yt Documentation

The yt project provides infrastructure for manipulating datasets to extract relevant information. Core analysis workflows include:

• Data Representation: Working with Fields, data objects, and units to represent and access physical quantities.
• Filtering: Selecting specific subsets of datasets based on defined criteria.
• Time Series Analysis: Analyzing chronological datasets from the same underlying simulation or source.
• Parallel Computation: Running tasks across multiple processors to accelerate analysis.
• Derived Fields: Creating new fields based on existing data.
• Data Management: Generating processed data and saving results for later use.
• Integrations: Utilizing astropy integrations for specialized astrophysical analysis.

The yt.visualization.plot_window.PlotWindow interface provides a high-level way to quickly inspect simulation outputs. yt supports several types of visualizations:

• Slices: 2D planes through the data.
• Projections: 2D images created by integrating data along a line of sight.
• 1D spatial line plots: Data values along a specific spatial path.
• 1D profiles: Data values as a function of a single coordinate.
• 2D profiles (phase plots): Plots showing the relationship between two different variables.

The yt project provides a wide range of built-in visualization methods for exploring datasets. Supported visualization techniques include:

• Slices and Projections: 2D views of data.
• Line Plots and Profiles: 1D data extractions.
• Phase Plots: Visualizing relationships between variables.
• Volume Rendering: 3D volumetric visualization.
• 3D Surfaces and Streamlines: Visualizing structures and flow.
• Interactive Interfaces: A Google Maps-like interface for interactive dataset exploration.
• Specialized Rendering: Unstructured mesh rendering and particle dataset visualization (via Firefly).

yt is a community-developed Python toolkit designed for the analysis and visualization of volumetric data. While frequently used for astrophysical simulation data, it is applicable to various scientific domains including seismology, radio telescope data, weather simulations, and nuclear engineering simulations.

Key capabilities include:

• Code Format Support: Support for many different data formats.
• Data Examination: Tools to load and examine various dataset types.
• Visualization: Capabilities for creating plots, projections, volume renderings, and movies.
• Analysis: General data manipulation and domain-specific analysis (e.g., astrophysical analysis, clump finding, and cosmology calculations).

If you need help or want to learn more about yt, use the following resources:

• Documentation: The latest recipes, tutorials, and API documentation are available at http://yt-project.org/docs/dev/.
• Users' Mailing List: Use the yt-users mailing list for general questions.
• Developers' Mailing List: Use the yt-dev mailing list for development-specific issues.
• Slack: Join the community on Slack at yt-project.slack.com (you may need to request an invite).

Beyond core yt and the official yt_astro_analysis and yt_attic packages, there is a wide ecosystem of independent extension packages. These include tools for:

• Visual effects
• Interactive widgets
• Synthetic absorption spectra
• X-ray observations
• Merger-trees

You can find a full list of available extension packages at the official yt Extensions page: http://yt-project.org/extensions.html

If you cannot run the notebooks interactively, you can view the pre-filled examples non-interactively. The quickstart covers the following topics through a series of notebooks:

• Introduction: Getting started and data setup.
• Data Inspection: Examining datasets.
• Simple Visualization: Basic plotting and rendering.
• Data Objects and Time Series: Working with core yt data structures.
• Derived Fields and Profiles: Creating new data from existing fields and generating profiles.
• Volume Rendering: Advanced 3D visualization techniques.

yt provides several methods for visualizing multidimensional data:

• Slice Plots: A 2D plane intersecting a 3D volume (or data object) to show field data at that intersection.
• Projection Plots: 3D volumes projected into 2D by summing or integrating the field along the line of sight (with or without a weighting field).
• Line Queries (Rays): A single line intersecting a 3D dataset.
• Profile Plots: 1D plots showing the relationship between a field and a variable (e.g., average density vs. radius).
• Phase Plots: 2D plots correlating two fields, typically weighted by a third field (e.g., density vs. temperature weighted by mass).
• Advanced Visualization: Includes streamlines for velocity flow, volume renderings (photorealistic isocontour images), and isosurfaces.

The yt testing suite is built on the pytest framework and consists of two primary test types to ensure functional consistency:

• Unit tests: Verify that small, isolated pieces of code behave as intended in predictable contexts.
• Answer tests: Generate outputs from the user-facing yt API and compare them against outputs from a known good, older version of yt to ensure consistency.

While developers are encouraged to run tests locally on changed features, the full test suite is run via Continuous Integration (CI) on GitHub Actions and a dedicated server at the University of Illinois. The full suite may take several hours to complete.