edx-enterprise-data 8.12.1__tar.gz → 8.13.0__tar.gz

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/CHANGELOG.rst

@@ -15,6 +15,10 @@ Unreleased
 ----------
 
 =========================
+[8.13.0] - 2024-09-23
+---------------------
+  * feat: convert the skills pandas code into sql queries for better performance
+
 [8.12.1] - 2024-09-16
 ---------------------
   * fix: Remove hyphens from enterprise customer UUID before database query.

{edx_enterprise_data-8.12.1/edx_enterprise_data.egg-info → edx_enterprise_data-8.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: edx-enterprise-data
-Version: 8.12.1
+Version: 8.13.0
 Summary: Enterprise Reporting
 Home-page: https://github.com/openedx/edx-enterprise-data
 Author: edX

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0/edx_enterprise_data.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: edx-enterprise-data
-Version: 8.12.1
+Version: 8.13.0
 Summary: Enterprise Reporting
 Home-page: https://github.com/openedx/edx-enterprise-data
 Author: edX

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/edx_enterprise_data.egg-info/SOURCES.txt

@@ -30,10 +30,12 @@ enterprise_data/admin_analytics/database/utils.py
 enterprise_data/admin_analytics/database/queries/__init__.py
 enterprise_data/admin_analytics/database/queries/fact_engagement_admin_dash.py
 enterprise_data/admin_analytics/database/queries/fact_enrollment_admin_dash.py
+enterprise_data/admin_analytics/database/queries/skills_daily_rollup_admin_dash.py
 enterprise_data/admin_analytics/database/tables/__init__.py
 enterprise_data/admin_analytics/database/tables/base.py
 enterprise_data/admin_analytics/database/tables/fact_engagement_admin_dash.py
 enterprise_data/admin_analytics/database/tables/fact_enrollment_admin_dash.py
+enterprise_data/admin_analytics/database/tables/skills_daily_rollup_admin_dash.py
 enterprise_data/api/__init__.py
 enterprise_data/api/urls.py
 enterprise_data/api/v0/__init__.py

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/__init__.py

@@ -2,4 +2,4 @@
 Enterprise data api application. This Django app exposes API endpoints used by enterprises.
 """
 
-__version__ = "8.12.1"
+__version__ = "8.13.0"

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/admin_analytics/data_loaders.py

@@ -147,51 +147,3 @@ def fetch_max_enrollment_datetime():
     if not results:
         return None
     return pandas.to_datetime(results[0][0])
-
-
-def fetch_skills_data(enterprise_uuid: str):
-    """
-    Fetch skills data from the database for the given enterprise customer.
-
-    Arguments:
-        enterprise_uuid (str): The UUID of the enterprise customer.
-
-    Returns:
-        (pandas.DataFrame): The skills data.
-    """
-
-    enterprise_uuid = enterprise_uuid.replace('-', '')
-
-    cols = [
-        'course_number',
-        'skill_type',
-        'skill_name',
-        'skill_url',
-        'confidence',
-        'skill_rank',
-        'course_title',
-        'course_key',
-        'level_type',
-        'primary_subject_name',
-        'date',
-        'enterprise_customer_uuid',
-        'enterprise_customer_name',
-        'enrolls',
-        'completions',
-    ]
-    query = get_select_query(
-        table='skills_daily_rollup_admin_dash', columns=cols, enterprise_uuid=enterprise_uuid
-    )
-
-    with timer('fetch_skills_data'):
-        skills = run_query(query=query)
-
-    if not skills:
-        raise Http404(f'No skills data found for enterprise {enterprise_uuid}')
-
-    LOGGER.info(f'[PLOTLY] Skills data fetched successfully. Records: {len(skills)}')
-    skills = pandas.DataFrame(numpy.array(skills), columns=cols)
-    LOGGER.info('[PLOTLY] Skills data converted to DataFrame.')
-    skills['date'] = skills['date'].astype('datetime64[ns]')
-
-    return skills

edx_enterprise_data-8.13.0/enterprise_data/admin_analytics/database/queries/skills_daily_rollup_admin_dash.py

