Command line interface#

Pharmpy has a command line interface for some of its functionality. The command line tool is an integrated part of the Pharmpy python package.

The main command is `pharmpy` and the functionality is divided into subcommands with two levels. The first and top level most often represents a type of object to perform on operation on. For example `model` or `data`. (The built in help system will give an overview of available subcommands with `pharmpy -h`. The second level is most often an operation or verb to perform. After the main command and the two subcommands follows the input and options of the particular command.

pharmpy#

Welcome to the command line interface of Pharmpy!

Functionality is split into various subcommands

try –help after a COMMAND
all keyword arguments can be abbreviated if unique

usage: pharmpy [-h] [--version] COMMAND ...

-h, --help#: show this help message and exit

--version#: show program’s version number and exit

Examples:

# Create 100 bootstrap datasets pharmpy data resample pheno_real.mod –resamples=100 –replace

# prettyprint model pharmpy model print pheno_real.mod

# version/install information pharmpy info

pharmpy data#

usage: pharmpy data [-h] ACTION ...

-h, --help#: show this help message and exit

pharmpy data append#

Append a column to dataset given an assignment expression.The expression can contain already present columns of the dataset.

usage: pharmpy data append [-h] [-f] [-o file] FILE expression

file#: input model or dataset file

expression#

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

pharmpy data deidentify#

Deidentify dataset by renumbering the id column and changing dates.

usage: pharmpy data deidentify [-h] [-f] [-o file] [--idcol COLUMN]
                               [--datecols COLUMNS]
                               FILE

file#: A csv file dataset

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

--idcol <column>#: id column name (default ID)

--datecols <columns>#: Comma separated list of date column names

pharmpy data filter#

Filter rows of dataset via expressions. All rows matching all expressions will be kept.abs A new model with the filtered dataset connected is created.

usage: pharmpy data filter [-h] [-f] [-o file] FILE ...

file#: input model or dataset file

expressions#

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

pharmpy data print#

Print whole dataset or selected columns from model or csv file via a pager to stdout. For NM-TRAN models the dataset will be filtered first.

usage: pharmpy data print [-h] [--columns COLUMNS] FILE

file#: input model or dataset file

-h, --help#: show this help message and exit

--columns <columns>#: Select specific columns (default is all)

pharmpy data reference#

Set values in specific columns to provided reference values

usage: pharmpy data reference [-h] [-f] [-o file]
                              FILE COLNAME=VALUE [COLNAME=VALUE ...]

file#: input model file

colname=value#: List of pairs of column names and reference values

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

pharmpy data resample#

Bootstrap resample datasetsMultiple new models and datasets will be created

usage: pharmpy data resample [-h] [--seed INTEGER] [--group COLUMN]
                             [--resamples NUMBER] [--stratify COLUMN]
                             [--replace] [--sample_size NUMBER]
                             FILE

file#: input model or dataset file

-h, --help#: show this help message and exit

--seed <integer>#: Provide a random seed. The seed must be an integer

--group <column>#: Column to use for grouping (default is ID)

--resamples <number>#: Number of resampled datasets (default 1)

--stratify <column>#: Column to use for stratification

--replace#: Sample with replacement (default is without)

--sample_size <number>#: Number of groups to sample for each resample

pharmpy data write#

Write a dataset from model as the model sees it. For NM-TRAN models this means to filter all IGNORE and ACCEPT statements in $DATA and to convert the dataset to csv format.

usage: pharmpy data write [-h] [-f] [-o file] FILE

file#: input model file

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

pharmpy info#

usage: pharmpy info [-h]

-h, --help#: show this help message and exit

pharmpy model#

usage: pharmpy model [-h] ACTION ...

