GitHub

Plotting

GMRFs.jl offers recipes to make plotting spatial and spatiotemporal GMRFs effortless. These recipes are contained in a package extension which gets loaded automatically when using Makie (or rather: one of its backends).

All of the following methods may also be called more simply by calling plot or plot! with appropriate arguments.

Note

Currently, we only provide recipes for 1D spatial and spatiotemporal GMRFs. This may change soon (feel free to open a PR!) Until then, note that any 2D or 3D FEM-based GMRF may be plotted indirectly through Ferrite's support of VTK files, which may then subsequently be opened e.g. in ParaView for visualization. See Spatial Modelling with SPDEs for an example.

GMRFs.gmrf_fem_1d_plot — Function
gmrf_fem_1d_plot(gmrf::AbstractGMRF, disc::FEMDiscretization{1})

Plot the GMRF gmrf based on the 1D FEM discretization disc.

Arguments

  • gmrf::AbstractGMRF: The GMRF to plot.
  • disc::FEMDiscretization{1}: The FEM discretization.

Keyword arguments

  • with_std::Bool=true: Whether to plot the confidence bands.
  • N_samples::Int=3: The number of samples to plot.
  • rng::AbstractRNG=Random.default_rng(): The random number generator.
  • field::Symbol=:default: The field to plot.
  • mean_color: The color of the mean line.
  • conf_band_color: The color of the confidence bands.
  • sample_color: The color of the samples.
source
GMRFs.gmrf_fem_1d_spatiotemporal_plot — Function
gmrf_fem_1d_spatiotemporal_plot(
    gmrf::Union{
        ConstantMeshSTGMRF{1}, LinearConditionalGMRF{<:ConstantMeshSTGMRF{1}}
        },
    t_idx::Int
)

Plot the 1D spatiotemporal GMRF gmrf at time index t_idx.

Arguments

  • gmrf: The GMRF to plot.
  • t_idx::Int: The time index.

Keyword arguments

  • with_std::Bool=true: Whether to plot the confidence bands.
  • N_samples::Int=3: The number of samples to plot.
  • rng::AbstractRNG=Random.default_rng(): The random number generator.
  • field::Symbol=:default: The field to plot.
  • mean_color: The color of the mean line.
  • conf_band_color: The color of the confidence bands.
  • sample_color: The color of the samples.
source