@@ -0,0 +1,118 @@
+"""
+Module containing queries for the skills_daily_rollup_admin_dash table.
+"""
+
+
+class SkillsDailyRollupAdminDashQueries:
+    """
+    Queries related to the skills_daily_rollup_admin_dash table.
+    """
+    @staticmethod
+    def get_top_skills():
+        """
+        Get the query to fetch the top skills for an enterprise customer.
+        """
+        return """
+            SELECT
+                skill_name,
+                skill_type,
+                SUM(enrolls) AS enrolls,
+                SUM(completions) AS completions
+            FROM
+                skills_daily_rollup_admin_dash
+            WHERE
+                enterprise_customer_uuid=%(enterprise_customer_uuid)s AND
+                date BETWEEN %(start_date)s AND %(end_date)s
+            GROUP BY
+                skill_name, skill_type
+            ORDER BY
+                enrolls DESC, completions DESC;
+        """
+
+    @staticmethod
+    def get_top_skills_by_enrollment():
+        """
+        Get the query to fetch the top skills by enrollment for an enterprise customer.
+        """
+        return """
+            WITH TopSkills AS (
+                -- Get top 10 skills by total enrollments
+                SELECT
+                    skill_name
+                FROM
+                    skills_daily_rollup_admin_dash
+                WHERE
+                    enterprise_customer_uuid=%(enterprise_customer_uuid)s
+                    AND date BETWEEN %(start_date)s AND %(end_date)s
+                GROUP BY
+                    skill_name
+                ORDER BY
+                    SUM(enrolls) DESC
+                LIMIT 10
+            )
+            SELECT
+                sd.skill_name,
+                CASE
+                    WHEN sd.primary_subject_name IN (
+                        'business-management', 'computer-science',
+                        'data-analysis-statistics', 'engineering', 'communication'
+                    ) THEN sd.primary_subject_name
+                    ELSE 'other'
+                END AS subject_name,
+                SUM(sd.enrolls) AS count
+            FROM
+                skills_daily_rollup_admin_dash AS sd
+            JOIN
+                TopSkills AS ts ON sd.skill_name = ts.skill_name
+            WHERE
+                sd.enterprise_customer_uuid=%(enterprise_customer_uuid)s
+                AND date BETWEEN %(start_date)s AND %(end_date)s
+            GROUP BY
+                sd.skill_name, subject_name
+            ORDER BY
+                subject_name;
+        """
+
+    @staticmethod
+    def get_top_skills_by_completion():
+        """
+        Get the query to fetch the top skills by completion for an enterprise customer.
+        """
+        return """
+            WITH TopSkills AS (
+                -- Get top 10 skills by total completions
+                SELECT
+                    skill_name
+                FROM
+                    skills_daily_rollup_admin_dash
+                WHERE
+                    enterprise_customer_uuid=%(enterprise_customer_uuid)s
+                    AND date BETWEEN %(start_date)s AND %(end_date)s
+                GROUP BY
+                    skill_name
+                ORDER BY
+                    SUM(completions) DESC
+                LIMIT 10
+            )
+            SELECT
+                sd.skill_name,
+                CASE
+                    WHEN sd.primary_subject_name IN (
+                        'business-management', 'computer-science',
+                        'data-analysis-statistics', 'engineering', 'communication'
+                    ) THEN sd.primary_subject_name
+                    ELSE 'other'
+                END AS subject_name,
+                SUM(sd.completions) AS count
+            FROM
+                skills_daily_rollup_admin_dash AS sd
+            JOIN
+                TopSkills AS ts ON sd.skill_name = ts.skill_name
+            WHERE
+                sd.enterprise_customer_uuid=%(enterprise_customer_uuid)s
+                AND date BETWEEN %(start_date)s AND %(end_date)s
+            GROUP BY
+                sd.skill_name, subject_name
+            ORDER BY
+                subject_name;
+        """

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/admin_analytics/database/tables/__init__.py

@@ -3,3 +3,4 @@ This module contains the database queries for the admin analytics.
 """
 from .fact_engagement_admin_dash import FactEngagementAdminDashTable
 from .fact_enrollment_admin_dash import FactEnrollmentAdminDashTable
+from .skills_daily_rollup_admin_dash import SkillsDailyRollupAdminDashTable

edx_enterprise_data-8.13.0/enterprise_data/admin_analytics/database/tables/skills_daily_rollup_admin_dash.py

