Package 'n2kanalysis'

Title: Generic Functions to Analyse Data from the 'Natura 2000' Monitoring
Description: All generic functions and classes for the analysis for the 'Natura 2000' monitoring. The classes contain all required data and definitions to fit the model without the need to access other sources. Potentially they might need access to one or more parent objects. An aggregation object might for example need the result of an imputation object. The actual definition of the analysis, using these generic function and classes, is defined in dedictated analysis R packages for every monitoring scheme. For example 'abvanalysis' and 'watervogelanalysis'.
Authors: Thierry Onkelinx [aut, cre] (<https://orcid.org/0000-0001-8804-4216>, Research Institute for Nature and Forest (INBO)), Research Institute for Nature and Forest (INBO) [cph, fnd]
Maintainer: Thierry Onkelinx <[email protected]>
License: GPL-3
Version: 0.3.2
Built: 2024-12-08 06:48:50 UTC
Source: https://github.com/inbo/n2kanalysis

Help Index

Combine multiple n2k objects

Description

Combine multiple n2k objects

Usage

combine(...)

## S4 method for signature 'n2kAnalysisVersion'
combine(...)

## S4 method for signature 'n2kParameter'
combine(...)

## S4 method for signature 'n2kAnomaly'
combine(...)

## S4 method for signature 'n2kResult'
combine(...)

Arguments

...

n2k objects

Delete an n2kModel object

Description

Delete an n2kModel object

Usage

delete_model(x, base, project)

## S4 method for signature 'ANY,character'
delete_model(x, base, project)

## S4 method for signature 'ANY,s3_bucket'
delete_model(x, base, project)

Arguments

x

the file fingerprint of the n2kModel
base

the base location
project

will be a relative path within the base location

Display a message.

Description

This is a short cut for if(verbose) message(x).

Usage

display(verbose, message, linefeed = TRUE)

Arguments

verbose

A logical. When TRUE print the message. When FALSE do nothing.
message

a vector passed to message().
linefeed

A logical. When TRUE append a newline character at the end of the message.

Extract the relevant coefficients

Description

Extract the relevant coefficients

Usage

extract(extractor, object, base, project)

## S4 method for signature 'ANY,character'
extract(extractor, object, base, project)

## S4 method for signature 'ANY,n2kInla'
extract(extractor, object, base = NULL, project = NULL)

Arguments

extractor

the extractor function
object

the n2kModel object
base

the optional base location of the object
project

the optional subdirectory

Value

the relevant coefficients

Fit the model to the analysis files

Description

Fit the model to the analysis files

Usage

fit_every_model(path, status, verbose = TRUE, n_cluster = 1, ...)

Arguments

path

The path containing the analysis files.
status

A vector with status levels to (re-)fit the model. Defaults to c("new", "waiting") when missing.
verbose

Show the name of the current analysis file on screen. Defaults to TRUE
n_cluster

The number of clusters to use.
...

Arguments passed to fit_model()

Fit an n2kModel object

Description

Fit an n2kModel object

Usage

fit_model(x, ...)

## S4 method for signature 'character'
fit_model(
  x,
  base,
  project,
  status = c("new", "waiting"),
  verbose = TRUE,
  ...,
  bucket
)

## S4 method for signature 'n2kAggregate'
fit_model(x, ...)

## S4 method for signature 'n2kComposite'
fit_model(x, base, project, status = "new", ...)

## S4 method for signature 'n2kHurdleImputed'
fit_model(x, base, project, status = c("new", "waiting"), ...)

## S4 method for signature 'n2kInla'
fit_model(
  x,
  status = "new",
  ...,
  timeout = NULL,
  seed = get_seed(x),
  num_threads = NULL,
  parallel_configs = TRUE
)

## S4 method for signature 'n2kInlaComparison'
fit_model(x, base, project, status = "new", verbose = TRUE, ...)

## S4 method for signature 'n2kManifest'
fit_model(
  x,
  base,
  project,
  status = c("new", "waiting"),
  verbose = TRUE,
  ...,
  local = tempfile("fit_model"),
  first = FALSE
)