-h, --help#: show this help message and exit

pharmpy model print#

Print an overview of a model.

usage: pharmpy model print [-h] [--explicit-odes] FILE [FILE ...]

file#: input model files

-h, --help#: show this help message and exit

--explicit-odes#: Print the ODE system explicitly instead of as a compartmental graph

pharmpy model sample#

Sample parameter initial estimates using uncertainty givenby covariance matrix.

usage: pharmpy model sample [-h] [--seed INTEGER] [--samples NUMBER] FILE

file#: input model file

-h, --help#: show this help message and exit

--seed <integer>#: Provide a random seed. The seed must be an integer

--samples <number>#: Number of sampled models

pharmpy model update_inits#

Update inits using modelfit results.

usage: pharmpy model update_inits [-h] [-f] [-o file] FILE

file#: input model file

-h, --help#: show this help message and exit

-f, --force#: remove existing destination files (all)

-o <file>, --output_file <file>#: output file

pharmpy psn#

usage: pharmpy psn [-h] ACTION ...

-h, --help#: show this help message and exit

pharmpy psn bootstrap#

Generate results from a PsN bootstrap run

usage: pharmpy psn bootstrap [-h] PsN directory

psn directory#: Path to PsN bootstrap run directory

-h, --help#: show this help message and exit

pharmpy psn cdd#

Generate results from a PsN cdd run

usage: pharmpy psn cdd [-h] PsN directory

psn directory#: Path to PsN cdd run directory

-h, --help#: show this help message and exit

pharmpy psn frem#

Generate results from a PsN frem run

usage: pharmpy psn frem [-h] [--method {cov_sampling,bipp}]
                        [--force_posdef_covmatrix]
                        [--force_posdef_samples FORCE_POSDEF_SAMPLES]
                        PsN directory

psn directory#: Path to PsN frem run directory

-h, --help#: show this help message and exit

--method {cov_sampling,bipp}#: Method to use for uncertainty of covariate effects

--force_posdef_covmatrix#: Should covariance matrix be forced to become positive definite

--force_posdef_samples <force_posdef_samples>#: Number of sampling tries to do before starting to force posdef

pharmpy psn linearize#

Generate results from a PsN linearize run

usage: pharmpy psn linearize [-h] PsN directory

psn directory#: Path to PsN linearize run directory

-h, --help#: show this help message and exit

pharmpy psn print#

Print results from PsN run to stdout

usage: pharmpy psn print [-h] file or directory

file or directory#: Path to directory containing results.json or directly to json results file

-h, --help#: show this help message and exit

pharmpy psn qa#

Generate results from a PsN qa run

usage: pharmpy psn qa [-h] PsN directory

psn directory#: Path to PsN qa run directory

-h, --help#: show this help message and exit

pharmpy psn report#

Generate results report for PsN run (currently only frem)

usage: pharmpy psn report [-h] PsN directory

psn directory#: Path to PsN run directory

-h, --help#: show this help message and exit

pharmpy psn ruvsearch#

Generate results from a PsN ruvsearch run

usage: pharmpy psn ruvsearch [-h] PsN directory

psn directory#: Path to PsN ruvsearch run directory

-h, --help#: show this help message and exit

pharmpy psn scm#

Generate results from a PsN scm run

usage: pharmpy psn scm [-h] PsN directory

psn directory#: Path to PsN scm run directory

-h, --help#: show this help message and exit

pharmpy psn simeval#

Generate results from a PsN simeval run

usage: pharmpy psn simeval [-h] PsN directory

psn directory#: Path to PsN simeval run directory

-h, --help#: show this help message and exit

pharmpy results#

usage: pharmpy results [-h] ACTION ...

-h, --help#: show this help message and exit

pharmpy results summary#

Print a summary of a model estimates to stdout.

usage: pharmpy results summary [-h] FILE [FILE ...]

file#: input model files

-h, --help#: show this help message and exit

pharmpy run#

usage: pharmpy run [-h] TOOL ...