@@ -0,0 +1,94 @@
+"""
+Module for interacting with the skills_daily_rollup_admin_dash table.
+"""
+from datetime import date
+from uuid import UUID
+
+from enterprise_data.admin_analytics.database.queries.skills_daily_rollup_admin_dash import (
+    SkillsDailyRollupAdminDashQueries,
+)
+from enterprise_data.admin_analytics.database.tables.base import BaseTable
+from enterprise_data.admin_analytics.database.utils import run_query
+
+
+class SkillsDailyRollupAdminDashTable(BaseTable):
+    """
+    Class for communicating with the skills_daily_rollup_admin_dash table.
+    """
+    queries = SkillsDailyRollupAdminDashQueries()
+
+    def get_top_skills(self, enterprise_customer_uuid: UUID, start_date: date, end_date: date):
+        """
+        Get the top skills for the given enterprise customer.
+
+        Arguments:
+            enterprise_customer_uuid (UUID): The UUID of the enterprise customer.
+            start_date (date): The start date.
+            end_date (date): The end date.
+
+        Returns:
+            list<dict>: A list of dictionaries containing the skill_name, skill_type, enrolls and completions.
+        """
+        return run_query(
+            query=self.queries.get_top_skills(),
+            params={
+                'enterprise_customer_uuid': enterprise_customer_uuid,
+                'start_date': start_date,
+                'end_date': end_date,
+            },
+            as_dict=True,
+        )
+
+    def get_top_skills_by_enrollment(
+        self,
+        enterprise_customer_uuid: UUID,
+        start_date: date,
+        end_date: date
+    ):
+        """
+        Get the top skills by enrollments for the given enterprise customer.
+
+        Arguments:
+            enterprise_customer_uuid (UUID): The UUID of the enterprise customer.
+            start_date (date): The start date.
+            end_date (date): The end date.
+
+        Returns:
+            list<dict>: A list of dictionaries containing the skill_name, subject_name, count.
+        """
+        return run_query(
+            query=self.queries.get_top_skills_by_enrollment(),
+            params={
+                'enterprise_customer_uuid': enterprise_customer_uuid,
+                'start_date': start_date,
+                'end_date': end_date,
+            },
+            as_dict=True,
+        )
+
+    def get_top_skills_by_completion(
+        self,
+        enterprise_customer_uuid: UUID,
+        start_date: date,
+        end_date: date
+    ):
+        """
+        Get the top skills by completion for the given enterprise customer.
+
+        Arguments:
+            enterprise_customer_uuid (UUID): The UUID of the enterprise customer.
+            start_date (date): The start date.
+            end_date (date): The end date.
+
+        Returns:
+            list<dict>: A list of dictionaries containing the skill_name, subject_name, count.
+        """
+        return run_query(
+            query=self.queries.get_top_skills_by_completion(),
+            params={
+                'enterprise_customer_uuid': enterprise_customer_uuid,
+                'start_date': start_date,
+                'end_date': end_date,
+            },
+            as_dict=True,
+        )

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/admin_analytics/utils.py

@@ -12,9 +12,7 @@ from enterprise_data.admin_analytics.data_loaders import (
     fetch_engagement_data,
     fetch_enrollment_data,
     fetch_max_enrollment_datetime,
-    fetch_skills_data,
 )
-from enterprise_data.utils import date_filter, primary_subject_truncate
 
 LOGGER = getLogger(__name__)
 
