TASK : 02

Diabetes Predic on Model using Logis c Regression

Objec ve:

 To understand and implement data preprocessing techniques for a medical dataset.

 To perform feature engineering for improved model performance.

 To build and train a Logis c Regression model for diabetes predic on.

 To evaluate the performance of the trained model using appropriate metrics.

Tools/So ware Required:

 Python 3.x

 Pandas

 Scikit-learn (sklearn)

 NumPy (for numerical opera ons)

 Anaconda (recommended for environment management) or Google Colab

Dataset:

A sample diabetes dataset is provided in the code. This dataset includes the following
features:

 Pregnancies: Number of pregnancies.

 Glucose: Glucose level.

 BloodPressure: Blood pressure.

 SkinThickness: Skin thickness.

 Insulin: Insulin level.

 BMI: Body mass index.

 DiabetesPedigreeFunc on: 1 Diabetes pedigree func on. 2

 Age: Age.

 Outcome: Diabetes status (1: Diabetes, 0: No Diabetes).

Procedure:

1. Data Preprocessing:

1. Load the Data:

 o Create a Pandas DataFrame from the provided sample diabetes dataset.

o Print the original DataFrame to inspect the raw data.

2. Handle Missing Values:

o Iden fy and handle missing values in the 'BMI' column using SimpleImputer
with the mean strategy.

o Print the DataFrame a er imputa on to verify the changes.

3. Scale Numerical Features:

o Scale the numerical features ('Pregnancies', 'Glucose', 'BloodPressure',

'SkinThickness', 'Insulin', 'BMI', 'DiabetesPedigreeFunc on', 'Age') using
StandardScaler.

o Print the DataFrame a er scaling to observe the transformed data.

2. Feature Engineering:

1. Create Interac on Feature:

o Create a new feature 'Glucose_BMI' by mul plying the 'Glucose' and 'BMI'
columns.

o Print the DataFrame a er adding the new feature.

3. Machine Learning Model Building and Evalua on:

1. Prepare Data for Model:

o Define the features (X) by dropping the 'Outcome' column.

o Define the target variable (y) as the 'Outcome' column.

2. Split Data:

o Use train_test_split to split the dataset into training and tes ng sets (80%
training, 20% tes ng) with random_state=42 for reproducibility.

o Print the shapes and contents of the training and tes ng sets.

3. Train the Model:

o Create a Logis cRegression model.

o Train the model using the training data (X_train, y_train).

4. Make Predic ons:

o Use the trained model to make predic ons on the tes ng data (X_test).
 o Print the predic ons (y_pred).

5. Evaluate the Model:

o Calculate the accuracy of the model using accuracy_score.

o Generate a classifica on report using classifica on_report to assess precision,

recall, and F1-score.

o Print the accuracy and the classifica on report.

Program :

Python

import pandas as pd

import numpy as np

from sklearn.model_selec on import train_test_split

from sklearn.preprocessing import StandardScaler

from sklearn.impute import SimpleImputer

from sklearn.linear_model import Logis cRegression

from sklearn.metrics import accuracy_score, classifica on_report

# 1. Sample Diabetes Dataset

data = {

'Pregnancies': [6, 1, 8, 1, 0, 5, 3, 10, 2, 4],

'Glucose': [148, 85, 183, 89, 137, 116, 78, 115, 197, 125],

'BloodPressure': [72, 66, 64, 66, 40, 74, 50, 0, 70, 96],

'SkinThickness': [35, 29, 0, 23, 35, 0, 32, 0, 45, 0],

'Insulin': [0, 0, 0, 94, 168, 0, 88, 0, 543, 0],

'BMI': [33.6, 26.6, 23.3, 28.1, 43.1, 25.6, 31.0, np.nan, 30.5, 0.0],

'DiabetesPedigreeFunc on': [0.627, 0.351, 0.672, 0.167, 2.288, 0.201, 0.248, 0.134, 0.158,
0.177],

'Age': [50, 31, 32, 21, 33, 30, 26, 29, 53, 41],

'Outcome': [1, 0, 1, 0, 1, 0, 0, 1, 1, 0]