## S4 method for signature 'n2kModelImputed'
fit_model(x, ...)

## S4 method for signature 's3_object'
fit_model(x, status = c("new", "waiting"), ...)

Arguments

x

the n2kModel
...

other arguments. See details
base

The root of a project. Can be either a directory on a file system or an AWS S3 bucket object. Extracted from bucket or x when missing.
project

The subdirectory of the project. Is relative the base. Extracted from x when missing.
status

A vector with status levels naming the levels which should be calculated. Defaults to "new".
verbose

A logical indicating if the function should display the name of the file and the status. Defaults to TRUE.
bucket

The name of the AWS S3 bucket. Only used when base is missing.
timeout

the optional number of second until the model will time out
seed

See the same argument in INLA::inla.qsample() for further information. In order to produce reproducible results, you ALSO need to make sure the RNG in R is in the same state, see the example in INLA::inla.posterior.sample(). When seed is non-zero, num_threads is forced to "1:1" and parallel_configs is set to FALSE, since parallel sampling would not produce a reproducible sequence of pseudo-random numbers.
num_threads

The number of threads to use in the format "A:B" defining the number threads in the outer (A) and inner (B) layer for nested parallelism. ⁠A "0"⁠ will be replaced intelligently. seed != 0 requires serial computations.
parallel_configs

Logical. If TRUE and not on Windows, then try to run each configuration in parallel (not Windows) using A threads (see num_threads), where each of them is using B:0 threads.
local

A local folder into which objects from an AWS S3 bucket are downloaded.
first

A logical. first = TRUE implies to fit only the first object in the manifest with matching status. first = FALSE implies to fit all objects in the manifest with matching status. Defaults to FALSE.

Details

  • status: A vector with status levels naming the levels which should be recalculated. Defaults to c("new", "waiting").

  • verbose: A logical indicating if the function should display the name of the file and the status. Defaults to TRUE.

Get the analysis date of an n2kModel object

Description

Get the analysis date of an n2kModel object

Usage

get_analysis_date(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_analysis_date(x)

Arguments

x

the n2kModel object

Value

the analysis date of the object

Get an analysis version

Description

Get an analysis version

Usage

get_analysis_version(version)

## S4 method for signature 'data.frame'
get_analysis_version(version)

## S4 method for signature 'n2kAnalysisMetadata'
get_analysis_version(version)

## S4 method for signature 'sessionInfo'
get_analysis_version(version)

Arguments

version

the object to extract the version

Get the anomalies from a model

Description

Get the anomalies from a model

Usage

get_anomaly(analysis, ...)

## S4 method for signature 'n2kInla'
get_anomaly(
  analysis,
  n = 20,
  expected_ratio = 5,
  expected_absent = 5,
  random_threshold = 1.05,
  verbose = TRUE,
  ...
)

## S4 method for signature 'n2kModel'
get_anomaly(analysis, verbose = TRUE, ...)

Arguments

analysis

The model to add
...

Extra options. See details.
n

the number of anomalies per category.
expected_ratio

Observations that have observed / fitted > expected_ratio or fitted / observed > expected_ratio are potential anomalies. Defaults to 5, which implies that observed values that are 5 times higher or lower than the fitted values are potential anomalies.
expected_absent

Zero observations where fitted > expected_absent are potential anomalies.
random_threshold

The minimal relative effect size of a random effect. Random effect with a smaller effect size will never be an anomaly. Defaults to 1.05 (5%).
verbose

Print extra information on the screen

Get the data of an n2kModel object

Description

Get the data of an n2kModel object

Usage

get_data(x)

## S4 method for signature 'n2kInla'
get_data(x)

Arguments

x

the n2kModel object

Value

the data of the object

Get the file fingerprint of an n2kModel object

Description

Get the file fingerprint of an n2kModel object

Usage

get_file_fingerprint(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_file_fingerprint(x)

## S4 method for signature 'n2kManifest'
get_file_fingerprint(x)

Arguments

x

the n2kModel object

Value

the file fingerprint of the object

Get the formula of an n2kModel object

Description

Get the formula of an n2kModel object

Usage

get_formula(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_formula(x)

Arguments

x

the n2kModel object

Value

the formula of the object

Get the location group id of an n2kModel object

Description

Get the location group id of an n2kModel object

Usage

get_location_group_id(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_location_group_id(x)

Arguments

x

the n2kModel object

Value

the location group id of the object

Get the model of a n2kModel object

Description

Get the model of a n2kModel object

Usage

get_model(x)

## S4 method for signature 'n2kInla'
get_model(x)

## S4 method for signature 'character'
get_model(x)

Arguments

x

the n2kModel object

Value

the model of the object

Add the model parameters from a model

Description

Add the model parameters from a model

Usage

get_model_parameter(analysis, ...)

