datawhisk 1.0.3

A lightweight, practical utility library for data scientists and ML engineers

Datawhisk

A lightweight, practical utility library for data scientists and ML engineers.

datawhisk provides a curated collection of analytical helpers designed to streamline common data science workflows. Built with speed, simplicity, and reliability in mind.

Why datawhisk?

• 🚀 Fast: Optimized implementations faster than manual approaches
• 🎯 Practical: Solves real problems data scientists face daily
• 🪶 Lightweight: Minimal dependencies, quick to install
• 📊 Reliable: 90%+ test coverage, production-ready
• 🧩 Intuitive: Clean APIs that "just work"

Installation

pip install datawhisk

With visualization support:

pip install datawhisk[viz]

Quick Start

Memory Optimizer

Automatically optimize DataFrame memory usage:

from datawhisk.analytical import optimize_memory

import pandas as pd

# Create a DataFrame
df = pd.DataFrame({
    'id': range(1000000),
    'category': ['A', 'B', 'C'] * 333334,
    'value': [1.5] * 1000000
})

# Optimize memory
optimized_df, report = optimize_memory(df, return_report=True)

print(f"Memory reduced by {report['reduction_percent']:.1f}%")
# Output: Memory reduced by 65.2%

Correlation Analyzer

Smart correlation analysis with multicollinearity detection:

from datawhisk.analytical import analyze_correlations

# Analyze correlations
results = analyze_correlations(
    df,
    target='price',
    threshold=0.8,
    method='pearson'
)

print(results.recommendations)
# Output: ['Remove feature_X (VIF=12.3)', 'Keep feature_Y (VIF=2.1)']

Quick EDA Reporter

Fast exploratory data analysis:

from datawhisk.analytical import quick_eda

# Generate EDA report (Standard)
report = quick_eda(df, visualize=True)

# Generate Custom Report (Modular)
# checks only missing values and structure
report_custom = quick_eda(
    df,
    check_missing=True,
    check_structure=True,
    check_outliers=False,
    check_distribution=False
)

print(report.summary)
# Dataset Shape: (1000, 5)
# Missing Values: 23 (2.3%)
# Outliers Detected: 156 total
# High cardinality columns: ['user_id', 'transaction_id']

Features

Analytical Helpers

• Memory Optimizer: Automatically downcast dtypes and optimize memory usage
• Correlation Analyzer: Calculate correlations with VIF and multicollinearity detection
• Quick EDA Reporter: Fast statistical summaries with anomaly detection

Comparison with Alternatives

Feature	datawhisk	Pandas	pandas-profiling
Memory Optimization	✅ Auto	Manual	❌
VIF Calculation	✅ Built-in	Manual	❌
Speed (EDA)	⚡ Fast	N/A	🐌 Slow
Dependencies	🪶 3 core	-	📦 20+
Learning Curve	📉 Low	Medium	Medium

Documentation

• Installation Guide
• Quick Start
• API Reference
• Tutorials
• Full Documentation

Requirements

• Python 3.8+
• numpy >= 1.20.0
• pandas >= 1.3.0
• scipy >= 1.7.0

Optional:

• matplotlib >= 3.3.0 (for visualizations)
• seaborn >= 0.11.0 (for enhanced visualizations)

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

# Clone the repository
git clone https://github.com/Ramku3639/datawhisk.git
cd datawhisk

# Install development dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run code quality checks
black datawhisk tests
flake8 datawhisk tests
mypy datawhisk

Roadmap

v0.2.0 (Next Release)

• Statistical utilities
• Data validation helpers
• Extended visualization options

v0.3.0

• Time series utilities
• Text processing helpers
• Performance benchmarking tools

See CHANGELOG.md for release history.

License

MIT License - see LICENSE for details.

Citation

If you use datawhisk in your research, please cite:

@software{datawhisk,
  author = {RamMohan Reddy K},
  title = {datawhisk: A Practical Utility Library for Data Scientists},
  year = {2025},
  url = {https://github.com/Ramku3639/datawhisk}
}

Support

• 📧 Email: [e-mail]
• 🐛 Issue Tracker
• 💬 Discussions

---

Built with ❤️ for the data science community