-h, --help#: show this help message and exit

pharmpy run allometry#

usage: pharmpy run allometry [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER]
                             [--allometric_variable ALLOMETRIC_VARIABLE]
                             [--reference_value REFERENCE_VALUE]
                             [--parameters PARAMETERS] [--initials INITIALS]
                             [--lower_bounds LOWER_BOUNDS]
                             [--upper_bounds UPPER_BOUNDS] [--non_fixed]
                             [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--allometric_variable <allometric_variable>#: Name of the variable to use for allometric scaling

--reference_value <reference_value>#: Reference value for the allometric variable

--parameters <parameters>#: Parameters to apply scaling to

--initials <initials>#: Initial estimates for the exponents

--lower_bounds <lower_bounds>#: Lower bounds for the exponents

--upper_bounds <upper_bounds>#: Upper bounds for the exponents

--non_fixed#: Should the exponents not be fixed

--path <path>#: Path to output directory

pharmpy run amd#

usage: pharmpy run amd [-h] [--seed INTEGER] [--broadcaster NAME]
                       [--dispatcher NAME] [--ncores NUMBER]
                       [--results RESULTS] [--modeltype MODELTYPE]
                       [--administration ADMINISTRATION] [--strategy STRATEGY]
                       [--cl_init CL_INIT] [--vc_init VC_INIT]
                       [--mat_init MAT_INIT] [--b_init B_INIT]
                       [--emax_init EMAX_INIT] [--ec50_init EC50_INIT]
                       [--met_init MET_INIT] [--search_space SEARCH_SPACE]
                       [--lloq_method LLOQ_METHOD] [--lloq_limit LLOQ_LIMIT]
                       [--allometric_variable ALLOMETRIC_VARIABLE]
                       [--occasion OCCASION] [--strictness STRICTNESS]
                       [--dv_types COLNAME=VALUE [COLNAME=VALUE ...]]
                       [--mechanistic_covariates MECHANISTIC_COVARIATES]
                       [--retries_strategy RETRIES_STRATEGY]
                       [--parameter_uncertainty_method PARAMETER_UNCERTAINTY_METHOD]
                       [--ignore_datainfo_fallback IGNORE_DATAINFO_FALLBACK]
                       [--resume RESUME] [--path PATH]
                       FILE

file#: input model or dataset path

-h, --help#: show this help message and exit

--seed <integer>#: Provide a random seed. The seed must be an integer

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--results <results>#: Reults of input if input is a model

--modeltype <modeltype>#: Type of model to build. Valid strings are “basic_pk”, “pkpd”, “drug_metabolite” and “tmdd”

--administration <administration>#: Route of administration. Either “iv”, “oral” or “ivoral”

--strategy <strategy>#: Run algorithm for AMD procedure. Valid options are “default”, “reevaluation”

--cl_init <cl_init>#: Initial estimate for the population clearance

--vc_init <vc_init>#: Initial estimate for the central compartment population volume

--mat_init <mat_init>#: Initial estimate for the mean absorption time (not for iv models)

--b_init <b_init>#: Initial estimate for the baseline (PKPD model)

--emax_init <emax_init>#: Initial estimate for E_max (PKPD model)

--ec50_init <ec50_init>#: Initial estimate for EC_50 (PKPD model)

--met_init <met_init>#: Initial estimate for mean equilibration time (PKPD model)

--search_space <search_space>#: MFL for search space for structural and covariate model

--lloq_method <lloq_method>#: Method for how to remove LOQ data

--lloq_limit <lloq_limit>#: Lower limit of quantification. If None LLOQ column from dataset will be used

--allometric_variable <allometric_variable>#: Variable to use for allometry

--occasion <occasion>#: Name of occasion column

--strictness <strictness>#: Strictness criteria

--dv_types <colname=value>#

--mechanistic_covariates <mechanistic_covariates>#: List of covariates or tuple of covariate and parameter combination to run in a separate proioritized covsearch run

--retries_strategy <retries_strategy>#: Whether or not to run retries tool

--parameter_uncertainty_method <parameter_uncertainty_method>#: Parameter uncertainty method

