mlhelper-ai 0.1.0

A Python library for machine learning preprocessing and utilities.

mlhelper

A lightweight Python library for Machine Learning preprocessing, feature engineering, visualization, model evaluation, and utilities.

---

Features

• Data preprocessing
• Missing value handling
• Outlier detection
• Feature engineering
• Encoding
• Feature scaling
• Model evaluation metrics
• Data visualization
• Model selection utilities
• Dataset loaders
• General utility functions

---

Installation

From PyPI

pip install mlhelper

From Source

git clone https://github.com/YOUR_USERNAME/mlhelper.git

cd mlhelper

pip install -e .

---

Requirements

• Python 3.10+
• NumPy
• Pandas
• Scikit-learn
• SciPy
• Matplotlib
• Seaborn
• Joblib

---

Quick Start

import pandas as pd
import mlhelper as ml

df = pd.read_csv("data.csv")

df = ml.clean_column_names(df)

df = ml.fill_mean(df)

df = ml.remove_iqr(df, "Age")

print(df.head())

---

Modules

preprocessing

ml.clean_column_names()

ml.remove_duplicates()

ml.drop_constant_columns()

ml.remove_whitespace()

ml.change_dtype()

---

missing_values

ml.fill_mean()

ml.fill_median()

ml.fill_mode()

ml.fill_constant()

ml.missing_report()

---

outlier

ml.detect_iqr()

ml.remove_iqr()

ml.cap_iqr()

ml.detect_zscore()

ml.remove_zscore()

---

feature_engineering

ml.extract_numbers()

ml.extract_text()

ml.split_unit()

ml.log_transform()

ml.sqrt_transform()

ml.frequency_encode()

---

encoding

ml.one_hot_encode()

ml.label_encode()

ml.ordinal_encode()

ml.binary_encode()

ml.target_encode()

---

scaling

ml.standard_scale()

ml.minmax_scale()

ml.robust_scale()

ml.normalize()

ml.quantile_transform()

---

metrics

ml.accuracy()

ml.precision()

ml.recall()

ml.f1()

ml.rmse()

ml.r2()

---

visualization

ml.histogram()

ml.boxplot()

ml.correlation_heatmap()

ml.scatter()

ml.learning_curve_plot()

---

model_selection

ml.split()

ml.cross_validation()

ml.grid_search()

ml.randomized_search()

ml.best_estimator()

---

datasets

iris = ml.load_iris()

wine = ml.load_wine()

digits = ml.load_digits()

housing = ml.load_california_housing()

df = ml.load_csv("train.csv")

---

utils

ml.data_summary()

ml.memory_usage()

ml.set_seed()

ml.save_pickle()

ml.load_pickle()

---

Example

import mlhelper as ml

iris = ml.load_iris()

print(iris.head())

X = iris.drop("target", axis=1)

y = iris["target"]

X_train, X_test, y_train, y_test = ml.split(X, y)

print(X_train.shape)

---

Testing

Install pytest

pip install pytest

Run all tests

python -m pytest -v

Coverage

python -m pytest --cov=mlhelper

---

Project Structure

mlhelper/
│
├── mlhelper/
│   ├── preprocessing.py
│   ├── missing_values.py
│   ├── outlier.py
│   ├── feature_engineering.py
│   ├── encoding.py
│   ├── scaling.py
│   ├── metrics.py
│   ├── visualization.py
│   ├── model_selection.py
│   ├── datasets.py
│   └── utils.py
│
├── tests/
├── examples/
├── docs/
├── README.md
├── LICENSE
├── pyproject.toml
└── requirements.txt

---

Roadmap

• Preprocessing
• Missing Values
• Outlier Detection
• Feature Engineering
• Encoding
• Scaling
• Metrics
• Visualization
• Model Selection
• Dataset Loader
• Documentation Improvements
• More Datasets
• More Visualizations
• Deep Learning Utilities

---

Contributing

Contributions are welcome.

1. Fork the repository
2. Create a new branch
3. Commit your changes
4. Open a Pull Request

---

License

MIT License

---

Author

Gautam

Computer Science Engineering (AI & ML)

Python • Machine Learning • Deep Learning • Open Source

---

If you find this project useful, consider giving it a ⭐ on GitHub.