PyBNF Module References

Modules:

• PyBNF entry point (pybnf.pybnf)
  • main()
• PyBNF algorithms (pybnf.algorithms)
  • Core execution
    • FailedSimulation
    • HybridJobGroup
    • Job
    • JobGroup
    • MultimodelJobGroup
    • Result
    • result_from_completed()
    • run_job()
    • Algorithm
    • exp10()
    • latin_hypercube()
  • Optimizers
    • Shared bases
    • Metaheuristic optimizers
    • Gradient-based optimizers
  • Bayesian samplers
    • BayesianAlgorithm
    • MCMCFamilyConfig
    • BasicBayesMCMCAlgorithm
    • BasicMCMCConfig
    • AdaptiveMCMCConfig
    • Adaptive_MCMC
    • DreamAlgorithm
    • DreamConfig
    • PDreamAlgorithm
    • PDreamConfig
    • HMCConfig
    • HMCSampler
  • Model checking
    • ModelCheck
• PyBNF cluster setup (pybnf.cluster)
  • Cluster
    • Cluster.local_cluster_kwargs()
    • Cluster.read_node_names()
    • Cluster.setup_cluster()
    • Cluster.teardown()
• PyBNF configuration (pybnf.config)
  • UnknownObjectiveFunctionError
  • UnmatchedExperimentalDataError
  • UnspecifiedConfigurationKeyError
  • reinit_logging()
• PyBNF constraint specification (pybnf.constraint)
  • AlwaysConstraint
  • AtConstraint
  • BetweenConstraint
  • Constraint
    • Constraint.bind_scale_param()
    • Constraint.coerce_penalty_model()
    • Constraint.find_keys()
    • Constraint.get_difference()
    • Constraint.get_key()
    • Constraint.get_log_likelihood()
    • Constraint.get_log_likelihood_logit()
    • Constraint.get_penalty()
    • Constraint.get_penalty_curvature()
    • Constraint.get_penalty_gradient()
    • Constraint.get_static_penalty()
    • Constraint.index()
    • Constraint.penalty()
    • Constraint.penalty_curvature()
    • Constraint.penalty_gradient()
  • ConstraintSet
    • ConstraintSet.load_constraint_file()
    • ConstraintSet.number_failed()
    • ConstraintSet.output_itemized_eval()
    • ConstraintSet.total_penalty()
  • OnceConstraint
  • SplitAtConstraint
• PyBNF data container (pybnf.data)
  • Data
    • Data.average()
    • Data.from_columns()
    • Data.gen_bootstrap_weights()
    • Data.get_row()
    • Data.load_data()
    • Data.load_rr_header()
    • Data.normalize()
    • Data.normalize_to_floor()
    • Data.normalize_to_init()
    • Data.normalize_to_peak()
    • Data.normalize_to_unit_scale()
    • Data.normalize_to_zero()
    • Data.rename_column()
  • DuplicateColumnError
  • NormalizationRecord
  • OutputSensitivities
    • OutputSensitivities.slice_for()
  • stack_scan_sensitivities()
• PyBNF convergence diagnostics (pybnf.diagnostics)
  • ess()
  • ess_from_chains()
  • rhat()
  • split_chain_rhat()
  • split_chains()
• PyBNF gradient plumbing (pybnf.gradient)
  • Errors
    • GradientNotSupported
  • Routing
    • Bind by id (ADR-0034)
    • Per-condition perturbation = local derivative (ADR-0028)
    • Reached through a condition (a per-condition estimated initial condition, ADR-0076)
    • The seed derivative is not assumed to be 1 (#530)
    • One contribution per native column (#537)
    • Scope
    • ExperimentRouting
    • ParamRoute
    • RouteContribution
    • SeedTerm
    • apply_routing()
    • apply_routings()
    • classify_bound_id()
    • classify_condition_target()
    • classify_free_param()
    • condition_factor()
    • route_experiment()
    • route_for_model()
  • Assembly
    • Convention pin (issue #449)
    • Native -> sampling space (once, ADR-0029)
    • Estimated noise scale (layer D, #451)
    • Trajectory transforms + normalization (layer F, #453)
    • Asymmetric / non-Gaussian families (layer G, #454; least-squares residual #459)
    • Constraint / qualitative penalties (layer I, #456)
    • Measurement-model layer (layer H, #455)
    • GradientResult
    • assemble_constraint_gradient()
    • assemble_constraint_hessian()
    • assemble_fisher_hessian()
    • assemble_gaussian_gradient()
    • assemble_gradient_and_fisher_hessian()
• PyBNF ArviZ bridge (pybnf.inference_data)
  • from_pybnf()
• PyBNF measurement models (pybnf.measurement)
  • MeasurementLayer