@@ -23,9 +21,6 @@ class ChartType(Enum):
     """
     Chart types.
     """
-    BUBBLE = 'bubble'
-    TOP_SKILLS_ENROLLMENT = 'top_skills_enrollment'
-    TOP_SKILLS_COMPLETION = 'top_skills_completion'
     COMPLETIONS_OVER_TIME = 'completions_over_time'
     TOP_COURSES_BY_COMPLETIONS = 'top_courses_by_completions'
     TOP_SUBJECTS_BY_COMPLETIONS = 'top_subjects_by_completions'
@@ -183,169 +178,3 @@ def fetch_and_cache_engagements_data(enterprise_id, cache_expiry):
             cache_key, engagements, get_cache_timeout(cache_expiry)
         )
         return engagements
-
-
-def fetch_and_cache_skills_data(enterprise_id, cache_expiry):
-    """
-    Helper method to fetch and cache skills data.
-
-    Arguments:
-        enterprise_id (str): UUID of the enterprise customer in string format.
-        cache_expiry (datetime): Datetime object denoting the cache expiry.
-
-    Returns:
-        (pandas.DataFrame): The skills data.
-    """
-    cache_key = get_cache_key(
-        resource='enterprise-admin-analytics-aggregate-skills',
-        enterprise_customer=enterprise_id,
-    )
-    cached_response = TieredCache.get_cached_response(cache_key)
-
-    if cached_response.is_found:
-        LOGGER.info(f"Skills data found in cache for Enterprise [{enterprise_id}]")
-        return cached_response.value
-    else:
-        skills = fetch_skills_data(enterprise_id)
-        TieredCache.set_all_tiers(
-            cache_key, skills, get_cache_timeout(cache_expiry)
-        )
-        return skills
-
-
-def get_skills_bubble_chart_df(skills_filtered):
-    """ Get the skills data for the bubble chart.
-
-    Args:
-        skills_filtered (pandas.DataFrame): The skills data.
-
-    Returns:
-        (pandas.DataFrame): The skills data for the bubble chart.
-    """
-
-    # Group by skill_name and skill_type, and aggregate enrolls and completions
-    skills_aggregated = (
-        skills_filtered.groupby(['skill_name', 'skill_type'], as_index=False)
-        .agg(enrolls=('enrolls', 'sum'), completions=('completions', 'sum'))
-    )
-
-    # Convert enrolls and completions to integers
-    skills_aggregated['enrolls'] = skills_aggregated['enrolls'].astype(int)
-    skills_aggregated['completions'] = skills_aggregated['completions'].astype(int)
-
-    # Sort the dataframe by enrolls and completions in descending order
-    skills_aggregated = skills_aggregated.sort_values(by=['enrolls', 'completions'], ascending=False)
-
-    return skills_aggregated
-
-
-def get_top_skills_enrollment(skills_filtered):
-    """ Get the top skills by enrolls.
-
-    Args:
-        skills_filtered (pandas.DataFrame): The skills data.
-
-    Returns:
-        (pandas.DataFrame): The top skills by enrolls data
-    """
-
-    # Get the top 10 skills by enrolls
-    top_skills = (
-        skills_filtered.groupby('skill_name')
-        .enrolls.sum()
-        .sort_values(ascending=False)
-        .head(10)
-        .index
-    )
-
-    # Apply primary_subject_truncate to the primary_subject_name column
-    skills_filtered['primary_subject_name'] = skills_filtered['primary_subject_name'].apply(primary_subject_truncate)
-
-    # Filter data for the top skills and aggregate enrolls by skill_name and primary_subject_name
-    top_skills_enrollment_data = (
-        skills_filtered[skills_filtered.skill_name.isin(top_skills)]
-        .groupby(['skill_name', 'primary_subject_name'], as_index=False)
-        .agg(count=('enrolls', 'sum'))
-    )
-
-    # Sort the dataframe by primary_subject_name
-    top_skills_enrollment_data = top_skills_enrollment_data.sort_values(by="primary_subject_name")
-
-    return top_skills_enrollment_data
-
-
-def get_top_skills_completion(skills_filtered):
-    """ Get the top skills by completions.
-
-    Args:
-        skills_filtered (pandas.DataFrame): The skills data.
-
-    Returns:
-        (pandas.DataFrame): The top skills by completions
-    """
-
-    # Get the top 10 skills by completions
-    top_skills = (
-        skills_filtered.groupby('skill_name')
-        .completions.sum()
-        .sort_values(ascending=False)
-        .head(10)
-        .index
-    )
-
-    # Apply primary_subject_truncate to the primary_subject_name column
-    skills_filtered['primary_subject_name'] = skills_filtered['primary_subject_name'].apply(primary_subject_truncate)
-
-    # Filter data for the top skills and aggregate completions by skill_name and primary_subject_name
-    top_skills_completion_data = (
-        skills_filtered[skills_filtered.skill_name.isin(top_skills)]
-        .groupby(['skill_name', 'primary_subject_name'], as_index=False)
-        .agg(count=('completions', 'sum'))
-    )
-
-    # Sort the dataframe by primary_subject_name
-    top_skills_completion_data = top_skills_completion_data.sort_values(by='primary_subject_name')
-
-    return top_skills_completion_data
-
-
-def get_skills_chart_data(chart_type, start_date, end_date, skills):
-    """
-    Get chart data for skill charts.
-
-    Arguments:
-        chart_type (ChartType): The type of chart to generate.
-        start_date (datetime): The start date for the date filter.
-        end_date (datetime): The end date for the date filter.
-        skills (pandas.DataFrame): The skills data.
-    """
-    skills_filtered = date_filter(start=start_date, end=end_date, data_frame=skills.copy(), date_column='date')
-    if chart_type == ChartType.BUBBLE:
-        return get_skills_bubble_chart_df(skills_filtered=skills_filtered.copy())
-    elif chart_type == ChartType.TOP_SKILLS_ENROLLMENT:
-        return get_top_skills_enrollment(skills_filtered=skills_filtered.copy())
-    elif chart_type == ChartType.TOP_SKILLS_COMPLETION:
-        return get_top_skills_completion(skills_filtered=skills_filtered.copy())
-    else:
-        raise ValueError(f"Invalid chart type: {chart_type}")
-
-
-def get_top_skills_csv_data(skills, start_date, end_date):
-    """ Get the top skills data for CSV download.
-
-    Args:
-        skills (pandas.DataFrame): The skills data.
-        start_date (str): The start date for the date filter.
-        end_date (str): The end date for the date filter.
-
-    Returns:
-        (pandas.DataFrame): The top skills data for CSV download.
-    """
-    dff = get_skills_chart_data(
-        chart_type=ChartType.BUBBLE,
-        start_date=start_date,
-        end_date=end_date,
-        skills=skills.copy(),
-    )
-    dff = dff.sort_values(by='enrolls', ascending=False)
-    return dff