}
df = pd.DataFrame(data)

print("Original DataFrame:\n", df)

# 2. Data Preprocessing

imputer = SimpleImputer(strategy='mean')

df['BMI'] = imputer.fit_transform(df[['BMI']])

print("\nDataFrame a er Impu ng BMI:\n", df)

scaler = StandardScaler()

numerical_cols = ['Pregnancies', 'Glucose', 'BloodPressure', 'SkinThickness', 'Insulin', 'BMI',

'DiabetesPedigreeFunc on', 'Age']

df[numerical_cols] = scaler.fit_transform(df[numerical_cols])

print("\nDataFrame a er Scaling:\n", df)

# 3. Feature Engineering

df['Glucose_BMI'] = df['Glucose'] * df['BMI']

print("\nDataFrame a er Feature Engineering:\n", df)

# 4. Model Building and Evalua on

X = df.drop('Outcome', axis=1)

y = df['Outcome']

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

print("\nTraining Data (X_train):\n", X_train)

print("\nTes ng Data (X_test):\n", X_test)

print("\nTraining Labels (y_train):\n", y_train)

print("\nTes ng Labels (y_test):\n", y_test)

model = Logis cRegression(random_state=42)

model.fit(X_train, y_train)

y_pred = model.predict(X_test)

print("\nPredic ons (y_pred):\n", y_pred)

accuracy = accuracy_score(y_test, y_pred)

report = classifica on_report(y_test, y_pred)

print("\nAccuracy:", accuracy)

print("\nClassifica on Report:\n", report)

Output :

Original DataFrame:

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

0 6 148 72 35 0 33.6

1 1 85 66 29 0 26.6

2 8 183 64 0 0 23.3

3 1 89 66 23 94 28.1

4 0 137 40 35 168 43.1

5 5 116 74 0 0 25.6

6 3 78 50 32 88 31.0

7 10 115 0 0 0 NaN

8 2 197 70 45 543 30.5

9 4 125 96 0 0 0.0
 DiabetesPedigreeFunc on Age Outcome

0 0.627 50 1

1 0.351 31 0

2 0.672 32 1

3 0.167 21 0

4 2.288 33 1

5 0.201 30 0

6 0.248 26 0

7 0.134 29 1

8 0.158 53 1

9 0.177 41 0

DataFrame a er Impu ng BMI:

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

0 6 148 72 35 0 33.600000

1 1 85 66 29 0 26.600000

2 8 183 64 0 0 23.300000

3 1 89 66 23 94 28.100000

4 0 137 40 35 168 43.100000

5 5 116 74 0 0 25.600000

6 3 78 50 32 88 31.000000

7 10 115 0 0 0 26.866667

8 2 197 70 45 543 30.500000

9 4 125 96 0 0 0.000000
 DiabetesPedigreeFunc on Age Outcome

0 0.627 50 1

1 0.351 31 0

2 0.672 32 1

3 0.167 21 0

4 2.288 33 1

5 0.201 30 0

6 0.248 26 0

7 0.134 29 1

8 0.158 53 1

9 0.177 41 0

DataFrame a er Scaling:

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

0 0.645497 0.544471 0.501265 0.885345 -0.553913 0.648853

1 -0.968246 -1.112615 0.254741 0.533552 -0.553913 -0.025697

2 1.290994 1.465074 0.172566 -1.166779 -0.553913 -0.343699

3 -0.968246 -1.007403 0.254741 0.181760 0.029153 0.118849

4 -1.290994 0.255139 -0.813528 0.885345 0.488163 1.564314

5 0.322749 -0.297223 0.583439 -1.166779 -0.553913 -0.122061

6 -0.322749 -1.296735 -0.402655 0.709449 -0.008064 0.398306

7 1.936492 -0.323526 -2.457018 -1.166779 -0.553913 0.000000

8 -0.645497 1.833316 0.419090 1.471666 2.814225 0.350124

9 0.000000 -0.060497 1.487359 -1.166779 -0.553913 -2.588989

 DiabetesPedigreeFunc on Age Outcome

0 0.200095 1.579674 1

1 -0.242777 -0.369274 0

2 0.272302 -0.266698 1

3 -0.538025 -1.395037 0

4 2.865348 -0.164122 1

5 -0.483468 -0.471851 0

6 -0.408052 -0.882156 0

7 -0.590977 -0.574427 1

8 -0.552466 1.887403 1

9 -0.521979 0.656488 0

DataFrame a er Feature Engineering:

