Grade Analytics - Machine Learning Project

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

Data Preprocessing

df=pd.read_csv("Grades.csv")
df
Seat No. PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 EL-102 EE-119 ... CS-312 CS-317 CS-403 CS-421 CS-406 CS-414 CS-419 CS-423 CS-412 CGPA
0 CS-97001 B- D+ C- C C- D+ D C- B- ... C- C- C- C- A- A C- B A- 2.205
1 CS-97002 A D D+ D B- C D A D+ ... D+ D C D A- B- C C B 2.008
2 CS-97003 A B A B- B+ A B- B+ A- ... B B A C A A A A- A 3.608
3 CS-97004 D C+ D+ D D A- D+ C- D ... D+ C D+ C- B- B C+ C+ C+ 1.906
4 CS-97005 A- A- A- B+ A A A- B+ A ... B- B+ B+ B- A- A A- A- A 3.448
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
566 CS-97567 B A A A- A+ A A- A- A+ ... A- A- A A A B+ B+ B A 3.798
567 CS-97568 A+ A A A A A A A- A ... B+ B+ A A A- B A- C A- 3.772
568 CS-97569 B A A- B+ A A A A A ... A- B A B+ A C B+ A- A- 3.470
569 CS-97570 A B+ D A D D+ B- C- B- ... D B B C- D C B B- C 2.193
570 CS-97571 C D D C C D+ B C+ C ... C+ C B- D F C- B+ D C- 1.753

571 rows × 43 columns

df.columns
Index(['Seat No.', 'PH-121', 'HS-101', 'CY-105', 'HS-105/12', 'MT-111',
       'CS-105', 'CS-106', 'EL-102', 'EE-119', 'ME-107', 'CS-107', 'HS-205/20',
       'MT-222', 'EE-222', 'MT-224', 'CS-210', 'CS-211', 'CS-203', 'CS-214',
       'EE-217', 'CS-212', 'CS-215', 'MT-331', 'EF-303', 'HS-304', 'CS-301',
       'CS-302', 'TC-383', 'MT-442', 'EL-332', 'CS-318', 'CS-306', 'CS-312',
       'CS-317', 'CS-403', 'CS-421', 'CS-406', 'CS-414', 'CS-419', 'CS-423',
       'CS-412', 'CGPA'],
      dtype='object')
drop_column=df.columns[:-1][8:]
print(drop_column)
df.drop(columns=drop_column,inplace=True)
seats = df['Seat No.'].to_dict()
df.drop(columns=['Seat No.'], inplace=True)
Index(['EL-102', 'EE-119', 'ME-107', 'CS-107', 'HS-205/20', 'MT-222', 'EE-222',
       'MT-224', 'CS-210', 'CS-211', 'CS-203', 'CS-214', 'EE-217', 'CS-212',
       'CS-215', 'MT-331', 'EF-303', 'HS-304', 'CS-301', 'CS-302', 'TC-383',
       'MT-442', 'EL-332', 'CS-318', 'CS-306', 'CS-312', 'CS-317', 'CS-403',
       'CS-421', 'CS-406', 'CS-414', 'CS-419', 'CS-423', 'CS-412'],
      dtype='object')
df.shape
(571, 8)
df.head()
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
0 B- D+ C- C C- D+ D 2.205
1 A D D+ D B- C D 2.008
2 A B A B- B+ A B- 3.608
3 D C+ D+ D D A- D+ 1.906
4 A- A- A- B+ A A A- 3.448 
df.tail()
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
566 B A A A- A+ A A- 3.798
567 A+ A A A A A A 3.772
568 B A A- B+ A A A 3.470
569 A B+ D A D D+ B- 2.193
570 C D D C C D+ B 1.753 
df.describe(include='all')
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
count 571 571 570 570 569 571 569 571.000000
unique 13 12 13 13 13 11 13 NaN
top A- A- A A A- A A- NaN
freq 112 82 177 96 105 151 116 NaN
mean NaN NaN NaN NaN NaN NaN NaN 2.954888
std NaN NaN NaN NaN NaN NaN NaN 0.620552
min NaN NaN NaN NaN NaN NaN NaN 0.800000
25% NaN NaN NaN NaN NaN NaN NaN 2.538000
50% NaN NaN NaN NaN NaN NaN NaN 3.029000
75% NaN NaN NaN NaN NaN NaN NaN 3.451000
max NaN NaN NaN NaN NaN NaN NaN 3.985000 
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 571 entries, 0 to 570
Data columns (total 8 columns):
 #   Column     Non-Null Count  Dtype  
---  ------     --------------  -----  
 0   PH-121     571 non-null    object 
 1   HS-101     571 non-null    object 
 2   CY-105     570 non-null    object 
 3   HS-105/12  570 non-null    object 
 4   MT-111     569 non-null    object 
 5   CS-105     571 non-null    object 
 6   CS-106     569 non-null    object 
 7   CGPA       571 non-null    float64
dtypes: float64(1), object(7)
memory usage: 35.8+ KB
df.isnull().sum()
PH-121       0
HS-101       0
CY-105       1
HS-105/12    1
MT-111       2
CS-105       0
CS-106       2
CGPA         0
dtype: int64
for column in df.columns:
  df[column].fillna(df[column].mode()[0], inplace=True)
df.isnull().sum()
PH-121       0
HS-101       0
CY-105       0
HS-105/12    0
MT-111       0
CS-105       0
CS-106       0
CGPA         0
dtype: int64