    • MeasurementLayer.apply()
  • MeasurementModel
    • MeasurementModel.materialize()
    • MeasurementModel.prediction_sensitivity()
  • PerMeasurementModel
    • PerMeasurementModel.prediction_sensitivity()
    • PerMeasurementModel.value()
• PyBNF noise models (pybnf.noise)
  • Base kernel
    • NoiseModel
  • Distribution families
    • Gaussian
    • Laplace
    • StudentT
    • NegBinomial
  • Noise axes
    • AdditiveNoiseScale
    • LocationInterpretation
  • Noise parameter sources
    • ColumnMeanSigma
    • ConstantSigma
    • DataColumnSigma
    • FormulaSigma
    • FreeParameterSigma
    • PerMeasurementFormulaSigma
    • PredictionFormulaSigma
    • RelativeSigma
    • SigmaSource
• PyBNF objective functions (pybnf.objective)
  • AveNormSumOfSquaresObjective
    • AveNormSumOfSquaresObjective.eval_point()
    • AveNormSumOfSquaresObjective.evaluate()
  • ChiSquareObjective
    • ChiSquareObjective.noise
  • ChiSquareObjective_Dynamic
    • ChiSquareObjective_Dynamic.noise
  • ColumnSummationObjective
    • ColumnSummationObjective.eval_column()
    • ColumnSummationObjective.evaluate()
    • ColumnSummationObjective.from_config()
  • ConstraintCounter
    • ConstraintCounter.evaluate()
    • ConstraintCounter.evaluate_multiple()
  • DirectPassObjective
    • DirectPassObjective.evaluate()
    • DirectPassObjective.evaluate_multiple()
  • InformationCriteria
    • InformationCriteria.aic
    • InformationCriteria.aicc
    • InformationCriteria.bic
    • InformationCriteria.k
    • InformationCriteria.log_likelihood
    • InformationCriteria.n
  • KLLikelihood
    • KLLikelihood.eval_column()
  • LaplaceObjective
    • LaplaceObjective.noise
  • LikelihoodObjective
    • LikelihoodObjective.aligned_prediction_data()
    • LikelihoodObjective.data_fit_grad_point()
    • LikelihoodObjective.eval_point()
    • LikelihoodObjective.evaluate_pointwise()
    • LikelihoodObjective.from_config()
    • LikelihoodObjective.has_least_squares_residual()
    • LikelihoodObjective.is_linear_gaussian()
    • LikelihoodObjective.location_fisher_point()
    • LikelihoodObjective.noise
    • LikelihoodObjective.noise_fisher_point()
    • LikelihoodObjective.noise_grad_point()
    • LikelihoodObjective.noise_profiling_plan()
    • LikelihoodObjective.overrides
    • LikelihoodObjective.required_free_noise_params()
    • LikelihoodObjective.residual_point()
    • LikelihoodObjective.set_default_location()
    • LikelihoodObjective.supports_pointwise_log_likelihood
  • LogNormalObjective
    • LogNormalObjective.noise
  • NegBinLikelihood
    • NegBinLikelihood.from_config()
    • NegBinLikelihood.noise
  • NegBinLikelihood_Dynamic
    • NegBinLikelihood_Dynamic.noise
  • NormSumOfSquaresObjective
    • NormSumOfSquaresObjective.eval_point()
  • ObjectiveCalculator
    • ObjectiveCalculator.evaluate_objective()
  • ObjectiveFunction
    • ObjectiveFunction.aligned_prediction_data()
    • ObjectiveFunction.data_fit_grad_point()
    • ObjectiveFunction.evaluate()
    • ObjectiveFunction.evaluate_multiple()
    • ObjectiveFunction.evaluate_pointwise()
    • ObjectiveFunction.from_config()
    • ObjectiveFunction.has_least_squares_residual()
    • ObjectiveFunction.is_linear_gaussian()
    • ObjectiveFunction.location_fisher_point()
    • ObjectiveFunction.measurement
    • ObjectiveFunction.noise_fisher_point()
    • ObjectiveFunction.noise_grad_point()
    • ObjectiveFunction.required_free_noise_params()
    • ObjectiveFunction.residual_point()
    • ObjectiveFunction.supports_pointwise_log_likelihood
  • SumOfDiffsObjective
    • SumOfDiffsObjective.eval_point()
  • SumOfSquaresObjective
    • SumOfSquaresObjective.eval_point()
  • SummationObjective
    • SummationObjective.analytic_scale_sensitivity()
    • SummationObjective.eval_point()
    • SummationObjective.evaluate()
    • SummationObjective.from_config()
    • SummationObjective.prediction_sensitivity()
  • WassersteinObjective
    • WassersteinObjective.eval_column()
  • build_named_objective()
