Datasets

The R package ships with several example datasets that are useful both for testing the wrapper and for learning the mizer workflow from Python.

List Available Datasets

import pymizer as mz

datasets = mz.list_datasets()
print(datasets)

Typical entries include:

• NS_species_params
• NS_species_params_gears
• NS_interaction
• NS_params
• NS_sim

Load A Dataset

species = mz.load_dataset("NS_species_params")
interaction = mz.load_dataset("NS_interaction")
params = mz.load_dataset("NS_params")
sim = mz.load_dataset("NS_sim")

Returned Python Types

load_dataset() maps dataset types to the closest Python representation:

• R data.frame -> pandas.DataFrame
• R matrix -> pandas.DataFrame with row and column labels
• MizerParams -> pymizer.MizerParams
• MizerSim -> pymizer.MizerSim

Load The North Sea Bundle

For the common North Sea example workflow, pymizer can load the matching datasets together:

import pymizer as mz

north_sea = mz.load_north_sea()

species = north_sea.species_params
interaction = north_sea.interaction
params = north_sea.params
sim = north_sea.sim

This is usually the most convenient entry point for a notebook because the tabular inputs and wrapped model objects stay grouped together. A runnable version of this workflow lives in examples/north_sea.py.

North Sea Example

You can rebuild the North Sea parameter object from Python inputs:

species = mz.load_dataset("NS_species_params")
interaction = mz.load_dataset("NS_interaction")

params = mz.new_multispecies_params(
    species_params=species,
    interaction=interaction,
)

This is often the most convenient starting point for notebooks because the inputs are already in pandas.