🚀 Quick Start

Installation

pip install rekha

Basic Usage

import rekha as rk
import pandas as pd

# Your data (any pandas DataFrame)
df = pd.read_csv('your_data.csv')

# One line = beautiful plot
fig = rk.scatter(df, x='x', y='y', color='category')
fig.show()

That's it! You now have publication-ready visualizations with zero configuration.

📊 Sample Graphs

Scatter Plot

rk.scatter(df, x='x', y='y', color='species')

Line Plot

rk.line(df, x='date', y='price', color='stock')

Bar Chart

rk.bar(df, x='category', y='sales', color='region')

Heatmap

rk.heatmap(correlation_matrix, text_auto=True)

Every plot type. Same simple API. Beautiful results.

📖 See all plot types in our documentation →

Why Rekha Exists

The Plotting Library Dilemma:

• Matplotlib = Powerful but painful. Every plot requires 20+ lines of boilerplate.
• Plotly Express = has great semantics but limited customization and poor aesthetics for publications.
• Seaborn = Good defaults but restrictive API and complicated to use.

Rekha's Solution: Get the best of all worlds. Plotly Express semantics + matplotlib quality + unlimited customization.

✨ What Makes Rekha Special

🎯 One-Line Plots - Create publication-ready visualizations with a single function call
🎨 Intelligent Defaults - Beautiful themes that work in light mode, dark mode, and print
📊 All Plot Types - Line, scatter, bar, histogram, box, heatmap, CDF - unified API
🖨️ Print Perfection - Automatic grayscale optimization with patterns for accessibility
⚡ Zero Config - Works beautifully out of the box, customize when you need to
🔧 Unlimited Power - Full matplotlib access when you need advanced customization

Plot Types

Rekha supports all major statistical visualizations with a unified interface:

• Line plots - Time series, trends, multiple series
• Scatter plots - Relationships, correlations, bubble charts
• Bar charts - Categorical comparisons, grouped/stacked bars
• Histograms - Data distributions, density plots
• Box plots - Statistical summaries, outlier detection
• Heatmaps - 2D data, correlation matrices
• CDF plots - Cumulative distributions, percentile analysis

Key Features

Smart Export Formats

fig.save('plot.pdf', format='paper')    # Publication PDF
fig.save('plot.svg', format='web')      # Web-ready SVG  
fig.save('plot.png', format='social')   # High-res PNG

Theme Support

fig = rk.scatter(df, x='x', y='y', dark_mode=True)           # Dark theme
fig = rk.bar(df, x='cat', y='val', grayscale_friendly=True)  # Print-ready

Direct Customization

# Set axis limits directly
fig = rk.scatter(df, x='x', y='y', xlim=(0, 100), ylim=(-50, 50))

# Position legend outside plot
fig = rk.line(df, x='time', y='value', color='category', 
              legend_bbox_to_anchor=(1.05, 1))

# Full matplotlib access when needed
fig.get_axes()[0].axhline(y=0, color='red', linestyle='--')

📖 Documentation

Comprehensive documentation is available at: https://project-vajra.github.io/rekha/

• Quick Start Guide - Get up and running in minutes
• User Guide - Comprehensive feature documentation
• API Reference - Complete function reference

Philosophy

Rekha follows these principles:

1. Beautiful defaults - Plots should look professional without tweaking
2. Simple API - Common tasks should be one-liners
3. Full control - Complex customization should be possible
4. Performance - No unnecessary overhead

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for detailed information on:

• Setting up development environment
• Code style and testing requirements
• Documentation standards
• Pull request process

Quick Development Setup

git clone https://github.com/project-vajra/rekha.git
cd rekha

# Install in development mode
make install-dev

# Install pre-commit hooks
pre-commit install

# Verify setup works
make format && make lint && make test

# See all available development commands
make help

Related Projects

Rekha is part of the Project Vajra ecosystem:

• Vajra - High-performance inference engine for large language models
• Rekha - Beautiful matplotlib visualizations (this project)

License

Apache License 2.0 - see LICENSE file for details.

Etymology

Rekha (रेखा) is a Sanskrit/Hindi word meaning "line", "stroke", or "diagram". It seemed fitting for a library focused on creating beautiful line plots, scatter plots, and other visualizations.