FeatureEng-arun8nov 0.0.2

A collection of feature engineering methods for preprocessing datasets.

📦 FeatureEngineering

FeatureEngineering is a Python package that provides multiple encoding and scaling techniques for preprocessing categorical and numerical data. It’s lightweight, dependency-free (except for pandas and numpy), and ideal for rapid feature engineering during model development.

---

📌 Key Features

✅ One-Hot Encoding
✅ Label Encoding
✅ Ordinal Encoding
✅ Target Encoding
✅ Frequency Encoding
✅ Binary Encoding
✅ Standard Scaling (Z-Score)
✅ Normalization Scaling (Min-Max)

---

📥 Installation

Install the package using pip:

pip install FeatureEng_arun8nov

Ensure you have pandas and numpy installed.

---

🔧 How to Use

Step 1: Import the Class

from Preprocess import FeatureEngineering
import pandas as pd

Step 2: Create Sample Data

df = pd.DataFrame({
    'Color': ['Red', 'Blue', 'Green', 'Red'],
    'Size': ['S', 'M', 'L', 'S']
})

Step 3: Initialize the Class

fe = FeatureEngineering()

---

🧠 Encoding Methods

1. One-Hot Encoding

ohe_df = fe.OneHot_En(df[['Color']])

2. Label Encoding

label_df = fe.Lable_En(df[['Color']].copy())

3. Ordinal Encoding

order_list = [['S', 'M', 'L']]
ord_df = fe.Order_En(order_list, df[['Size']].copy())

4. Target Encoding

target = pd.DataFrame({'Price': [100, 200, 300, 150]})
target_encoded_df = fe.Target_En(df[['Color']], target)

5. Frequency Encoding

freq_df = fe.Freq_En(df[['Color']].copy())

6. Binary Encoding

bin_df = fe.Bin_En(df[['Color']].copy())

---

📏 Scaling Methods

For numerical columns:

num_df = pd.DataFrame({'Age': [20, 25, 30], 'Salary': [30000, 50000, 70000]})

7. Standard Scaling (Z-Score)

std_df = fe.Stannd_Scale(num_df)

8. Normalization (Min-Max)

norm_df = fe.Normm_Scale(num_df)

---

🧾 Function Summary

Method	Purpose	Input Type
OneHot_En	One-Hot Encode categorical data	DataFrame
Lable_En	Label Encode categorical data	DataFrame
Order_En	Ordinal Encode based on user order	DataFrame + list
Target_En	Encode using target mean per category	DataFrame + target column as DataFrame
Freq_En	Encode based on frequency count	DataFrame
Bin_En	Encode as binary numbers	DataFrame
Stannd_Scale	Apply Z-score scaling	Numerical DataFrame
Normm_Scale	Apply Min-Max scaling	Numerical DataFrame

---

🔐 License

This package is licensed under the MIT License.

---

👨‍💻 Author

Arunprakash B
GitHub: [arun8nov]
LinkedIn: [https://www.linkedin.com/in/arun8nov/]

---

📢 Note

This package is ideal for:

• ML beginners looking to understand encodings manually
• Rapid prototyping pipelines
• Educational and academic projects

---

🙋 Need Help?

If you encounter any bugs or have suggestions, please open an issue on GitHub.