Usage

Usage:

• Examples
  • Model simulation, optimization, analysis
    • Prerequisites
    • Model and data
    • Standard optimization Ipopt
    • Changing optimizers - Ceres
  • Hierarchical optimization
    • Prerequisites
    • Hierarchical optimization
    • Standard optimization Ipopt
    • Standard optimization vs. hierarchical optimization
  • Mini-batch optimization
    • Prerequisites
    • Mini-batch optimization
• Parameter estimation for PEtab/SBML models using parPE
  • Introduction
  • Workflow overview
  • Notation
  • Building PEtab
  • Model processing
  • Building parameter estimation executable
  • Generating an HDF5 input file for parPE parameter optimization
  • Running parameter optimization and further analyses
• Running parameter optimization for an AMICI model using parpe::OptimizationApplication
  • Command line options
  • Output explained
    • Optimizer iterations
    • Simulations during cost function evaluation
  • Environment variables
• Running AMICI simulations for parameters from an optimization result file using main_simulator.cpp
• Using parPE with charliecloud
  • parPE base docker image
    • Generating parPE base docker image
    • Using a provided docker image
  • Running the example model
    • In an interactive session
  • Optimizing a PEtab model
• Using parPE with Snakemake
  • Parameter estimation
    • An example
• HDF5 data file structure
  • Parameter estimation input files
    • Overview
    • /amiciOptions/
    • /fixedParameters/
    • /measurements/
    • /parameters/
    • Optimization options
    • Hierarchical optimization
  • Parameter estimation output files
    • Results per optimizer run
  • Simulation result files
• Mini-batch optimization with parPE
  • Using the mini-batch optimization solver of parPE
  • Setting mini-batch optimization specific hyperparameters
    • Optimization algorithms
    • Learning rate scheduling
    • ODE-model specific implementations: Rescue interceptor and line-search
  • Choosing hyperparameters
    • The learning rate
    • The mini-batch size
  • Reporting of objective function values and gradients: Important differences