main.py

# Import necessary libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC
from sklearn.metrics import (
    confusion_matrix, accuracy_score, precision_score, 
    recall_score, f1_score, roc_auc_score, classification_report
)

# ------------------------- 1. Load and Explore Data -------------------------

# Load the dataset
df = pd.read_csv('./admit-train.csv')  # Replace with your file path
print("Dataset Preview:")
print(df.head())

# Check for missing values
print("\nMissing Values:")
print(df.isnull().sum())

# Describe the dataset
print("\nStatistical Summary:")
print(df.describe())

# Check data types
print("\nData Types:")
print(df.dtypes)

# ------------------------- 2. Preprocess the Data -------------------------

# Convert 'rank' to categorical
df['rank'] = df['rank'].astype('category')

# Separate features and target
X = df[['gre', 'gpa', 'rank']]
y = df['admit']

# Preprocessing: Scale numerical variables, one-hot encode 'rank'
preprocessor = ColumnTransformer(
    transformers=[
        ('num', StandardScaler(), ['gre', 'gpa']),
        ('cat', OneHotEncoder(drop='first'), ['rank'])
    ]
)

# Split into training and test sets (80/20 split)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)
print("\nTraining and Test Set Sizes:")
print(f"Training set: {X_train.shape}, Test set: {X_test.shape}")

# Preprocess training and test data
X_train = preprocessor.fit_transform(X_train)
X_test = preprocessor.transform(X_test)

# ------------------------- 3. Train Models -------------------------

# Logistic Regression
logreg = LogisticRegression()
logreg.fit(X_train, y_train)
y_pred_logreg = logreg.predict(X_test)

# Random Forest
rf = RandomForestClassifier(random_state=42)
rf.fit(X_train, y_train)
y_pred_rf = rf.predict(X_test)

# Support Vector Machine (SVM)
svm = SVC(probability=True, random_state=42)
svm.fit(X_train, y_train)
y_pred_svm = svm.predict(X_test)

# ------------------------- 4. Evaluate Models -------------------------

# Define a function for metrics
def evaluate_model(y_true, y_pred, model_name):
    print(f"\nEvaluation Metrics for {model_name}:")
    print("Accuracy:", accuracy_score(y_true, y_pred))
    print("Precision:", precision_score(y_true, y_pred))
    print("Recall:", recall_score(y_true, y_pred))
    print("F1 Score:", f1_score(y_true, y_pred))
    print("ROC-AUC Score:", roc_auc_score(y_true, y_pred))
    print("Confusion Matrix:\n", confusion_matrix(y_true, y_pred))

# Evaluate each model
evaluate_model(y_test, y_pred_logreg, "Logistic Regression")
evaluate_model(y_test, y_pred_rf, "Random Forest")
evaluate_model(y_test, y_pred_svm, "SVM")

# Compare accuracies
models = ['Logistic Regression', 'Random Forest', 'SVM']
accuracies = [
    accuracy_score(y_test, y_pred_logreg),
    accuracy_score(y_test, y_pred_rf),
    accuracy_score(y_test, y_pred_svm)
]
comparison = pd.DataFrame({'Model': models, 'Accuracy': accuracies})
print("\nModel Comparison:")
print(comparison)

# Plot accuracy comparison
comparison.plot(x='Model', y='Accuracy', kind='bar', legend=False)
plt.title("Model Accuracy Comparison")
plt.ylabel("Accuracy")
plt.show()

# ------------------------- 5. Predict on New Data -------------------------

# Load new test data
new_data = pd.read_csv('./admit-test.csv')  # Replace with test file path
new_data['rank'] = new_data['rank'].astype('category')  # Ensure rank is categorical

# Preprocess new data
new_data_preprocessed = preprocessor.transform(new_data)

# Predict probabilities and final classes using the best model (Random Forest here)
new_data['admit_prob'] = rf.predict_proba(new_data_preprocessed)[:, 1]
new_data['admit_pred'] = (new_data['admit_prob'] >= 0.5).astype(int)

# Save predictions
new_data[['admit_prob', 'admit_pred']].to_csv("predictions.csv", index=False)
print("\nPredictions saved to 'predictions.csv'.")

# ------------------------- 6. Summary -------------------------

print("\nTask Completed Successfully!")
print("1. Models trained: Logistic Regression, Random Forest, and SVM.")
print("2. Metrics evaluated: Accuracy, Precision, Recall, F1-Score, ROC-AUC.")
print("3. Predictions made and saved for new test data.")