "N/A"
+                    else "Contract Basis" if participation_type == 4 else "N/A"
                 )
             )
             role = get_role_name(annotator.role)
@@ -776,9 +774,7 @@ def user_analytics(self, request, pk=None):
                     else (
                         "Part Time"
                         if participation_type == 2
-                        else "Contract Basis"
-                        if participation_type == 4
-                        else "N/A"
+                        else "Contract Basis" if participation_type == 4 else "N/A"
                     )
                 )
                 role = get_role_name(annotator.role)
@@ -902,9 +898,9 @@ def iter_items(items, pseudo_buffer):
                     status=status.HTTP_200_OK,
                     content_type="text/csv",
                 )
-                response[
-                    "Content-Disposition"
-                ] = f'attachment; filename="{organization.title}_user_analytics.csv"'
+                response["Content-Disposition"] = (
+                    f'attachment; filename="{organization.title}_user_analytics.csv"'
+                )
                 return response
 
             return Response(data=final_result, status=status.HTTP_200_OK)

backend/projects/utils.py

@@ -530,7 +530,7 @@ def add_extra_task_data(t, project):
         .filter(review_user__isnull=True)
     )
     total_ratings, seen = [], {}
-    max_rating, curr_response = float("-inf"), ""
+    max_rating, curr_response, min_rating = float("-inf"), "", float("inf")
     for st in similar_tasks:
         ann = Annotation.objects.filter(task=st, annotation_status=LABELED)[0]
         for r in ann.result:
@@ -558,14 +558,18 @@ def add_extra_task_data(t, project):
             seen[st.id] = curr_response
             total_ratings.append(curr_response)
             max_rating = max(max_rating, curr_response)
+            min_rating = min(min_rating, curr_response)
     t.data["avg_rating"] = -1
     t.data["curr_rating"] = -1
     t.data["inter_annotator_difference"] = -1
     if t.id in seen:
         t.data["avg_rating"] = sum(total_ratings) / len(total_ratings)
+        t.data["total_rating"] = sum(total_ratings)
         t.data["curr_rating"] = seen[t.id]
         t.data["inter_annotator_difference"] = (
-            max_rating - seen[t.id] if max_rating > float("-inf") else -1
+            max_rating - min_rating
+            if (max_rating > float("-inf") and min_rating < float("inf"))
+            else -1
         )
     t.save()
 

backend/projects/views.py

@@ -636,9 +636,9 @@ def get_supercheck_reports(proj_id, userid, start_date, end_date):
         result["Rejected Word Count"] = rejected_word_count
     elif proj_type in get_audio_project_types():
         result["Validated Segments Duration"] = validated_audio_duration
-        result[
-            "Validated With Changes Segments Duration"
-        ] = validated_with_changes_audio_duration
+        result["Validated With Changes Segments Duration"] = (
+            validated_with_changes_audio_duration
+        )
         result["Rejected Segments Duration"] = rejected_audio_duration
         result["Total Raw Audio Duration"] = total_raw_audio_duration
         result["Average Word Error Rate R/S"] = round(avg_word_error_rate, 2)
@@ -3790,11 +3790,11 @@ def download(self, request, pk=None, *args, **kwargs):
                 if include_input_data_metadata_json:
                     dataset_type = project.dataset_id.all()[0].dataset_type
                     dataset_model = getattr(dataset_models, dataset_type)
-                    task_dict["data"][
-                        "input_data_metadata_json"
-                    ] = dataset_model.objects.get(
-                        pk=task_dict["input_data"]
-                    ).metadata_json
+                    task_dict["data"]["input_data_metadata_json"] = (
+                        dataset_model.objects.get(
+                            pk=task_dict["input_data"]
+                        ).metadata_json
+                    )
                 del task_dict["annotation_users"]
                 del task_dict["review_user"]
                 tasks_list.append(OrderedDict(task_dict))