--ignore_datainfo_fallback <ignore_datainfo_fallback>#: Ignore using datainfo to get information not given by the user

--resume <resume>#: Whether to allow resuming previous run

--path <path>#: Path to run AMD in

pharmpy run bootstrap#

usage: pharmpy run bootstrap [-h] [--seed INTEGER] [--broadcaster NAME]
                             [--dispatcher NAME] [--ncores NUMBER]
                             [--samples SAMPLES] [--dofv] [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--seed <integer>#: Provide a random seed. The seed must be an integer

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--samples <samples>#: Number of bootstrap datasets

--dofv#: Also run evaluation of the bootstrap models on the original dataset

--path <path>#: Path to output directory

pharmpy run covsearch#

usage: pharmpy run covsearch [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--search_space SEARCH_SPACE]
                             [--p_forward P_FORWARD] [--p_backward P_BACKWARD]
                             [--max_steps MAX_STEPS] [--algorithm ALGORITHM]
                             [--max_eval MAX_EVAL]
                             [--adaptive_scope_reduction ADAPTIVE_SCOPE_REDUCTION]
                             [--strictness STRICTNESS]
                             [--naming_index_offset NAMING_INDEX_OFFSET]
                             [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--search_space <search_space>#: MFL of covariate effects to try

--p_forward <p_forward>#: The p-value to use in the likelihood ratio test for forward steps

--p_backward <p_backward>#: The p-value to use in the likelihood ratio test for backward steps

--max_steps <max_steps>#: The maximum number of search steps to make

--algorithm <algorithm>#: The search algorithm to use

--max_eval <max_eval>#: Limit the number of function evaluations to 3.1 times that of the base model

--adaptive_scope_reduction <adaptive_scope_reduction>#: Stash all non-significant parameter-covariate effects to be tested after all significant effects have been tested. Once all these have been tested, try adding the stashed effects once more with a regular forward approach

--strictness <strictness>#: Strictness criteria

--naming_index_offset <naming_index_offset>#: Index offset for naming of runs

--path <path>#: Path to output directory

pharmpy run estmethod#

usage: pharmpy run estmethod [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--methods METHODS]
                             [--solvers SOLVERS]
                             [--parameter_uncertainty_methods PARAMETER_UNCERTAINTY_METHODS]
                             [--path PATH]
                             FILE algorithm compare_ofv

file#: input model file

algorithm#: Algorithm to use

compare_ofv#: Whether to compare the OFV between candidates

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--methods <methods>#: List of estimation methods to test. Can be specified as “all”, a list of estimation methods, or not specify (to not test any estimation method)

--solvers <solvers>#: List of solvers to test. Can be specified as “all”, a list of solvers, or not specify (to not test any solver)

--parameter_uncertainty_methods <parameter_uncertainty_methods>#: List of parameter uncertainty methods to test. Can be specified as “all”, a list of uncertainty methods, or not specify (to not test any uncertainty method)

--path <path>#: Path to output directory

pharmpy run execute#

usage: pharmpy run execute [-h] [--broadcaster NAME] [--dispatcher NAME]
                           [--ncores NUMBER]
                           FILE [FILE ...]

file#: input model files

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

pharmpy run iivsearch#

usage: pharmpy run iivsearch [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--algorithm ALGORITHM]
                             [--iiv_strategy IIV_STRATEGY]
                             [--rank_type RANK_TYPE] [--cutoff CUTOFF]
                             [--linearize] [--keep KEEP]
                             [--strictness STRICTNESS]
                             [--correlation_algorithm CORRELATION_ALGORITHM]
                             [--e_p E_P] [--e_q E_Q] [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--algorithm <algorithm>#: Which algorithm to run when determining number of IIVs

--iiv_strategy <iiv_strategy>#: If/how IIV should be added to start model

--rank_type <rank_type>#: Which ranking type should be used

--cutoff <cutoff>#: Cutoff for which value of the ranking function that is considered significant

