Scribd
Upload
16 views3 pages

ML (Experiment No 3)

The document outlines a Python program that performs car evaluation using Decision Tree and Random Forest classifiers. It preprocesses the dataset by encoding categorical variables and splits the data into training and testing sets. The program evaluates model performance using accuracy, precision, recall, and F1 score, and visualizes the decision trees for both classifiers.

Uploaded by

kjasus520
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views3 pages

ML (Experiment No 3)

The document outlines a Python program that performs car evaluation using Decision Tree and Random Forest classifiers. It preprocesses the dataset by encoding categorical variables and splits the data into training and testing sets. The program evaluates model performance using accuracy, precision, recall, and F1 score, and visualizes the decision trees for both classifiers.

Uploaded by

kjasus520
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Program :

import pandas as pd
from [Link] import DecisionTreeClassifier, plot_tree
from [Link] import RandomForestClassifier
from [Link] import LabelEncoder
from [Link] import accuracy_score, precision_score, recall_score, f1_score
from sklearn.model_selection import train_test_split
import [Link] as plt

df = pd.read_csv('car_evaluation.csv')
# Replace '5more' with 5 in doors and 'more' with 5 in persons
df['doors'] = df['doors'].replace('5more', '5').astype(int)
df['persons'] = df['persons'].replace('more', '5').astype(int)

# Columns to encode (all except 'doors' and 'persons')


cols_to_encode = [col for col in [Link] if col not in ['doors', 'persons', 'class']]

# Encode categorical columns except doors and persons


le_dict = {}
for col in cols_to_encode + ['class']: # Also encode the target 'class'
le = LabelEncoder()
df[col] = le.fit_transform(df[col])
le_dict[col] = le # Store LabelEncoders in case you need to decode later
df

# Split features and target


X = [Link]('class', axis=1)
y = df['class']

# Split data into train and test sets for evaluation


X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)

# Decision Tree Classifier


dt_clf = DecisionTreeClassifier(criterion='entropy', max_depth=4, random_state=0)
dt_clf.fit(X_train, y_train)
y_pred_dt = dt_clf.predict(X_test)

# Train Random Forest with limited depth


rf_clf = RandomForestClassifier(n_estimators=100, max_depth=4, random_state=42)
rf_clf.fit(X_train, y_train)
y_pred_rf = rf_clf.predict(X_test)

# Evaluation function
def evaluate_model(y_true, y_pred, model_name):
print(f"--- {model_name} Evaluation ---")
print(f"Accuracy: {accuracy_score(y_true, y_pred):.4f}")
print(f"Precision: {precision_score(y_true, y_pred, average='weighted'):.4f}")
print(f"Recall: {recall_score(y_true, y_pred, average='weighted'):.4f}")
print(f"F1 Score: {f1_score(y_true, y_pred, average='weighted'):.4f}")
print("\n")

# Print evaluations
evaluate_model(y_test, y_pred_dt, "Decision Tree")
evaluate_model(y_test, y_pred_rf, "Random Forest")

# Plot Decision Tree


[Link](figsize=(20,10))
plot_tree(dt_clf, feature_names=[Link], class_names=le_dict['class'].classes_, filled=True,
rounded=True, fontsize=12)
[Link]("Decision Tree")
[Link]()

# Plot first tree of RF with full plot_tree (no max_depth here, tree is already limited)
[Link](figsize=(16,8))
plot_tree(
rf_clf.estimators_[0],
feature_names=[Link],
class_names=le_dict['class'].classes_,
filled=True,
rounded=True,
fontsize=10
)
[Link]("Random Forest - Tree 1 (Limited depth)")
[Link]()
Output :

You might also like