CODEWITHSUNDEEP
pythonnumpy
TheUnreal
TheUnreal

Reputation: 24492

NumPy - index of highest average

I have the following values:

grade_list = [[99 73 97 98] [98 71 70 99]]
student_ids = [521 597 624 100]

Each array in the grade_list represents a list of grades for each student (student #521 grades are 99, 98, student #597 grades are 73, 71, and so on..)

My goal is to return the student ID of the one with the highest average grade (should be 521 since the average is 98.5 which is the highest)

I have tried this:

def find_student_with_max_avg(grade_list, student_ids):
    mean_grades = np.mean(grade_list, axis=1)
    best_student_index = student_ids[mean_grades.argmax()]
    return best_student_index

which returned 521, but I have tried to play with the grade_list and I see that even when I set this student grades to 0, it still return 521`.

not sure how to get the student ID of the best student, any idea?

Upvotes: 3

Views: 186

Answers (3)

alec_djinn
alec_djinn

Reputation: 10819

I would use Counter for this.

from collections import Counter
import numpy as np


grade_list = [[99, 73, 97, 98], [98, 71, 70, 99]] 
student_ids = [521, 597, 624, 100]                                     

c = Counter() 
for i,n in enumerate(student_ids): 
    c.update({n:np.mean([item[i] for item in grade_list])}) 
                                                                   

c.most_common()                                                        
[(521, 98.5), (100, 98.5), (624, 83.5), (597, 72.0)]

As you see, you get 2 best students in this case.

In [17]: best_score = c.most_common(1)[0][1]                                    

In [18]: best_score                                                             
Out[18]: 98.5

In [19]: best_students = [k for k,v in c.items() if v == best_score]            

In [20]: best_students                                                          
Out[20]: [521, 100]

Upvotes: 0

sushanth
sushanth

Reputation: 8302

try this,

import numpy as np

grade_list = [[99,73,97,98] ,[98,71,70,99]]
student_ids = [521,597,624,100]

# axis 0 is row-wise, and 1 is column-wise
student_ids[np.array(grade_list).mean(axis=0).argmax()]

Upvotes: 2

Feodoran
Feodoran

Reputation: 1822

You are taking the mean along the wrong axis:

>>> np.mean(grade_list, axis=1)
array([91.75, 84.5 ])

Change the axis, and check that you get as many mean values as there are student IDs.

>>> np.mean(grade_list, axis=0)
array([98.5, 72. , 83.5, 98.5])

Upvotes: 1

Related Questions