--linearize#: Whether or not use linearization when running the tool

--keep <keep>#: List of IIVs to keep

--strictness <strictness>#: Strictness criteria

--correlation_algorithm <correlation_algorithm>#: Which algorithm to run for the determining block structure of added IIVs

--e_p <e_p>#: Expected number of predictors for diagonal elements (used for mBIC)

--e_q <e_q>#: Expected number of predictors for off-diagonal elements (used for mBIC)

--path <path>#: Path to output directory

pharmpy run iovsearch#

usage: pharmpy run iovsearch [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--column COLUMN]
                             [--list_of_parameters LIST_OF_PARAMETERS]
                             [--rank_type RANK_TYPE] [--cutoff CUTOFF]
                             [--distribution DISTRIBUTION]
                             [--strictness STRICTNESS] [--e E] [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--column <column>#: Name of column in dataset to use as occasion column (default is “OCC”)

--list_of_parameters <list_of_parameters>#: List of parameters to test IOV on (if not specified, allparameters with IIV will be tested)

--rank_type <rank_type>#: Which ranking type should be used

--cutoff <cutoff>#: Cutoff for which value of the ranking function that is considered significant

--distribution <distribution>#: Which distribution added IOVs should have (default is same-as-iiv)

--strictness <strictness>#: Strictness criteria

--e <e>#: Expected number of predictors (used for mBIC)

--path <path>#: Path to output directory

pharmpy run linearize#

usage: pharmpy run linearize [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--path <path>#: Path to output directory

pharmpy run modelsearch#

usage: pharmpy run modelsearch [-h] [--broadcaster NAME] [--dispatcher NAME]
                               [--ncores NUMBER] [--algorithm ALGORITHM]
                               [--iiv_strategy IIV_STRATEGY]
                               [--rank_type RANK_TYPE] [--cutoff CUTOFF]
                               [--strictness STRICTNESS] [--e E] [--path PATH]
                               FILE mfl

file#: input model file

mfl#: Search space to test

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--algorithm <algorithm>#: Algorithm to use

--iiv_strategy <iiv_strategy>#: If/how IIV should be added to candidate models

--rank_type <rank_type>#: Which ranking type should be used

--cutoff <cutoff>#: Cutoff for which value of the ranking function that is considered significant

--strictness <strictness>#: Strictness criteria

--e <e>#: Expected number of predictors (used for mBIC)

--path <path>#: Path to output directory

pharmpy run ruvsearch#

usage: pharmpy run ruvsearch [-h] [--broadcaster NAME] [--dispatcher NAME]
                             [--ncores NUMBER] [--groups GROUPS]
                             [--p_value P_VALUE] [--skip SKIP]
                             [--max_iter MAX_ITER] [--dv DV]
                             [--strictness STRICTNESS] [--path PATH]
                             FILE

file#: input model file

-h, --help#: show this help message and exit

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--groups <groups>#: The number of bins to use for the time varying models

--p_value <p_value>#: The p-value to use for the likelihood ratio test

--skip <skip>#: List of models to not test

--max_iter <max_iter>#: Number of iterations to run (1, 2, or 3)

--dv <dv>#: Which DV to assess the error model for

--strictness <strictness>#: Strictness criteria

--path <path>#: Path to output directory

pharmpy run vpc#

usage: pharmpy run vpc [-h] [--seed INTEGER] [--broadcaster NAME]
                       [--dispatcher NAME] [--ncores NUMBER] --samples SAMPLES
                       [--stratify STRATIFY] [--path PATH]
                       FILE

file#: input model file

-h, --help#: show this help message and exit

--seed <integer>#: Provide a random seed. The seed must be an integer

--broadcaster <name>#: Name of the broadcaster to use for log messages

--dispatcher <name>#: Name of the dispatcher to use

--ncores <number>#: Number of cores to use

--samples <samples>#: Number of samples

--stratify <stratify>#: Column to stratify on

--path <path>#: Path to output directory