## S4 method for signature 'n2kAggregate'
get_model_parameter(analysis, ...)

## S4 method for signature 'n2kComposite'
get_model_parameter(analysis, ...)

## S4 method for signature 'n2kHurdleImputed'
get_model_parameter(analysis, ...)

## S4 method for signature 'n2kInla'
get_model_parameter(analysis, verbose = TRUE, ...)

## S4 method for signature 'n2kInlaComparison'
get_model_parameter(analysis, ...)

## S4 method for signature 'n2kModelImputed'
get_model_parameter(analysis, ...)

Arguments

analysis

The model to add.
...

extra options
verbose

Print extra information on the screen

Get the model type of an n2kModel object

Description

Get the model type of an n2kModel object

Usage

get_model_type(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_model_type(x)

Arguments

x

the n2kModel object

Value

the model type of the object

Get the object of the parent analyses of an n2kModel object

Description

Get the object of the parent analyses of an n2kModel object

Usage

get_parents(child, base, project)

Arguments

child

the child object
base

the base location to read the model
project

will be a relative path within the base location

Add the results from an analysis

Description

Add the results from an analysis

Usage

get_result(x, ...)

## S4 method for signature 'character'
get_result(x, n_cluster = 1, verbose = TRUE, ...)

## S4 method for signature 'n2kInla'
get_result(x, verbose = TRUE, ...)

## S4 method for signature 'n2kModel'
get_result(x, verbose = TRUE, ...)

## S4 method for signature 'n2kManifest'
get_result(x, ..., base, project, verbose = TRUE)

## S4 method for signature 's3_object'
get_result(x, ...)

Arguments

x

object with the current results
...

further arguments (see Details)
n_cluster

The number of clusters to run this function in parallel. Defaults to 1 (= no parallel computing).
verbose

Print extra information on the screen
base

the base location to read the model
project

will be a relative path within the base location

Get the scheme_id of an n2kModel object

Description

Get the scheme_id of an n2kModel object

Usage

get_scheme_id(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_scheme_id(x)

Arguments

x

the n2kModel object

Value

the scheme_id of the object

Get the seed of an n2kModel object

Description

Get the seed of an n2kModel object

Usage

get_seed(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_seed(x)

Arguments

x

the n2kModel object

Value

the seed of the object

Get the species group id of an n2kModel object

Description

Get the species group id of an n2kModel object

Usage

get_species_group_id(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_species_group_id(x)

Arguments

x

the n2kModel object

Value

the species group id of the object

Get the status fingerprint of an n2kModel object

Description

Get the status fingerprint of an n2kModel object

Usage

get_status_fingerprint(x)

## S4 method for signature 'n2kAnalysisMetadata'
get_status_fingerprint(x)

Arguments

x

the n2kModel object

Value

the status fingerprint of the object

Calculate the mean and 95\ marginal

Description

This is useful for calculating the variance of a random effect when the marginal gives the precision of the random effect.

Usage

inla_inverse(marginal)

Arguments

marginal

The INLA marginal

Convert a manifest yaml file into a bash script

Description

Convert a manifest yaml file into a bash script

Usage

manifest_yaml_to_bash(
  base,
  project,
  hash,
  shutdown = FALSE,
  split = 1,
  status = c("new", "waiting"),
  limit = FALSE,
  timeout = 4
)

## S4 method for signature 's3_bucket'
manifest_yaml_to_bash(
  base,
  project,
  hash,
  shutdown = FALSE,
  split = 1,
  status = c("new", "waiting"),
  limit = FALSE,
  timeout = 4
)