  • build_profile_objective()
  • build_whole_fit_noise_objective()
  • information_criteria()
  • likelihood_information_criteria()
• PyBNF configuration parsing (pybnf.parse)
  • parse_normalization_chain()
  • parse_normalization_def()
• PyBNF PEtab v2 interoperability (pybnf.petab)
  • Parameters
    • PetabParameterRow
    • free_parameter_from_row()
    • free_parameters_from_file()
    • free_parameters_from_table()
    • petab_parameter_row()
    • read_parameter_table()
    • write_parameter_table()
  • Observables and noise
    • PetabObservableRow
    • noise_model_from_row()
    • noise_models_from_file()
    • noise_models_from_table()
    • petab_observable_row()
    • read_observable_table()
    • write_observable_table()
  • Measurements
    • PetabMeasurementRow
    • data_from_measurement_rows()
    • dose_response_measurement_rows()
    • measurement_param_bindings()
    • measurement_rows_from_data()
    • noise_parameter_ids_by_observable()
    • noise_parameters_by_observable()
    • observable_parameters_by_observable()
    • read_measurement_table()
    • reconstruct_dose_responses()
    • reconstruct_preequilibrated_dose_responses()
    • row_varying_noise_ids()
    • row_varying_noise_param_ids()
    • row_varying_observable_ids()
    • write_measurement_table()
  • Conditions and experiments
    • CONDITION_ID_PREFIX
    • PetabConditionRow
    • PetabExperimentRow
    • PetabMappingRow
    • REF_MARKER
    • SPECIES_ID_PREFIX
    • build_dose_response_conditions()
    • build_experiment_conditions()
    • build_preequilibrated_dose_response_conditions()
    • build_preequilibration_conditions()
    • condition_name_from_id()
    • conditions_from_rows()
    • is_species_target()
    • mutation_target_value()
    • read_condition_table()
    • read_experiment_table()
    • read_mapping_table()
    • species_target_id()
    • surrogate_name()
    • write_condition_table()
    • write_experiment_table()
    • write_mapping_table()
  • Measurement formulas
    • bngl_body_to_petab_math()
    • compile_objective_expression()
    • compile_petab_formula()
    • compile_petab_formula_derivatives()
    • formula_free_symbols()
    • inline_assignment_rules()
    • inline_constants()
    • substitute_placeholders()
  • BNGL model loader
    • BnglModel
    • MODEL_TYPE_BNGL
    • register_bngl()
  • Importing a problem
    • ImportedExperiment
    • import_job()
    • read_problem_yaml()
  • Exporting a job
    • clean_model_for_petab()
    • export_job()
    • write_problem_yaml()
• PyBNF printing functions (pybnf.printing)
  • PybnfError
  • print0()
  • print1()
  • print2()
• PyBNF priors (pybnf.priors)
  • Package facade
    • Bijector
    • FrozenPrior
    • Linear
    • Ln
    • Log10
    • NoPrior
    • Prior
    • Scale
    • TruncatedPrior
    • bijector_for_support()
    • build_prior()
    • var_keyword_grammar()
  • The Prior abstraction
    • FrozenPrior
    • NoPrior
    • Prior
  • Parameter scale
    • Linear
    • Ln
    • Log10
    • Scale
  • Bounded support
    • TruncatedPrior
    • Bijector
    • BoxBijector
    • IdentityBijector
    • LowerBoundedBijector
    • UpperBoundedBijector
    • bijector_for_support()
• PyBNF model and parameter containers (pybnf.pset)
  • Action
    • Action.output_length()
  • BNGLModel
    • BNGLModel.add_action()
    • BNGLModel.copy_with_param_set()
    • BNGLModel.execute()
    • BNGLModel.find_t_length()
    • BNGLModel.get_suffixes()
    • BNGLModel.model_text()
    • BNGLModel.save()
    • BNGLModel.save_all()
  • FailedSimulationError
  • FreeParameter
    • FreeParameter.add()
    • FreeParameter.add_rand()
    • FreeParameter.d_from_sampling_space()
    • FreeParameter.diff()
    • FreeParameter.from_sampling_space()
    • FreeParameter.from_sampling_space_jax()
    • FreeParameter.has_bounded_initialization
    • FreeParameter.has_bounded_support
    • FreeParameter.has_prior
    • FreeParameter.initial_value_from_quantile()
    • FreeParameter.multiply()
    • FreeParameter.prior_logpdf()
