Examples are provided as Jupyter notebooks in a separate freud-examples repository. These notebooks may be launched interactively on Binder or downloaded and run on your own system. Visualization of data is done via Matplotlib [Matplotlib] and Bokeh [Bokeh], unless otherwise noted.
There are a few critical concepts, algorithms, and data structures that are central to all of freud. The box module defines the concept of a periodic simulation box, and the locality module defines methods for finding nearest neighbors for particles. Since both of these are used throughout freud, we recommend familiarizing yourself with these first, before delving into the workings of specific freud analysis modules.
These introductory examples showcase the functionality of specific modules in freud, showing how they can be used to perform specific types of analyses of simulations.
- Density - ComplexCF
- RDF: Accumulating g(r) for a Fluid
- RDF: Choosing Bin Widths
- LocalDescriptors: Steinhardt Order Parameters
- Hexatic Order Parameter
- LocalQl, LocalWl
- Shifting Example
The examples below go into greater detail about specific applications of freud and use cases that its analysis methods enable, such as user-defined analyses, machine learning, and data visualization.
- Implementing Common Neighbor Analysis as a custom method
- Analyzing simulation data from HOOMD-blue at runtime
- Analyzing data from LAMMPS
- Using Machine Learning for Structural Identification
- Calculating Strain via Voxelization
- Visualizing analyses with fresnel
- Visualization with plato
- Visualizing 3D Voronoi and Voxelization
Performance is a central consideration for freud. Below are some benchmarks comparing freud to other tools offering similar analysis methods.