Clean scores and return top 5 students

Problem Statement

You are provided with a DataFrame df_scores containing columns ['student_id', 'subject', 'score']. Some scores are missing (NaN). Your task is to process this data to identify the top 5 performing students based on their average score across all subjects.

Requirements:

1. Data Cleaning: Fill missing scores with the median score of that specific subject. If a subject has no available scores, assume the median is 0.
2. Aggregation: Calculate the average score for each student across all subjects.
3. Ranking: Identify the top 5 students by their calculated average score. In the event of a tie in average scores, use student_id (ascending) as a tie-breaker.
4. Output: Return a list of student_ids.

Function Signature

def get_top_students(df_scores: pd.DataFrame) -> list:
    # implementation here

Example

Input:

data = {
    'student_id': [1, 1, 2, 2, 3, 3],
    'subject': ['Math', 'Science', 'Math', 'Science', 'Math', 'Science'],
    'score': [80, None, 90, 85, None, 70]
}

Processing:

• Math median: (80+90)/2 = 85. Science median: (85+70)/2 = 77.5.
• Student 1 scores: Math 80, Science 77.5. Avg: 78.75.
• Student 2 scores: Math 90, Science 85. Avg: 87.5.
• Student 3 scores: Math 85, Science 70. Avg: 77.5.

Output: [2, 1, 3]

Constraints

• student_id is an integer.
• score is a float or NaN.
• Handle cases where a student might not have any scores.