## S4 method for signature 'character'
manifest_yaml_to_bash(
  base,
  project,
  hash,
  shutdown = FALSE,
  split = 1,
  status = c("new", "waiting"),
  limit = FALSE
)

Arguments

base

the base location to store the manifest
project

will be a relative path within the base location
hash

Fingerprint of the manifest yamlfile.
shutdown

Append a shutdown command at the end of the script. Defaults to FALSE.
split

Number of scripts over which to splits the analyses. Default to 1.
status

A vector with status levels naming the levels which should be calculated. Defaults to c("new", "waiting").
limit

Limit bandwidth and CPU usage. Defaults to FALSE.
timeout

number of hours to time out the docker container. Defaults to 4.

Find and mark obsolete datasets

Description

A dataset is obsolete when a newer version is available

Usage

mark_obsolete_dataset(channel)

Arguments

channel

An open ODBC channel

Create an n2kAggregate object

Description

A new n2kAggregate model.

Usage

n2k_aggregate(...)

## S4 method for signature 'ANY'
n2k_aggregate(...)

Arguments

...

other arguments. See below

Details

  • scheme_id: a string holding the id of the scheme.

  • species_group_id: a string identifying the species group.

  • location_group_id: a string identifying the location group.

  • model_type: a string identifying the type of model to fit to the data.

  • first_imported_year: Oldest year considered in the data.

  • last_imported_year: Most recent year considered in the data.

  • duration: The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.

  • last_analysed_year: Most recent year in the window. Defaults to last_imported_year.

  • analysis_date: A POSIXct date indicating the date that the dataset was imported.

  • seed: A single integer used as a seed for all calculations. A random seed will be inserted when missing.

Create an n2kComposite object

Description

A new n2kComposite model is created when parent is a data.frame.

Usage

n2k_composite(parent_status, ...)

## S4 method for signature 'data.frame'
n2k_composite(
  parent_status,
  status = "waiting",
  result_datasource_id,
  scheme_id,
  formula,
  species_group_id,
  location_group_id,
  model_type,
  first_imported_year,
  last_imported_year,
  duration,
  last_analysed_year,
  analysis_date,
  extractor,
  ...,
  seed
)

Arguments

parent_status

A data.frame with columns parent_analysis (the file fingerprint of the parent), parentstatus_fingerprint (the status fingerprint of the parent), and parent_status (the status of the parent).
...

other arguments
status

A single character indicating the status of the model. Defaults to "waiting".
result_datasource_id

A string identifying the data source.
scheme_id

A single integer holding the id of the scheme.
formula

A single character identifying the comparison.
species_group_id

A string identifying the species group.
location_group_id

A string identifying the location group.
model_type

The type of the models. Must start with "inla comparison:".
first_imported_year

Oldest year considered in the data.
last_imported_year

Most recent year considered in the data.
duration

The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.
last_analysed_year

Most recent year in the window. Defaults to last_imported_year.
analysis_date

A POSIXct date indicating the date that the dataset was imported.
extractor

a function to extract the relevant parameters from the model
seed

A single integer used as a seed for all calculations. A random seed will be inserted when missing.

Create an n2kHurdleImputed object

Description

A new n2kHurdleImputed model.

Usage

n2k_hurdle_imputed(presence, count, verbose = FALSE)

## S4 method for signature 'n2kInla'
n2k_hurdle_imputed(presence, count, verbose = FALSE)

Arguments

presence

the n2kInla object for the presence model.
count

the n2kInla object for the count model.
verbose

display the location group ID and species group ID. Defaults to FALSE.

Create an n2kImport object

Description

A new n2kImport model.

Usage

n2k_import(...)

## S4 method for signature 'ANY'
n2k_import(...)

Arguments

...

other arguments. See below

Details

  • scheme_id: a string holding the id of the scheme.

  • species_group_id: a string identifying the species group.

  • location_group_id: a string identifying the location group.

  • model_type: a string identifying the type of model to fit to the data.

  • first_imported_year: Oldest year considered in the data.

  • last_imported_year: Most recent year considered in the data.

  • duration: The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.

  • last_analysed_year: Most recent year in the window. Defaults to last_imported_year.

  • analysis_date: A POSIXct date indicating the date that the dataset was imported.

  • seed: A single integer used as a seed for all calculations. A random seed will be inserted when missing.

  • formula: a string holding the model formula.

  • dataset: A data.frame with filename, fingerprint and import_date.

Create an n2kInla object

Description

A new n2kInla model is created when data is a data.frame.

In case data is an n2kInla object, then only the model and status are updated. All other slots are unaffected.

Usage

n2k_inla(data, ..., model_fit)

## S4 method for signature 'data.frame,ANY'
n2k_inla(
  data,
  status = "new",
  result_datasource_id,
  scheme_id,
  family = "poisson",
  formula,
  species_group_id,
  location_group_id,
  model_type,
  first_imported_year,
  last_imported_year,
  duration,
  last_analysed_year,
  analysis_date,
  lin_comb = NULL,
  minimum = "",
  imputation_size,
  parent = character(0),
  seed,
  replicate_name = list(),
  control = list(),
  parent_status = "converged",
  parent_statusfingerprint,
  extra,
  ...,
  model_fit
)

## S4 method for signature 'n2kInla,inla'
n2k_inla(data, status, raw_imputed = NULL, ..., model_fit)

Arguments

data

a data.frame with the data to analyse
...

other arguments. See below
model_fit

The fitted model
status

A single character indicating the status of the model. Defaults to "waiting".
result_datasource_id

A string identifying the data source.
scheme_id

A single integer holding the id of the scheme.
family

the family to use in the INLA model.
formula

A single character identifying the comparison.
species_group_id

A string identifying the species group.
location_group_id

A string identifying the location group.
model_type

The type of the models. Must start with "inla comparison:".
first_imported_year

Oldest year considered in the data.
last_imported_year

Most recent year considered in the data.
duration

The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.
last_analysed_year

Most recent year in the window. Defaults to last_imported_year.
analysis_date

A POSIXct date indicating the date that the dataset was imported.
lin_comb

A model matrix to calculate linear combinations.
minimum

The name of the variable which holds the minimum counts. Only relevant in case of multiple imputation.
imputation_size

The required number of imputations defaults to 0.
parent

The file fingerprint of the optional parent analysis.
seed

A single integer used as a seed for all calculations. A random seed will be inserted when missing.
replicate_name

A list with the names of replicates. Defaults to an empty list. Used in case of f(X, ..., replicate = Z). Should be a named list like e.g. list(X = c("a", "b", "c")).
control

A named list passed to INLA::inla() when fitting the model.
parent_status

The status of the parent analysis.
parent_statusfingerprint

The status fingerprint of the parent analysis.
extra

a data.frame with extra observations not used in the model. They will be added in subsequent analyses.
raw_imputed

the optional rawImputed object

Create an n2kInlaComparison object

Description

A new n2kInlaComparison model is created when parent is a character.

Usage

n2k_inla_comparison(parent_status, ...)

## S4 method for signature 'data.frame'
n2k_inla_comparison(
  parent_status,
  status = "waiting",
  result_datasource_id,
  scheme_id,
  formula,
  species_group_id,
  location_group_id,
  model_type,
  first_imported_year,
  last_imported_year,
  duration,
  last_analysed_year,
  analysis_date,
  ...,
  seed
)

Arguments

parent_status

A data.frame with columns parent_analysis (the file fingerprint of the parent), parentstatus_fingerprint (the status fingerprint of the parent), and parent_status (the status of the parent).
...

other arguments
status

A single character indicating the status of the model. Defaults to "waiting".
result_datasource_id

A string identifying the data source.
scheme_id

A single integer holding the id of the scheme.
formula

A single character identifying the comparison.
species_group_id

A string identifying the species group.
location_group_id

A string identifying the location group.
model_type

The type of the models. Must start with "inla comparison:".
first_imported_year

Oldest year considered in the data.
last_imported_year

Most recent year considered in the data.
duration

The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.
last_analysed_year

Most recent year in the window. Defaults to last_imported_year.
analysis_date

A POSIXct date indicating the date that the dataset was imported.
seed

A single integer used as a seed for all calculations. A random seed will be inserted when missing.

Create an n2kManifest object

Description

A new n2kManifest object is created when manifest is a data.frame.

Usage

n2k_manifest(manifest)

## S4 method for signature 'data.frame'
n2k_manifest(manifest)

Arguments

manifest

a data.frame with the manifest content

Create an n2kModelImputed object

Description

A new n2kModelImputed model.

Usage

n2k_model_imputed(...)

## S4 method for signature 'ANY'
n2k_model_imputed(...)

Arguments

...

other arguments. See below

Details

  • scheme_id: a string holding the id of the scheme.

  • species_group_id: a string identifying the species group.

  • location_group_id: a string identifying the location group.

  • model_type: a string identifying the type of model to fit to the data.

  • first_imported_year: Oldest year considered in the data.

  • last_imported_year: Most recent year considered in the data.

  • duration: The width of the moving window. Defaults to the last_imported_year - first_imported_year + 1.

  • last_analysed_year: Most recent year in the window. Defaults to last_imported_year.

  • analysis_date: A POSIXct date indicating the date that the dataset was imported.

  • seed: A single integer used as a seed for all calculations. A random seed will be inserted when missing.

The n2kAggregate class

Description

It holds analysis data based on an aggregated imputation

Slots

RawImputed

A rawImputed object with multiple imputations.

Function

The function to apply on each group.

Filter

The object to pass to the filter argument of multimput::aggregate_impute().

Join

The object to pass to the join argument of multimput::aggregate_impute().

AggregatedImputed

An aggregatedImputed object with multiple imputations.

The n2kAnalysisMetadata class

Description

The n2kAnalysisMetadata class

The n2kAnalysisVersion class

Description

The n2kAnalysisVersion class

The n2kAnomaly class

Description

The n2kAnomaly class

The n2kComposite class

Description

Calculate composite indices from multiple analysis

Slots

Extractor

A function to extract the relevant parameters from the model.

Parameter

A data.frame with the relevant parameter estimates of each parent analysis.

Index

The composite index based on the parameters.

The n2kContrast class

Description

The n2kContrast class

The n2kHurdleImputed class

Description

It holds the model2 of hurdle imputed data.

Slots

Presence

A rawImputed object with the presence component.

Count

A rawImputed object with the count component.

Hurdle

A aggregatedImputed object with the resulting hurdle object.

The n2kImport class

Description

It hold the imported datasets

Slots

Dataset

A data.frame with the datasets.

The n2kInla class

Description

It hold analysis data based on an INLA Poisson model

Slots

Data

A data.frame with the data.

LinearCombination

An optional matrix with the linear combinations.

ReplicateName

An optional list with names of replicates.

Model

Either NULL or the resulting INLA model.

Family

The family of the INLA model.

Control

A named list with options passed to the arguments of INLA::inla().

ImputationSize

The number of multiple imputations. Defaults to 0, indication no multiple imputation.

Minimum

An optional string containing the name of the variable in Data holding the minimal values for imputation.

RawImputed

A rawImputed object with multiple imputations.

Extra

A data.frame with extra data to add to the imputations. This data is not used in the imputation model. It must contain the same variables as the original data.

The n2kInlaComparison class

Description

Compare multiple models using the WAIC criterion

Slots

WAIC

A data.frame with WAIC values per model.

The n2kManifest class

Description

The n2kManifest class

The n2kModel class

Description

A virtual superclass to contain the analysis models

The n2kModelImputed class

Description

It holds the model of aggregated imputed data

Slots

Function

The object to pass to the model.fun argument of multimput::model_impute().

Package

A vector of package names which must be loaded to run the function.

ModelArgs

The object to pass to the model.args argument of multimput::model_impute().

PrepareModelArgs

An optional list containing a single function that will be applied to the object. The result of the function will be appended to the ModelsArgs.

Extractor

The object to pass to the extractor argument of multimput::model_impute().

ExtractorArgs

The object to pass to the extractor.args argument of multimput::model_impute().

Filter

The object to pass to the filter argument of multimput::model_impute().

Mutate

The object to pass to the mutate argument of multimput::model_impute().

AggregatedImputed

An aggregatedImputed object with multiple imputations.

Results

The data.frame with the results of multimput::model_impute().

The n2kParameter class

Description

The n2kParameter class

The n2kResult class

Description

The n2kResult class

Get the parent status of an n2kModel

Description

Get the parent status of an n2kModel

Usage

parent_status(x)

## S4 method for signature 'n2kAnalysisMetadata'
parent_status(x)

Arguments

x

the n2kModel object

Value

the parent status of the object

Overwrite the status of an n2kAnalysisMetadata

Description

Overwrite the status of an n2kAnalysisMetadata

Usage

parent_status(x) <- value

## S4 replacement method for signature 'n2kComposite'
parent_status(x) <- value

Arguments

x

the n2kAnalysisMetadata object
value

the new values for the status

Read a n2kManifest object

Description

Read a n2kManifest object

Usage

read_manifest(base, project, hash)

## S4 method for signature 'character'
read_manifest(base, project, hash)

## S4 method for signature 's3_bucket'
read_manifest(base, project, hash)

Arguments

base

The base location to read the manifest.
project

Will be a relative path within the base location.
hash

Optional the sha1 of the manifest. This can be abbreviated to to first unique characters. The function will return an error in case of multiple matches. If missing, then most recent manifest will be returned.

Read an n2kModel object

Description

Read an n2kModel object

Usage

read_model(x, base, project)

## S4 method for signature 'ANY,character'
read_model(x, base, project)

## S4 method for signature 'ANY,s3_bucket'
read_model(x, base, project)

## S4 method for signature 'ANY,ANY'
read_model(x, base, project)

Arguments

x

the file fingerprint of the n2kModel
base

the base location to read the model
project

will be a relative path within the base location

Read an n2kResult object

Description

Read an n2kResult object

Usage

read_result(x, base, project)

## S4 method for signature 'ANY,character'
read_result(x, base, project)

## S4 method for signature 'ANY,s3_bucket'
read_result(x, base, project)

## S4 method for signature 'ANY,ANY'
read_result(x, base, project)

Arguments

x

the file fingerprint of the n2kResult
base

the base location to read the results
project

will be a relative path within the base location

Aggregate all results estimates in a single dataframe

Description

Aggregate all results estimates in a single dataframe

Usage

result_estimate(x, ...)

## S4 method for signature 'character'
result_estimate(x, ..., base, project)

## S4 method for signature 'n2kResult'
result_estimate(x, ...)

Arguments

x

object with the current results
...

further arguments (see Details)
base

the base location to read the results
project

will be a relative path within the base location

Aggregate all results meta data in a single dataframe

Description

Aggregate all results meta data in a single dataframe

Usage

result_metadata(x, ...)

## S4 method for signature 'character'
result_metadata(x, ..., base, project)

## S4 method for signature 'n2kResult'
result_metadata(x, ...)

Arguments

x

object with the current results
...

further arguments (see Details)
base

the base location to read the results
project

will be a relative path within the base location

Select data based on the number of presences per category

Description

Presences have Count > 0.

Usage

select_factor_count_strictly_positive(
  observation,
  variable,
  threshold,
  relative = FALSE,
  dimension = 1
)

Arguments

observation

the data.frame with observations
variable

the name of the factor
threshold

the minimal threshold
relative

When FALSE the threshold is the number of non-zero observations. When TRUE the threshold is the proportion of non-zero observations. Defaults to FALSE.
dimension

Indicates which element of variable is used for the final aggregation.

Examples

observation <- data.frame(
  Count = c(4, 4, 4, 4, 3, 3, 3, 0, 2, 2, 0, 0),
  LocationID = rep(1:3, each = 4),
  Year = rep(c(1, 1, 1, 1, 2, 2), 2)
)
# Select the locations with at least 3 prescenses
select_factor_count_strictly_positive(
  observation,
  variable = "LocationID",
  threshold = 3
)
# Select those locations in which the species is present in at least 2 years
select_factor_count_strictly_positive(
  observation, variable = c("LocationID", "Year"), threshold = 2
)
# Select those years in which the species is present in at least 2 locations
select_factor_count_strictly_positive(
  observation, variable = c("LocationID", "Year"),
  threshold = 2,
  dimension = 2
)

Select the observations based on the average of a factor

Description

The negative binomial average of the Count variable is calculated for each level of variable. Only the levels which are equal or larger than threshold times the maximal average (in the original scale) are retained.

Usage

select_factor_threshold(observation, variable, threshold)

Arguments

observation

the data.frame with observations
variable

the name of the factor
threshold

the minimal threshold

Examples

observation <- data.frame(
  Count = c(100, 101, 50, 51, 1, 0, 0, 0),
  LocationID = factor(rep(1:4, each = 2))
)
select_factor_threshold(observation, "LocationID", threshold = 0.05)

Limit the observation to the range in which the species is present

Description

Limit the observation to the range in which the species is present

Usage

select_observed_range(observation, variable)

Arguments

observation

the data.frame with observations
variable

the name of the factor

Examples

observation <- data.frame(
  Count = c(0, 0, 100, 101, 0, 51, 1, 0, 0, 0),
  Year = 1:10
)
select_observed_range(observation, "Year")

Convert a sessionInfo() to a data.frame of packages

Description

Convert a sessionInfo() to a data.frame of packages

Usage

session_package(session)

## S4 method for signature 'sessionInfo'
session_package(session)

Arguments

session

The output of sessionInfo()

Value

a data.frame with the packages of a sessionInfo()

Get the status of an n2kModel

Description

Get the status of an n2kModel

Usage

status(x)

## S4 method for signature 'n2kAnalysisMetadata'
status(x)

## S4 method for signature 'character'
status(x)

Arguments

x

the n2kModel object

Value

the status of the object

Overwrite the status of an n2kAnalysisMetadata

Description

Overwrite the status of an n2kAnalysisMetadata

Usage

status(x) <- value

## S4 replacement method for signature 'n2kInla'
status(x) <- value

## S4 replacement method for signature 'n2kComposite'
status(x) <- value

## S4 replacement method for signature 'n2kInlaComparison'
status(x) <- value

## S4 replacement method for signature 'n2kAggregate'
status(x) <- value

## S4 replacement method for signature 'n2kModelImputed'
status(x) <- value

## S4 replacement method for signature 'n2kHurdleImputed'
status(x) <- value

Arguments

x

the n2kAnalysisMetadata object
value

the new values for the status

Store an n2kManifest object

Description

Store an n2kManifest object

Usage

store_manifest(x, base, project)

## S4 method for signature 'ANY,character'
store_manifest(x, base, project)

## S4 method for signature 'ANY,s3_bucket'
store_manifest(x, base, project)

Arguments

x

the n2kManifest
base

the base location to store the manifest
project

will be a relative path within the base location

Store a Docker configuration

Description

Store a Docker configuration

Usage

store_manifest_yaml(x, base, project, docker, dependencies)

## S4 method for signature 'ANY,s3_bucket'
store_manifest_yaml(x, base, project, docker, dependencies)

## S4 method for signature 'ANY,character'
store_manifest_yaml(x, base, project, docker, dependencies)

Arguments

x

the n2kManifest
base

the base location to store the manifest
project

will be a relative path within the base location
docker

the docker image to use
dependencies

extra GitHub packages to install

Store an n2kModel object

Description

Store an n2kModel object

Usage

store_model(x, base, project, overwrite = TRUE, validate = TRUE)

## S4 method for signature 'ANY,character'
store_model(x, base, project, overwrite = TRUE, validate = TRUE)

## S4 method for signature 'ANY,s3_bucket'
store_model(x, base, project, overwrite = TRUE, validate = TRUE)

Arguments

x

The n2kModel.
base

The base location to store the model.
project

Will be a relative path within the base location.
overwrite

Should an existing object be overwritten? Defaults to TRUE.
validate

Check that the object is valid before storing it. Defaults to TRUE.

Make a union of AnalysisVersions

Description

Make a union of AnalysisVersions

Usage

union(...)

Arguments

...

the n2kAnalysisVersions to union

Get the validity of objects in files

Description

Get the validity of objects in files

Usage

## S4 method for signature 'character'
validObject(object, test = FALSE, complete = FALSE)

Arguments

object

a single filename, a single path or a vector with filenames.
test

See methods::validObject()
complete

See methods::validObject()