graph.py

import pandas as pd
import matplotlib.pyplot as plt
import joblib

MODEL_FILE = "password_strength_model.pkl"
DATASET_FILE = "password_dataset.csv"

df = pd.read_csv(DATASET_FILE)
model = joblib.load(MODEL_FILE)

X = df[["length", "has_upper", "has_lower", "has_digit", "has_symbol", "entropy"]]

y_pred = model.predict(X)

plt.figure(figsize=(7,5))
colors = ['red' if score < 0.4 else 'orange' if score < 0.7 else 'green' for score in y_pred]  
plt.scatter(df["entropy"], df["length"], c=colors, label=colors, alpha=0.7)
plt.xlabel("Entropy (bits)")
plt.ylabel("Password Length")
plt.title("Password Complexity vs Model Predicted Strength")
plt.show()

plt.figure(figsize=(7,5))
plt.scatter(df["entropy"], y_pred, alpha=0.7, c='blue')
plt.xlabel("Entropy (bits)")
plt.ylabel("Predicted Strength Score (0=weak, 1=strong)")
plt.title("AI Predicted Strength vs Password Entropy")
plt.ylim(0,1)
plt.show()

classical_time_sec = 2**df["entropy"] / 1e9
quantum_time_sec = 2**(df["entropy"]/2) / 1e9

plt.figure(figsize=(7,5))
plt.scatter(df["entropy"], classical_time_sec, label="Classical", alpha=0.6)
plt.scatter(df["entropy"], quantum_time_sec, label="Quantum", alpha=0.6)
plt.yscale("log")
plt.xlabel("Entropy (bits)")
plt.ylabel("Crack Time (seconds, log scale)")
plt.title("Classical vs Quantum Crack Time")
plt.legend()
plt.show()