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

0 0.645497 0.544471 0.501265 0.885345 -0.553913 0.648853

1 -0.968246 -1.112615 0.254741 0.533552 -0.553913 -0.025697

2 1.290994 1.465074 0.172566 -1.166779 -0.553913 -0.343699

3 -0.968246 -1.007403 0.254741 0.181760 0.029153 0.118849

4 -1.290994 0.255139 -0.813528 0.885345 0.488163 1.564314

5 0.322749 -0.297223 0.583439 -1.166779 -0.553913 -0.122061

6 -0.322749 -1.296735 -0.402655 0.709449 -0.008064 0.398306

7 1.936492 -0.323526 -2.457018 -1.166779 -0.553913 0.000000

8 -0.645497 1.833316 0.419090 1.471666 2.814225 0.350124

9 0.000000 -0.060497 1.487359 -1.166779 -0.553913 -2.588989

 DiabetesPedigreeFunc on Age Outcome Glucose_BMI

0 0.200095 1.579674 1 0.353282

1 -0.242777 -0.369274 0 0.028591

2 0.272302 -0.266698 1 -0.503545

3 -0.538025 -1.395037 0 -0.119729

4 2.865348 -0.164122 1 0.399117

5 -0.483468 -0.471851 0 0.036280

6 -0.408052 -0.882156 0 -0.516497

7 -0.590977 -0.574427 1 -0.000000

8 -0.552466 1.887403 1 0.641887

9 -0.521979 0.656488 0 0.156625

Training Data (X_train):

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

5 0.322749 -0.297223 0.583439 -1.166779 -0.553913 -0.122061

0 0.645497 0.544471 0.501265 0.885345 -0.553913 0.648853

7 1.936492 -0.323526 -2.457018 -1.166779 -0.553913 0.000000

2 1.290994 1.465074 0.172566 -1.166779 -0.553913 -0.343699

9 0.000000 -0.060497 1.487359 -1.166779 -0.553913 -2.588989

4 -1.290994 0.255139 -0.813528 0.885345 0.488163 1.564314

3 -0.968246 -1.007403 0.254741 0.181760 0.029153 0.118849

6 -0.322749 -1.296735 -0.402655 0.709449 -0.008064 0.398306

DiabetesPedigreeFunc on Age Glucose_BMI

5 -0.483468 -0.471851 0.036280

0 0.200095 1.579674 0.353282

7 -0.590977 -0.574427 -0.000000

2 0.272302 -0.266698 -0.503545

9 -0.521979 0.656488 0.156625

4 2.865348 -0.164122 0.399117

3 -0.538025 -1.395037 -0.119729

6 -0.408052 -0.882156 -0.516497

Tes ng Data (X_test):

Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \

8 -0.645497 1.833316 0.419090 1.471666 2.814225 0.350124

1 -0.968246 -1.112615 0.254741 0.533552 -0.553913 -0.025697

DiabetesPedigreeFunc on Age Glucose_BMI

8 -0.552466 1.887403 0.641887

1 -0.242777 -0.369274 0.028591

Training Labels (y_train):

5 0

0 1

7 1

2 1

9 0

4 1

3 0

6 0

Name: Outcome, dtype: int64

Tes ng Labels (y_test):

8 1
1 0

Name: Outcome, dtype: int64

Predic ons (y_pred):

[1 0]

Accuracy: 1.0

Results:

 Record the original DataFrame.

 Record the DataFrame a er imputa on and scaling.

 Record the DataFrame a er feature engineering.

 Record the shapes and content of the training and tes ng sets.

 Record the predic ons made by the model.

 Record the accuracy and classifica on report of the model.