# ===== Logistic Regression on Real Dataset =====

# 1) Load libraries
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix
import matplotlib.pyplot as plt
import seaborn as sns

# 2) Load Pima Indians Diabetes dataset (public URL)
url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv"
col_names = [
    "pregnancies", "glucose", "blood_pressure", "skin_thickness",
    "insulin", "bmi", "diabetes_pedigree", "age", "outcome"
]
data = pd.read_csv(url, names=col_names)

# Inspect dataset
print("First 5 rows:")
print(data.head(), "\n")

print("Dataset shape:", data.shape)

# 3) Separate features and labels
X = data.drop("outcome", axis=1)
y = data["outcome"]

# 4) Train/Test split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42, stratify=y
)

# 5) Feature scaling (important for many ml models)
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

# 6) Train Logistic Regression model
model = LogisticRegression(max_iter=1000)
model.fit(X_train_scaled, y_train)

# 7) Predictions
y_pred = model.predict(X_test_scaled)

# 8) Evaluation
print("Accuracy:", accuracy_score(y_test, y_pred))
print("\nClassification Report:\n", classification_report(y_test, y_pred))
print("\nConfusion Matrix:\n", confusion_matrix(y_test, y_pred))

# 9) Optional: Probability outputs
y_proba = model.predict_proba(X_test_scaled)[:, 1]
print("\nPredicted probabilities (first 10):\n", y_proba[:10])


# 10) Visualizing the Sigmoid (S-Curve)
# We calculate the "Logits" (z) which is the raw output: z = wX + b
# Then we plot the probability p = 1 / (1 + exp(-z))

# Calculate logits for the test set
logits = model.decision_function(X_test_scaled)

plt.figure(figsize=(10, 6))

# Plot the S-curve
# Sorting is necessary to make the line continuous
sorted_indices = np.argsort(logits)
plt.plot(logits[sorted_indices], y_proba[sorted_indices], color='blue', lw=3, label='Sigmoid Curve')

# Scatter plot of actual outcomes (0 or 1)
plt.scatter(logits, y_test, color='red', alpha=0.5, label='Actual Data Points')

plt.title('Logistic Regression S-Curve (Sigmoid Function)')
plt.xlabel('Logits (Linear combination of features)')
plt.ylabel('Probability of Diabetes')
plt.grid(True, linestyle='--', alpha=0.6)
plt.legend()
plt.show()