Building a Sex Classifier from Workclass Categorical Features

import pandas as pd
import numpy as np

# Load the dataset
df = pd.read_csv('your_data.csv')

# Remove rows with missing values
df = df.dropna()

# Define the categorical columns to be encoded
cat_cols = ['workclass', 'marital-status', 'occupation', 'relationship', 
             'race', 'sex', 'native-country']

# Encode sex column
df["Value"] = np.where((df["sex"] == 'Female'), 0, 1)

# Encode categorical columns
data = df.copy()
for col in cat_cols:
    data = pd.get_dummies(data, columns=[col], prefix=[col])

# Define the workclass columns to be selected
workclass_cols = [i for i in data.columns if 'workclass' in i]

# Create a new dataframe with only the workclass columns and target variable
X_workclass = data[workclass_cols]
y_workclass = df['Value']

# Print the shape of X_workclass and y_workclass
print(f"Shape of X_workclass: {X_workclass.shape}")
print(f"Shape of y_workclass: {y_workclass.shape}")

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score, classification_report

# Split the data into training and testing sets
X_train_workclass, X_test_workclass, y_train_workclass, y_test_workclass = train_test_split(X_workclass, y_workclass, test_size=0.2, random_state=42)

# Create a logistic regression model
model_workclass = LogisticRegression()

# Train the model on the training data
model_workclass.fit(X_train_workclass, y_train_workclass)

# Make predictions on the testing data
y_pred_workclass = model_workclass.predict(X_test_workclass)

# Evaluate the model's performance
print(f"Accuracy: {accuracy_score(y_test_workclass, y_pred_workclass)}")
print(f"Classification Report:\n{classification_report(y_test_workclass, y_pred_workclass)}")

# Define all feature columns (excluding target variable)
feature_cols = [col for col in data.columns if col != 'Value']

# Create a new dataframe with only the feature columns and target variable
X_all_features = data[feature_cols]
y_all_features = df['Value']

# Split the data into training and testing sets
X_train_all, X_test_all, y_train_all, y_test_all = train_test_split(X_all_features, y_all_features, test_size=0.2, random_state=42)

# Create a logistic regression model
model_all_features = LogisticRegression()

# Train the model on the training data
model_all_features.fit(X_train_all, y_train_all)

# Make predictions on the testing data
y_pred_all_features = model_all_features.predict(X_test_all)

# Evaluate the model's performance
print(f"Accuracy: {accuracy_score(y_test_all, y_pred_all_features)}")
print(f"Classification Report:\n{classification_report(y_test_all, y_pred_all_features)}")

from sklearn.ensemble import BaggingClassifier, GradientBoostingClassifier

# Define a bagging classifier with logistic regression as base estimator
bagging_classifier = BaggingClassifier(base_estimator=LogisticRegression(), n_estimators=100)

# Train the bagging classifier on the training data
bagging_classifier.fit(X_train_workclass, y_train_workclass)

# Make predictions on the testing data
y_pred_bagging = bagging_classifier.predict(X_test_workclass)

# Define a gradient boosting classifier with logistic regression as base estimator
gb_classifier = GradientBoostingClassifier(base_estimator=LogisticRegression(), n_estimators=100)

# Train the gradient boosting classifier on the training data
gb_classifier.fit(X_train_all, y_train_all)

# Make predictions on the testing data
y_pred_gb = gb_classifier.predict(X_test_all)

# Evaluate the performance of both models
print(f"Bagging Accuracy: {accuracy_score(y_test_workclass, y_pred_bagging)}")
print(f"Gradient Boosting Accuracy: {accuracy_score(y_test_all, y_pred_gb)}")