Decision Tree Classification
সম্পূর্ণ বাংলা লেকচার নোট

# প্রয়োজনীয় লাইব্রেরি
import pandas as pd
import numpy as np
from sklearn.tree import DecisionTreeClassifier, export_text, plot_tree
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
import matplotlib.pyplot as plt

# Dataset তৈরি
data = {
    'study_hours':    [1,2,2,3,4,4,5,5,6,6,7,7,8,8,9,3,5,7,2,8],
    'attendance':     [40,50,45,55,60,70,65,75,80,85,90,88,95,92,98,48,72,88,42,96],
    'previous_marks': [30,40,35,45,50,55,60,58,65,70,75,72,80,78,88,38,62,79,33,84],
    'pass_fail':      [0,0,0,0,0,1,0,1,1,1,1,1,1,1,1,0,1,1,0,1]
}

df = pd.DataFrame(data)
print("Dataset তৈরি হয়েছে!")
print(df.head(8))

# Features ও Target
X = df[['study_hours', 'attendance', 'previous_marks']]
y = df['pass_fail']

# Train-Test Split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)
print("Training:", X_train.shape, "| Testing:", X_test.shape)

# Decision Tree তৈরি (max_depth=3 দিয়ে Overfitting রোধ)
dt_model = DecisionTreeClassifier(
    max_depth=3,
    criterion='gini',   # Gini Impurity ব্যবহার
    random_state=42
)
dt_model.fit(X_train, y_train)
print("✅ Decision Tree Training সম্পন্ন!")
print(f"Tree-এর গভীরতা: {dt_model.get_depth()}")
print(f"Leaf Node সংখ্যা: {dt_model.get_n_leaves()}")

# Tree-এর Text উপস্থাপনা
feature_names = ['study_hours', 'attendance', 'previous_marks']
tree_rules = export_text(dt_model, feature_names=feature_names)
print("🌳 Decision Tree Rules:")
print(tree_rules)

# Tree ভিজ্যুয়াল প্লট
plt.figure(figsize=(14, 7))
plot_tree(
    dt_model,
    feature_names=feature_names,
    class_names=['ফেল', 'পাস'],
    filled=True,
    rounded=True,
    fontsize=12
)
plt.title("Decision Tree — ছাত্র পাস/ফেল Prediction")
plt.savefig('decision_tree.png', bbox_inches='tight', dpi=150)
plt.show()
print("✅ Tree চিত্র সংরক্ষণ হয়েছে!")

y_pred = dt_model.predict(X_test)
print("Predicted:", y_pred)
print("Actual:   ", y_test.values)
print(f"\n✅ Accuracy: {accuracy_score(y_test, y_pred):.2%}")
print("\n📊 Classification Report:")
print(classification_report(y_test, y_pred, target_names=['ফেল', 'পাস']))

# কোন Feature সবচেয়ে গুরুত্বপূর্ণ?
importances = dt_model.feature_importances_
print("📊 Feature Importance:")
for feat, imp in zip(feature_names, importances):
    print(f"  {feat}: {imp:.4f} ({imp*100:.1f}%)")

# বিভিন্ন max_depth দিয়ে Accuracy তুলনা
print("Max Depth | Train Accuracy | Test Accuracy")
print("-" * 45)

for depth in [1, 2, 3, 4, 5, None]:
    model = DecisionTreeClassifier(max_depth=depth, random_state=42)
    model.fit(X_train, y_train)
    train_acc = accuracy_score(y_train, model.predict(X_train))
    test_acc  = accuracy_score(y_test,  model.predict(X_test))
    print(f"  {str(depth):9} | {train_acc:.2%}         | {test_acc:.2%}")