Exporting
For added convenience, InformationGeometry.jl already provides several methods, which can be used to export results like confidence regions or geodesics.
InformationGeometry.SaveConfidence
— FunctionSaveConfidence(sols::AbstractVector{<:AbstractODESolution}, N::Int=500; sigdigits::Int=7, adaptive::Bool=true) -> Matrix
SaveConfidence(Planes::AbstractVector{<:Plane}, sols::AbstractVector{<:AbstractODESolution}, N::Int=500; sigdigits::Int=7, adaptive::Bool=true) -> Matrix
Returns a Matrix
of with N
rows corresponding to the number of evaluations of each ODESolution
in sols
. The colums correspond to the various components of the evaluated solutions. E.g. for an ODESolution
with 3 components, the 4. column in the Matrix
corresponds to the evaluated first components of sols[2]
.
In particular, choosing the keyword adaptive=true
samples the ODESolution
objects roughly proportional to their curvature (instead of equidistant), which means that more samples are provided from tight bends than from segments that are straight, leading to more faithful representations of the confidence boundary when plotting.
InformationGeometry.SaveDataSet
— FunctionSaveDataSet(DS::DataSet; sigdigits::Int=0)
Returns a DataFrame
whose columns respectively constitute the x-values, y-values and standard distributions associated with the data points. For sigdigits > 0
the values are rounded to the specified number of significant digits.