PlotSmith Documentation

PlotSmith is a layered plotting library for ML models with strict architectural boundaries. It provides clean, minimalist visualizations following best practices for analytical figures.

Contents:

• Installation
  • Requirements
  • Installing PlotSmith
  • Development Installation
• Quick Start
  • Basic Usage
  • Time Series with Confidence Bands
  • Residual Analysis
  • Heatmap
  • Next Steps
• Architecture
  • Layer 1: Data and Representations
  • Layer 2: Primitives
  • Layer 3: Tasks
  • Layer 4: Workflows
  • Public API
• API Reference
  • Workflows
  • View Objects
  • Primitives
• Examples
  • Jupyter Notebooks
  • Python Scripts
  • New Chart Types
• Performance Guide
  • Performance Characteristics
  • Optimization Tips
  • Memory Usage
  • Benchmarking
  • Performance Regression Testing
  • Known Performance Considerations
• Troubleshooting Guide
  • Installation Issues
  • Plotting Issues
  • Data Validation Issues
  • Type Errors
  • Performance Issues
  • Styling Issues
  • CI/CD Issues
  • Getting Help
• Contributing
  • Development Setup
  • Architecture Guidelines
  • Code Style
  • Testing

Features

• 4-Layer Architecture: Strict separation of concerns with clear boundaries

• Minimalist Styling: Clean, publication-ready plots with serif fonts and minimal clutter

• Type-Safe: Full type hints and immutable data structures

• Comprehensive: Time series, scatter plots, histograms, bar charts, heatmaps, residuals, and more

• ML-Focused: Built for model evaluation, backtesting, and analysis workflows

Indices and tables

• Index

• Module Index

• Search Page