Exploratory Data Analysis

grade_list = ['A+', 'A', 'A-', 'B+', 'B', 'B-', 'C+', 'C', 'C-', 'D+', 'D', 'F', 'WU']

Countplot

for i in df.columns[:-1]:
    sns.countplot(x=i, data=df, palette='Set2', legend=False, hue=i, order=grade_list)
    plt.title(f'Countplot for {i}')
    plt.show()

Scatter/Strip Plot

sns.stripplot(data=df, x=df['PH-121'], y=df['HS-101'], color='green', order=grade_list)
plt.xlabel('PH-121')
plt.ylabel('HS-101')
plt.title('Strip Plot of PH-121 vs HS-101')
plt.show()

sns.stripplot(data=df, x=df['HS-101'], y=df['CGPA'], color='green', order=grade_list)
plt.show()
sns.stripplot(data=df, x=df['CS-105'], y=df['CGPA'], color='green', order=grade_list)
plt.show()
sns.stripplot(data=df, x=df['CS-106'], y=df['CGPA'], color='green', order=grade_list)
plt.show()

Student Performance

roll = input("Enter roll number: ")
roll = [i for i in seats if seats[i]==roll]

if len(roll)==1:
    roll = roll[0]
    rolldf = df2.loc[roll][:-1]
    ax = rolldf.plot(kind='bar', color=['blue']*7+['green'])
    bx = ax.bar_label(ax.containers[0], labels=df.loc[roll][:-1].tolist())
    #cx = df2.loc[roll][-2:-1].plot(kind='bar')
else:
    print("Roll number not found.")
Enter roll number:  CS-97002

Box and Violin Plots

df2 = df.copy()
df2
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
0 B- D+ C- C C- D+ D 2.205
1 A D D+ D B- C D 2.008
2 A B A B- B+ A B- 3.608
3 D C+ D+ D D A- D+ 1.906
4 A- A- A- B+ A A A- 3.448
... ... ... ... ... ... ... ... ...
566 B A A A- A+ A A- 3.798
567 A+ A A A A A A 3.772
568 B A A- B+ A A A 3.470
569 A B+ D A D D+ B- 2.193
570 C D D C C D+ B 1.753

571 rows × 8 columns

print(grade_list)
grade_mapping = {}
for i in range(len(grade_list)):
    grade_mapping[grade_list[i]] = len(grade_list)-i-1
print(grade_mapping)
['A+', 'A', 'A-', 'B+', 'B', 'B-', 'C+', 'C', 'C-', 'D+', 'D', 'F', 'WU']
{'A+': 12, 'A': 11, 'A-': 10, 'B+': 9, 'B': 8, 'B-': 7, 'C+': 6, 'C': 5, 'C-': 4, 'D+': 3, 'D': 2, 'F': 1, 'WU': 0}
for col in df2.columns[:-1]:
  df2[col] = df2[col].map(grade_mapping)
df2
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
0 7 3 4 5 4 3 2 2.205
1 11 2 3 2 7 5 2 2.008
2 11 8 11 7 9 11 7 3.608
3 2 6 3 2 2 10 3 1.906
4 10 10 10 9 11 11 10 3.448
... ... ... ... ... ... ... ... ...
566 8 11 11 10 12 11 10 3.798
567 12 11 11 11 11 11 11 3.772
568 8 11 10 9 11 11 11 3.470
569 11 9 2 11 2 3 7 2.193
570 5 2 2 5 5 3 8 1.753