    • FreeParameter.prior_logpdf_jax()
    • FreeParameter.prior_support()
    • FreeParameter.sample_initial_value()
    • FreeParameter.sample_value()
    • FreeParameter.set_value()
    • FreeParameter.to_sampling_space()
    • FreeParameter.value_from_quantile()
  • Model
    • Model.add_mutant()
    • Model.copy_with_param_set()
    • Model.emit_suffixes
    • Model.execute()
    • Model.get_suffixes()
    • Model.save()
  • ModelError
  • MutationSet
  • NetModel
    • NetModel.copy_with_param_set()
    • NetModel.save()
  • OutOfBoundsException
  • PSet
    • PSet.get_param()
    • PSet.keys()
    • PSet.keys_to_string()
    • PSet.values_to_string()
  • ParamScan
    • ParamScan.set_preequilibration()
  • SbmlModel
    • SbmlModel.execute()
  • SbmlModelNoTimeout
    • SbmlModelNoTimeout.copy_with_param_set()
    • SbmlModelNoTimeout.execute()
    • SbmlModelNoTimeout.get_suffixes()
    • SbmlModelNoTimeout.model_text()
    • SbmlModelNoTimeout.save()
  • TimeCourse
    • TimeCourse.set_preequilibration()
  • Trajectory
    • Trajectory.add()
    • Trajectory.best_fit()
    • Trajectory.best_fit_name()
    • Trajectory.best_score()
    • Trajectory.load_trajectory()
    • Trajectory.write_to_file()
  • clear_sim_registry()
  • reap_active_sims()
  • run_subprocess()
  • set_sim_registry()
• Multiple shooting (pybnf.shooting)
  • Knot placement
    • Three placements, and one of them is the default for a measured reason
    • Names, because the homotopy transfers by name
    • Which segment owns a data point
    • EQUAL_OBSERVATIONS
    • EQUAL_TIME
    • EXPLICIT
    • KNOT
    • OBSERVATIONS_PER_SEGMENT
    • SegmentGrid
    • max_segments()
  • The segment-simulation seam
    • What a segment simulation is
    • Segment 0 is not special-cased
    • SegmentBackend
    • SegmentSimulationFailed
    • SegmentTrace
    • trace_from_data()
    • Why the SBML/Antimony path, and what the .net path would need
    • Simulator reuse
    • BngsimSegmentBackend
    • The cost, which is real and is the model’s, not the method’s
    • NetSegmentBackend
  • The segment pass
    • Threads, because the integration releases the GIL
    • A lane is a warm engine + simulator, and it is not free
    • What parallel gives up
    • SERIAL
    • SegmentPool
    • SegmentTask
  • The transcription
    • The objective half needs no new residual math
    • The constraint terms never enter f
    • AUX_DECADES
    • MultipleShootingProblem
    • STATE_FLOOR
    • SegmentedExperiment
    • ShootingExperiment
    • seed_stage()
  • The inner solver
    • Reusing the fit type rather than reimplementing it
    • Two tolerances, and why the outer loop’s is only a floor away from being obeyed
    • GaussNewtonSolver
  • The ladder
    • Why the stages are callables
    • The ladder starts in the middle
    • coarsening_stages()
    • feasible_ladder()
    • run_multiple_shooting()
• Constrained transcription (pybnf.transcription)
  • Errors
    • TranscriptionError
  • Augmented variable layout
    • Space
    • The homotopy seam
    • AugmentedLayout
    • QUALIFIER
    • VariableBlock
  • Equality residuals and Jacobians
    • BlockJacobian
    • EqualityModel
    • EqualitySystem
    • JacobianBlock
  • The augmented Lagrangian at a point
    • The three forms, and why the least-squares one is exact
    • The invariant the noise scale depends on
    • AugmentedModel
    • AugmentedSubproblem
    • Multipliers
    • ObjectiveModel
    • TranscriptionProblem
  • The outer loop
    • The inner-solver contract
    • The schedule, and why it starts tight
    • Stalling
    • The defect report
    • Best-iterate certification
    • AugmentedLagrangian
    • Certificate
    • CertifiedBest
    • CertifiedIterate
    • InnerOutcome
    • OuterResult
    • PenaltySchedule
    • WORST_DEFECTS
    • projected_gradient_norm()
  • The transcription homotopy
    • Two rules the driver follows, and the asymmetry between them
    • HomotopyResult
    • StageResult
    • coarsening_ladder()
    • run_homotopy()