OdeModelFit: An ODE model fit (R6 class)
In jtimonen/odemodeling: Building, fitting, and validating Bayesian ODE models in 'Stan'
OdeModelFit R Documentation
An ODE model fit (R6 class)
Description
The fields cmdstanr_time, cmdstanr_summary,
and cmdstanr_draws store the output of cmdstanr_fit's
methods $time(), $summary(), and $draws() methods, respectively,
in memory in case cmdstanr_fit gets corrupted (for example
if the CSV files that it reads the data from are destroyed).
Public fields
modelAn object of class OdeModel.
t0Used initial time.
tUsed time points.
solverUsed solver.
dataGiven additional data.
cmdstanr_fitA cmdstanr::CmdStanMCMC or cmdstanr::CmdStanGQ
object.
cmdstanr_timeA list containing output of the $time() method
of cmdstanr_fit.
cmdstanr_summaryA tibble containing output of the $summary()
method of cmdstanr_fit.
cmdstanr_drawsA posterior::draws_array object containing the
output of the $draws() method of cmdstanr_fit.
cmdstanr_metadataA list containing output of the $metadata()
method of cmdstanr_fit.
cmdstanr_outputA list containing output of the $output()
method of cmdstanr_fit.
setup_timeTime it took to call $initialize() when the
OdeModelFit object was created (in seconds).
Methods
Public methods
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Method new()
Create an OdeModelFit object.
Usage
OdeModelFit$new(model, t0, t, solver, data, cmdstanr_fit)
Arguments
modelAn object of class OdeModel (will be deepcopied).
t0Used initial time.
tUsed time points.
solverUsed solver. An object of class OdeSolver.
dataGiven additional data.
cmdstanr_fitA cmdstanr::CmdStanMCMC or cmdstanr::CmdStanGQ
object (will be deepcopied).
Method time()
Get time information.
Usage
OdeModelFit$time()
Returns
A list.
Method info()
Get various information.
Usage
OdeModelFit$info()
Returns
A string.
Method draws()
Get draws (parameters and generated quantities).
Usage
OdeModelFit$draws(variable = NULL, iteration = NULL)
Arguments
variableName of variable.
iterationIndex of iteration.
Returns
A posterior::draws_array object.
Method summary()
Get summary
Usage
OdeModelFit$summary()
Returns
A tibble.
Method niterations()
Get number of post-warmup iterations per MCMC chain.
Usage
OdeModelFit$niterations()
Method nchains()
Get number of MCMC chains.
Usage
OdeModelFit$nchains()
Method ndraws()
Get total number of post-warmup draws.
Usage
OdeModelFit$ndraws()
Method cmdstan_seed()
Get used 'CmdStan' rng seed.
Usage
OdeModelFit$cmdstan_seed()
Method cmdstan_version()
Get used 'CmdStan' version.
Usage
OdeModelFit$cmdstan_version()
Returns
A string.
Method get_t()
Get time points where the model was fitted.
Usage
OdeModelFit$get_t(include_t0 = FALSE)
Arguments
include_t0Should the initial time point be included?
Returns
A numeric vector of length N. If include_t0 is TRUE, length
will be N+1.
Method get_t0()
Get used initial time point t0.
Usage
OdeModelFit$get_t0()
Returns
A numeric value.
Method dim()
Get dimensions of a variable.
Usage
OdeModelFit$dim(variable)
Arguments
variableName of variable.
Returns
A numeric vector, which is the 'Stan' variable dimension,
obtained as metadata$stan_variable_dims[[variable]], where
metadata is the metadata of the cmdstanr::CmdStanMCMC or
cmdstanr::CmdStanGQ object.
Method dim_odesol()
Extract the dimensions of the ODE solution variable.
Usage
OdeModelFit$dim_odesol()
Returns
A numeric vector of length 2, where first element is the
number of time points and second element is the ODE system dimension.
Method extract_unflattened()
Extract array variable draws so that the array is unflattened.
Usage
OdeModelFit$extract_unflattened(variable)
Arguments
variableName of variable.
Returns
A base R array of dimension c(num_draws, ...) where num_draws
is the total number of draws and ... is the 'Stan' variable dimension,
obtained as self$dim(variable).
Method extract_odesol()
Extract the ODE solutions using each parameter draw, in an
unflattened base R array format.
Usage
OdeModelFit$extract_odesol(include_y0 = FALSE)
Arguments
include_y0Should the initial state be included?
Returns
A base R array of dimension c(num_draws, N, D) where
num_draws is the total number of draws and N is the number of
time points and D is the number of ODE system dimensions. If
include_y0 is TRUE, then the N dimension grows to N+1.
Method extract_y0()
Extract the ODE initial states using each parameter draw, in a
base R array format.
Usage
OdeModelFit$extract_y0()
Returns
A base R array of dimension c(num_draws, D) where
num_draws is the total number of draws and D is the number of ODE
system dimensions.
Method extract_odesol_quantile()
Extract quantiles of the ODE solutions in a base R array
format.
Usage
OdeModelFit$extract_odesol_quantile(p, include_y0 = FALSE)
Arguments
pPercentile. A number between 0 and 1. For example p=0.5
corresponds to median.
include_y0Should the initial state be included?
Returns
A base R array of dimension c(N, D) where N is the number of
time points and D is the number of ODE system dimensions. If
include_y0 is TRUE, then the N dimension grows to N+1.
Method loglik()
Extract the log likelihood using each parameter draw.
Usage
OdeModelFit$loglik()
Returns
A posterior::draws_array.
Method extract_odesol_df()
Extract the ODE solutions using each parameter draw, in a
flattened data frame format that is easy to pass as data
to for example ggplot2::ggplot().
Usage
OdeModelFit$extract_odesol_df(
draw_inds = NULL,
include_y0 = FALSE,
ydim_names = NULL
)
Arguments
draw_indsIf this is not NULL, returns ode solutions
corresponding only to given draws.
include_y0Should the initial state be included?
ydim_namesNames of the ODE dimensioins. If NULL, these
are automatically set as "y1", "y2", etc.
Returns
A data.frame.
Method extract_odesol_df_dist()
Extract (quantiles of) the marginal distribution of ODE solutions in a
data frame format that is easy to pass as data
to for example ggplot2::ggplot().
Usage
OdeModelFit$extract_odesol_df_dist(
probs = c(0.1, 0.5, 0.9),
include_y0 = FALSE,
ydim_names = NULL
)
Arguments
probsThe percentile values. A numeric vector where all values
are between 0 and 1.
include_y0Should the initial state be included?
ydim_namesNames of the ODE dimensions. If NULL, these
are automatically set as "y1", "y2", etc.
Returns
A data.frame.
Method plot_odesol()
A quick way to plot the ODE solutions.
Usage
OdeModelFit$plot_odesol(
draw_inds = NULL,
alpha = 0.75,
color = "firebrick",
...
)
Arguments
draw_indsIf this numeric and positive, plots ODE solutions
corresponding only to given draws. If this is 0, all draws are plotted.
If this is NULL, a random subset of at most 100 draws are plotted.
alphaline alpha
colorline color
...other arguments passed to extract_odesol_df
Returns
A ggplot object.
Method plot_odesol_dist()
A quick way to plot the marginal distribution of ODE solutions at
each time point.
Usage
OdeModelFit$plot_odesol_dist(
p = 0.8,
alpha = 0.75,
color = "firebrick",
fill_color = "firebrick",
...
)
Arguments
pWhich percentage central interval to plot?
alphafill alpha
colorline color
fill_colorfill color
...other arguments passed to extract_odesol_df_dist
Returns
A ggplot object.
Method clone()
The objects of this class are cloneable with this method.
Usage
OdeModelFit$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
See Also
Other model fit classes:
OdeModelGQ,
OdeModelMCMC
jtimonen/odemodeling documentation built on Sept. 15, 2024, 4:29 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| OdeModelFit | R Documentation |
An ODE model fit (R6 class)
Description
The fields cmdstanr_time, cmdstanr_summary,
and cmdstanr_draws store the output of cmdstanr_fit's
methods $time(), $summary(), and $draws() methods, respectively,
in memory in case cmdstanr_fit gets corrupted (for example
if the CSV files that it reads the data from are destroyed).
Public fields
modelAn object of class OdeModel.
t0Used initial time.
tUsed time points.
solverUsed solver.
dataGiven additional data.
cmdstanr_fitA cmdstanr::CmdStanMCMC or cmdstanr::CmdStanGQ object.
cmdstanr_timeA list containing output of the
$time()method ofcmdstanr_fit.cmdstanr_summaryA tibble containing output of the
$summary()method ofcmdstanr_fit.cmdstanr_drawsA posterior::draws_array object containing the output of the
$draws()method ofcmdstanr_fit.cmdstanr_metadataA list containing output of the
$metadata()method ofcmdstanr_fit.cmdstanr_outputA list containing output of the
$output()method ofcmdstanr_fit.setup_timeTime it took to call
$initialize()when the OdeModelFit object was created (in seconds).
Methods
Public methods
Method new()
Create an OdeModelFit object.
Usage
OdeModelFit$new(model, t0, t, solver, data, cmdstanr_fit)
Arguments
modelAn object of class OdeModel (will be deepcopied).
t0Used initial time.
tUsed time points.
solverUsed solver. An object of class OdeSolver.
dataGiven additional data.
cmdstanr_fitA cmdstanr::CmdStanMCMC or cmdstanr::CmdStanGQ object (will be deepcopied).
Method time()
Get time information.
Usage
OdeModelFit$time()
Returns
A list.
Method info()
Get various information.
Usage
OdeModelFit$info()
Returns
A string.
Method draws()
Get draws (parameters and generated quantities).
Usage
OdeModelFit$draws(variable = NULL, iteration = NULL)
Arguments
variableName of variable.
iterationIndex of iteration.
Returns
A posterior::draws_array object.
Method summary()
Get summary
Usage
OdeModelFit$summary()
Returns
A tibble.
Method niterations()
Get number of post-warmup iterations per MCMC chain.
Usage
OdeModelFit$niterations()
Method nchains()
Get number of MCMC chains.
Usage
OdeModelFit$nchains()
Method ndraws()
Get total number of post-warmup draws.
Usage
OdeModelFit$ndraws()
Method cmdstan_seed()
Get used 'CmdStan' rng seed.
Usage
OdeModelFit$cmdstan_seed()
Method cmdstan_version()
Get used 'CmdStan' version.
Usage
OdeModelFit$cmdstan_version()
Returns
A string.
Method get_t()
Get time points where the model was fitted.
Usage
OdeModelFit$get_t(include_t0 = FALSE)
Arguments
include_t0Should the initial time point be included?
Returns
A numeric vector of length N. If include_t0 is TRUE, length
will be N+1.
Method get_t0()
Get used initial time point t0.
Usage
OdeModelFit$get_t0()
Returns
A numeric value.
Method dim()
Get dimensions of a variable.
Usage
OdeModelFit$dim(variable)
Arguments
variableName of variable.
Returns
A numeric vector, which is the 'Stan' variable dimension,
obtained as metadata$stan_variable_dims[[variable]], where
metadata is the metadata of the cmdstanr::CmdStanMCMC or
cmdstanr::CmdStanGQ object.
Method dim_odesol()
Extract the dimensions of the ODE solution variable.
Usage
OdeModelFit$dim_odesol()
Returns
A numeric vector of length 2, where first element is the number of time points and second element is the ODE system dimension.
Method extract_unflattened()
Extract array variable draws so that the array is unflattened.
Usage
OdeModelFit$extract_unflattened(variable)
Arguments
variableName of variable.
Returns
A base R array of dimension c(num_draws, ...) where num_draws
is the total number of draws and ... is the 'Stan' variable dimension,
obtained as self$dim(variable).
Method extract_odesol()
Extract the ODE solutions using each parameter draw, in an unflattened base R array format.
Usage
OdeModelFit$extract_odesol(include_y0 = FALSE)
Arguments
include_y0Should the initial state be included?
Returns
A base R array of dimension c(num_draws, N, D) where
num_draws is the total number of draws and N is the number of
time points and D is the number of ODE system dimensions. If
include_y0 is TRUE, then the N dimension grows to N+1.
Method extract_y0()
Extract the ODE initial states using each parameter draw, in a base R array format.
Usage
OdeModelFit$extract_y0()
Returns
A base R array of dimension c(num_draws, D) where
num_draws is the total number of draws and D is the number of ODE
system dimensions.
Method extract_odesol_quantile()
Extract quantiles of the ODE solutions in a base R array format.
Usage
OdeModelFit$extract_odesol_quantile(p, include_y0 = FALSE)
Arguments
pPercentile. A number between 0 and 1. For example
p=0.5corresponds to median.include_y0Should the initial state be included?
Returns
A base R array of dimension c(N, D) where N is the number of
time points and D is the number of ODE system dimensions. If
include_y0 is TRUE, then the N dimension grows to N+1.
Method loglik()
Extract the log likelihood using each parameter draw.
Usage
OdeModelFit$loglik()
Returns
A posterior::draws_array.
Method extract_odesol_df()
Extract the ODE solutions using each parameter draw, in a
flattened data frame format that is easy to pass as data
to for example ggplot2::ggplot().
Usage
OdeModelFit$extract_odesol_df( draw_inds = NULL, include_y0 = FALSE, ydim_names = NULL )
Arguments
draw_indsIf this is not
NULL, returns ode solutions corresponding only to given draws.include_y0Should the initial state be included?
ydim_namesNames of the ODE dimensioins. If
NULL, these are automatically set as"y1","y2", etc.
Returns
A data.frame.
Method extract_odesol_df_dist()
Extract (quantiles of) the marginal distribution of ODE solutions in a
data frame format that is easy to pass as data
to for example ggplot2::ggplot().
Usage
OdeModelFit$extract_odesol_df_dist( probs = c(0.1, 0.5, 0.9), include_y0 = FALSE, ydim_names = NULL )
Arguments
probsThe percentile values. A numeric vector where all values are between 0 and 1.
include_y0Should the initial state be included?
ydim_namesNames of the ODE dimensions. If
NULL, these are automatically set as"y1","y2", etc.
Returns
A data.frame.
Method plot_odesol()
A quick way to plot the ODE solutions.
Usage
OdeModelFit$plot_odesol( draw_inds = NULL, alpha = 0.75, color = "firebrick", ... )
Arguments
draw_indsIf this numeric and positive, plots ODE solutions corresponding only to given draws. If this is
0, all draws are plotted. If this isNULL, a random subset of at most 100 draws are plotted.alphaline alpha
colorline color
...other arguments passed to
extract_odesol_df
Returns
A ggplot object.
Method plot_odesol_dist()
A quick way to plot the marginal distribution of ODE solutions at each time point.
Usage
OdeModelFit$plot_odesol_dist( p = 0.8, alpha = 0.75, color = "firebrick", fill_color = "firebrick", ... )
Arguments
pWhich percentage central interval to plot?
alphafill alpha
colorline color
fill_colorfill color
...other arguments passed to
extract_odesol_df_dist
Returns
A ggplot object.
Method clone()
The objects of this class are cloneable with this method.
Usage
OdeModelFit$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
See Also
Other model fit classes:
OdeModelGQ,
OdeModelMCMC
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.