571 rows × 8 columns

for i in df2.columns:
    sns.boxplot(x=i, data=df2, orient='h')
    plt.show()

for i in df2.columns:
    sns.violinplot(x=i, data=df2)
    plt.show()

Correlation Matrix

df2.corr()
PH-121 HS-101 CY-105 HS-105/12 MT-111 CS-105 CS-106 CGPA
PH-121 1.000000 0.408386 0.626572 0.549676 0.494110 0.672645 0.303391 0.680941
HS-101 0.408386 1.000000 0.362929 0.490250 0.463471 0.444827 0.362976 0.560989
CY-105 0.626572 0.362929 1.000000 0.583896 0.576232 0.618138 0.403414 0.685564
HS-105/12 0.549676 0.490250 0.583896 1.000000 0.526262 0.540200 0.526946 0.657252
MT-111 0.494110 0.463471 0.576232 0.526262 1.000000 0.434376 0.543269 0.749554
CS-105 0.672645 0.444827 0.618138 0.540200 0.434376 1.000000 0.302261 0.644594
CS-106 0.303391 0.362976 0.403414 0.526946 0.543269 0.302261 1.000000 0.573354
CGPA 0.680941 0.560989 0.685564 0.657252 0.749554 0.644594 0.573354 1.000000 
plt.figure(figsize=(6,5))
sns.heatmap(df2.corr(), annot=True, cmap='coolwarm')
<Axes: >

Clustering

from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler

X = df2.drop(['CGPA'], axis=1)
scaler = StandardScaler()
X_std = scaler.fit_transform(X)

num_clusters = 3

kmeans = KMeans(n_clusters=num_clusters, random_state=42, n_init='auto')
df2['Cluster'] = kmeans.fit_predict(X)

plt.figure(figsize=(10, 6))
plt.scatter(df2['CGPA'], df2['CS-106'], c=df2['Cluster'], cmap='viridis')
plt.title('K-Means Clustering of Students')
plt.xlabel('CGPA')
plt.ylabel('CS-106')
plt.show()

CGPA Prediction (Using 4 Subjects)

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from random import shuffle

cols = df2.columns[:-2].tolist()
print(f'Available Subjects: \t\t\t {cols}')
shuffle(cols)
cols = cols[0:4]
print(f'Selected Subjects: \t\t\t {cols}')

X = df2[cols]
y = df2['CGPA']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

model = LinearRegression()

model.fit(X_train, y_train)
y_pred = model.predict(X_test)

mse = mean_squared_error(y_test, y_pred)
score = model.score(X_test, y_test)

print(f'Mean Squared Error: \t\t\t {mse}')
print(f'Coefficient of determination (R^2): \t {score}\n')

output = pd.DataFrame({'Actual CGPA': y_test, 'Predicted CGPA': y_pred})
plt.figure(figsize=(10, 6))
sns.regplot(x='Actual CGPA', y='Predicted CGPA', data=output, ci=95)
plt.title('Actual vs Predicted CGPA\n')
plt.xlabel('Actual CGPA')
plt.ylabel('Predicted CGPA')
plt.show()
Available Subjects:              ['PH-121', 'HS-101', 'CY-105', 'HS-105/12', 'MT-111', 'CS-105', 'CS-106']
Selected Subjects:           ['HS-105/12', 'HS-101', 'CS-106', 'CS-105']
Mean Squared Error:              0.12502848801062927
Coefficient of determination (R^2):      0.6242993778884239


!quarto render Grade_Analytics.ipynb
pandoc 
  to: html
  output-file: Grade_Analytics.html
  standalone: true
  section-divs: true
  html-math-method: mathjax
  wrap: none
  default-image-extension: png
  
metadata
  document-css: false
  link-citations: true
  date-format: long
  lang: en
  title: Grade Analytics - Machine Learning Project
  
Output created: Grade_Analytics.html