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/api/v1/views/enterprise_admin.py

@@ -1,7 +1,7 @@
 """
 Views for enterprise admin api v1.
 """
-from datetime import datetime, timedelta
+from datetime import datetime
 
 from edx_rbac.decorators import permission_required
 from edx_rest_framework_extensions.auth.jwt.authentication import JwtAuthentication
@@ -10,16 +10,11 @@ from rest_framework.response import Response
 from rest_framework.status import HTTP_200_OK, HTTP_404_NOT_FOUND
 from rest_framework.views import APIView
 
-from django.http import HttpResponse
-
 from enterprise_data.admin_analytics.data_loaders import fetch_max_enrollment_datetime
-from enterprise_data.admin_analytics.database.tables import FactEngagementAdminDashTable, FactEnrollmentAdminDashTable
-from enterprise_data.admin_analytics.utils import (
-    ChartType,
-    fetch_and_cache_enrollments_data,
-    fetch_and_cache_skills_data,
-    get_skills_chart_data,
-    get_top_skills_csv_data,
+from enterprise_data.admin_analytics.database.tables import (
+    FactEngagementAdminDashTable,
+    FactEnrollmentAdminDashTable,
+    SkillsDailyRollupAdminDashTable,
 )
 from enterprise_data.api.v1 import serializers
 from enterprise_data.models import (
@@ -157,61 +152,35 @@ class EnterpriseAdminAnalyticsSkillsView(APIView):
         Returns:
             response(HttpResponse): response object
         """
-        serializer = serializers.AdminAnalyticsAggregatesQueryParamsSerializer(
-            data=request.GET
-        )
+        enterprise_id = enterprise_id.replace('-', '')
+        serializer = serializers.AdvanceAnalyticsQueryParamSerializer(data=request.GET)
         serializer.is_valid(raise_exception=True)
-        last_updated_at = fetch_max_enrollment_datetime()
-        cache_expiry = (
-            last_updated_at + timedelta(days=1) if last_updated_at else datetime.now()
-        )
 
-        enrollment = fetch_and_cache_enrollments_data(
-            enterprise_id, cache_expiry
-        ).copy()
+        if (start_date := serializer.data.get('start_date')) is None:
+            start_date, _ = FactEnrollmentAdminDashTable().get_enrollment_date_range(enterprise_id)
 
-        start_date = serializer.data.get('start_date', enrollment.enterprise_enrollment_date.min())
         end_date = serializer.data.get('end_date', datetime.now())
 