backend/tasks/views.py

@@ -1215,12 +1215,16 @@ def get_users_recent_tasks(self, request):
                     "Project ID": annotation.task.project_id.id,
                     "Task ID": annotation.task.id,
                     "Updated at": utc_to_ist(annotation.updated_at),
-                    "Annotated at": utc_to_ist(annotation.annotated_at)
-                    if annotation.annotated_at
-                    else None,
-                    "Created at": utc_to_ist(annotation.created_at)
-                    if annotation.created_at
-                    else None,
+                    "Annotated at": (
+                        utc_to_ist(annotation.annotated_at)
+                        if annotation.annotated_at
+                        else None
+                    ),
+                    "Created at": (
+                        utc_to_ist(annotation.created_at)
+                        if annotation.created_at
+                        else None
+                    ),
                 }
 
                 response.append(data)
@@ -1656,16 +1660,16 @@ def partial_update(self, request, pk=None):
                         and len(annotation_obj.result) > len(request.data["result"])
                     ):
                         request.data["result"] = annotation_obj.result
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            annotation_obj.result
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                annotation_obj.result
+                            )
                         )
                     else:
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            request.data["result"]
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                request.data["result"]
+                            )
                         )
                 annotation_response = super().partial_update(request)
                 if is_IDC:
@@ -1844,16 +1848,16 @@ def partial_update(self, request, pk=None):
                         and len(annotation_obj.result) > len(request.data["result"])
                     ):
                         request.data["result"] = annotation_obj.result
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            annotation_obj.result
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                annotation_obj.result
+                            )
                         )
                     else:
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            request.data["result"]
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                request.data["result"]
+                            )
                         )
                 annotation_response = super().partial_update(request)
                 if is_IDC:
@@ -2053,16 +2057,16 @@ def partial_update(self, request, pk=None):
                         and len(annotation_obj.result) > len(request.data["result"])
                     ):
                         request.data["result"] = annotation_obj.result
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            annotation_obj.result
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                annotation_obj.result
+                            )
                         )
                     else:
-                        request.data[
-                            "meta_stats"
-                        ] = compute_meta_stats_for_instruction_driven_chat(
-                            request.data["result"]
+                        request.data["meta_stats"] = (
+                            compute_meta_stats_for_instruction_driven_chat(
+                                request.data["result"]
+                            )
                         )
                 annotation_response = super().partial_update(request)
                 if is_IDC:
@@ -2197,9 +2201,11 @@ def convert_chitralekha_format_to_LSF(
             text_dict = {
                 "origin": "manual",
                 "to_name": "audio_url",
-                "from_name": "transcribed_json"
-                if not is_acoustic
-                else "verbatim_transcribed_json",
+                "from_name": (
+                    "transcribed_json"
+                    if not is_acoustic
+                    else "verbatim_transcribed_json"
+                ),
                 "original_length": audio_duration,
             }
 
@@ -2479,19 +2485,19 @@ def get_celery_tasks(request):
     for i in filtered_tasks:
         if filtered_tasks[i]["succeeded"] is not None:
             filtered_tasks[i]["succeeded"] = datetime.fromtimestamp(
-                filtered_tasks[i]["succeeded"],tz=timezone.utc
+                filtered_tasks[i]["succeeded"], tz=timezone.utc
             ).strftime("%Y-%m-%dT%H:%M:%S.%fZ")
         if filtered_tasks[i]["failed"] is not None:
             filtered_tasks[i]["failed"] = datetime.fromtimestamp(
-                filtered_tasks[i]["failed"],tz=timezone.utc
+                filtered_tasks[i]["failed"], tz=timezone.utc
             ).strftime("%Y-%m-%dT%H:%M:%S.%fZ")
         if filtered_tasks[i]["started"] is not None:
             filtered_tasks[i]["started"] = datetime.fromtimestamp(
-                filtered_tasks[i]["started"],tz=timezone.utc
+                filtered_tasks[i]["started"], tz=timezone.utc
             ).strftime("%Y-%m-%dT%H:%M:%S.%fZ")
         if filtered_tasks[i]["received"] is not None:
             filtered_tasks[i]["received"] = datetime.fromtimestamp(
-                filtered_tasks[i]["received"],tz=timezone.utc
+                filtered_tasks[i]["received"], tz=timezone.utc
             ).strftime("%Y-%m-%dT%H:%M:%S.%fZ")
 
     if "error" in filtered_tasks: