Types

Newtrinos

A Julia package for global analysis of neutrino data. Provides modular physics models, experimental likelihoods, and inference tools (profile likelihood, Bayesian sampling).

Physics

Abstract type for physics modules. Subtypes must have params::NamedTuple and priors::NamedTuple fields.

Experiment

Abstract type for experiment modules. Subtypes must have physics, params, priors, assets, forward_model, and plot fields.

NewtrinosResult(; axes, values, meta=Dict())

Container for the results of a scan or profile run.

Fields

• axes::NamedTuple: scan grid coordinates, one entry per scanned parameter. Each entry is a vector of grid-point values.
• values::NamedTuple: per-grid-point outputs — likelihood values (llh, log_posterior) and optimized nuisance parameter arrays.
• meta::Dict: execution metadata (hostname, date, git commit, timing). Populated by add_meta!.

Wrapper <: Newtrinos.Experiment

Wraps a Newtrinos.Experiment with parameter name aliases.

When two experiments share a physics parameter under different names (e.g. theta13 vs sin2theta13), a Wrapper renames parameters seen by the outer analysis while transparently translating back to the original names before calling the inner experiment's forward_model and plot functions.

Construct with Wrapper(experiment, aliases) where aliases is a Dict{Symbol,Symbol} mapping original parameter names to their aliases.

Fields

• x::Newtrinos.Experiment: the wrapped experiment.
• aliases::Dict{Symbol,Symbol}: map from original name → alias.
• translated_keys::Vector{Symbol}: all parameter keys after alias substitution.
• reverse_lookup::Dict{Symbol,Symbol}: map from alias → original name.

Examples

aliases = Dict(:dm21_sq => :Dm21)
wrapped = Wrapper(dayabay_experiment, aliases)