-        skills = fetch_and_cache_skills_data(enterprise_id, cache_expiry).copy()
-
-        if serializer.data.get('response_type') == 'csv':
-            csv_data = get_top_skills_csv_data(skills, start_date, end_date)
-            response = HttpResponse(content_type='text/csv')
-            filename = f"Skills by Enrollment and Completion, {start_date} - {end_date}.csv"
-            response['Content-Disposition'] = f'attachment; filename="{filename}"'
-            response['Access-Control-Expose-Headers'] = 'Content-Disposition'
-            csv_data.to_csv(path_or_buf=response, index=False)
-            return response
-
-        with timer('skills_all_charts_data'):
-            top_skills = get_skills_chart_data(
-                chart_type=ChartType.BUBBLE,
-                start_date=start_date,
-                end_date=end_date,
-                skills=skills,
-            )
-            top_skills_enrollments = get_skills_chart_data(
-                chart_type=ChartType.TOP_SKILLS_ENROLLMENT,
-                start_date=start_date,
-                end_date=end_date,
-                skills=skills,
+        with timer('top_skills'):
+            skills = SkillsDailyRollupAdminDashTable().get_top_skills(enterprise_id, start_date, end_date)
+
+        with timer('top_skills_by_enrollments'):
+            top_skills_by_enrollments = SkillsDailyRollupAdminDashTable().get_top_skills_by_enrollment(
+                enterprise_id,
+                start_date,
+                end_date
             )
-            top_skills_by_completions = get_skills_chart_data(
-                chart_type=ChartType.TOP_SKILLS_COMPLETION,
-                start_date=start_date,
-                end_date=end_date,
-                skills=skills,
+        with timer('top_skills_by_completions'):
+            top_skills_by_completions = SkillsDailyRollupAdminDashTable().get_top_skills_by_completion(
+                enterprise_id,
+                start_date,
+                end_date
             )
 
         response_data = {
-            "top_skills": top_skills.to_dict(orient="records"),
-            "top_skills_by_enrollments": top_skills_enrollments.to_dict(
-                orient="records"
-            ),
-            "top_skills_by_completions": top_skills_by_completions.to_dict(
-                orient="records"
-            ),
+            "top_skills": skills,
+            "top_skills_by_enrollments": top_skills_by_enrollments,
+            "top_skills_by_completions": top_skills_by_completions,
         }
 
         return Response(data=response_data, status=HTTP_200_OK)

{edx_enterprise_data-8.12.1 → edx_enterprise_data-8.13.0}/enterprise_data/tests/admin_analytics/mock_analytics_data.py

@@ -537,3 +537,59 @@ def engagements_csv_content():
         b'paul77@example.org,Synergized reciprocal encoding,2021-07-26,business-management,4.4\r\n'
         b'samanthaclarke@example.org,Synergized reciprocal encoding,2021-07-19,business-management,0.0\r\n'
     )
+
+
+TOP_SKILLS = [
+    {
+        "skill_name": "Python (Programming Language)",
+        "skill_type": "Specialized Skill",
+        "enrolls": 19027.0,
+        "completions": 3004.0
+    },
+    {
+        "skill_name": "Data Science",
+        "skill_type": "Specialized Skill",
+        "enrolls": 13756.0,
+        "completions": 1517.0
+    },
+    {
+        "skill_name": "Algorithms",
+        "skill_type": "Specialized Skill",
+        "enrolls": 12776.0,
+        "completions": 1640.0
+    },
+]
+TOP_SKILLS_BY_ENROLLMENTS = [
+    {
+        "skill_name": "Python (Programming Language)",
+        "subject_name": "business-management",
+        "count": 313.0
+    },
+    {
+        "skill_name": "Machine Learning",
+        "subject_name": "business-management",
+        "count": 442.0
+    },
+    {
+        "skill_name": "Computer Science",
+        "subject_name": "business-management",
+        "count": 39.0
+    },
+]
+TOP_SKILLS_BY_COMPLETIONS = [
+    {
+        "skill_name": "Python (Programming Language)",
+        "subject_name": "business-management",
+        "count": 21.0
+    },
+    {
+        "skill_name": "SQL (Programming Language)",
+        "subject_name": "business-management",
+        "count": 11.0
+    },
+    {
+        "skill_name": "Algorithms",
+        "subject_name": "business-management",
+        "count": 15.0
+    },
+]