pyturbo-analytics 0.1.2

A high-performance Python library for blazing-fast data analysis

<<<<<<< HEAD

pyturbo-analytics

=======

PyTurbo: Unleashing the Speed of Data Analysis 🚀

PyTurbo is a high-performance Python library designed to dramatically accelerate data analysis tasks by leveraging multiple computing paradigms including multithreading, multiprocessing, GPU acceleration, and compiled code optimization.

Features

• Fast DataFrame Operations: Parallelized Pandas-style operations with GPU acceleration
• Smart Task Optimization: Automatic workload distribution across CPU cores and GPUs
• Performance Profiling: Built-in analysis tools for code optimization
• High-Speed Data Loading: Optimized I/O for CSV, JSON, SQL, and Parquet formats
• GPU-Accelerated Visualizations: Real-time plotting of massive datasets
• Customizable Accelerators: Easy-to-use APIs for custom optimized operations
• Distributed Processing: Seamless scaling with Dask and Ray integration

Installation

pip install pyturbo

For development installation:

git clone https://github.com/pyturbo/pyturbo.git
cd pyturbo
pip install -e ".[dev]"

Quick Start

import pyturbo as pt

# Create a TurboFrame (high-performance DataFrame)
tf = pt.TurboFrame.from_csv("large_dataset.csv")

# Perform accelerated operations
result = tf.groupby("category").agg({
    "value": ["mean", "sum", "count"]
}).compute()

# Use GPU acceleration
with pt.use_gpu():
    result = tf.merge(other_tf, on="key")

Requirements

• Python 3.8+
• CUDA-capable GPU (optional, for GPU acceleration)
• CUDA Toolkit 11.x (for GPU features)

Documentation

Visit our documentation for:

• Detailed API reference
• Performance optimization guides
• Examples and tutorials
• Best practices

Contributing

We welcome contributions! Please see our Contributing Guide for details.

License

MIT License - see the LICENSE file for details.

Citation

If you use PyTurbo in your research, please cite:

@software{pyturbo2025,
  author = {PyTurbo Team},
  title = {PyTurbo: High-Performance Data Analysis Library},
  year = {2025},
  url = {https://github.com/pyturbo/pyturbo}
}

373cfb017 (Initial commit)