Skip to contents

Fit a Bayesian Model

Source: R/fitter.R
fit.Rd

Main fitting method for Bayesian model fitters.

Usage

fit(fitter, data_bundle, fit_spec, seed, task_ctx)

Arguments

fitter

An S7 Fitter object

data_bundle

A list containing:

  • train: Training data (data.frame)

  • test: Test data (data.frame, may be NULL)

  • response: Name of response variable (character)

  • true_params: True parameter values (named numeric, may be NULL)

  • Additional model-specific data

fit_spec

A list or single-row data.frame from fit_grid containing:

  • Model specification parameters

  • Prior specifications

  • Algorithm settings

seed

Integer seed for reproducibility

task_ctx

A list with simulation context:

  • task_id: Unique task identifier

  • data_idx: Index of data generation settings

  • fit_idx: Index of fitting settings

  • rep_idx: Replication index

Value

A bayesim_fit_result S3 object containing:

  • success: TRUE/FALSE

  • fit: backend object

  • draws: matrix with colnames or NULL

  • diagnostics: named list

  • timing: list with total, warmup, sample

  • warnings: character vector

  • error: NULL or condition