import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import confusion_matrix

# 1. Generate Synthetic Data (NumPy)
np.random.seed(42)
n_samples = 200
temp = np.random.normal(300, 20, n_samples)
pressure = np.random.normal(50, 10, n_samples)

# Define stability condition: unstable if temp > 330 or pressure > 65
stability = ((temp > 330) | (pressure > 65)).astype(int)

# 2. Structure Data (Pandas)
df = pd.DataFrame({
  'Temperature': temp,
  'Pressure': pressure,
  'Unstable': stability
})

print("First 5 samples of the dataset:")
print(df.head())

# 3. Model Training (Scikit-Learn)
X = df[['Temperature', 'Pressure']]
y = df['Unstable']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25)

model = RandomForestClassifier(n_estimators=50)
model.fit(X_train, y_train)

# 4. Comprehensive Visualization (Matplotlib/Seaborn)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(15, 6))

# Plot A: The raw data distribution
sns.scatterplot(data=df, x='Temperature', y='Pressure', hue='Unstable', ax=ax1, palette='coolwarm')
ax1.set_title("Reactor States: Temperature vs Pressure")

# Plot B: The Model Performance (Confusion Matrix)
y_pred = model.predict(X_test)
cm = confusion_matrix(y_test, y_pred)
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', ax=ax2)
ax2.set_xlabel('Predicted')
ax2.set_ylabel('Actual')
ax2.set_title("Model Accuracy: Confusion Matrix")

plt.tight_layout()
plt.show()

score = model.score(X_test, y_test)
print(f"\nModel Accuracy on Test Set: {score*100:.1f}%")