R: Functions to extract informations from 'BIOMOD.models.out',...

getters.out {biomod2}

R Documentation

Functions to extract informations from `BIOMOD.models.out`, `BIOMOD.projection.out` or `BIOMOD.ensemble.models.out` objects

Description

These functions allow the user to easily retrieve informations stored in the different biomod2 objects from the different modeling steps, such as modeling options and formated data, models used or not, predictions, evaluations, variables importance.

Usage

## S4 method for signature 'BIOMOD.formated.data'
get_species_data(obj)

## S4 method for signature 'BIOMOD.formated.data.PA'
get_species_data(obj)

## S4 method for signature 'BIOMOD.formated.data'
get_eval_data(obj)

## S4 method for signature 'BIOMOD.models.out'
get_options(obj)

## S4 method for signature 'BIOMOD.models.out'
get_calib_lines(obj, as.data.frame = FALSE, PA = NULL, run = NULL)

## S4 method for signature 'BIOMOD.models.out'
get_formal_data(obj, subinfo = NULL)

## S4 method for signature 'BIOMOD.models.out'
get_predictions(
  obj,
  evaluation = FALSE,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  model.as.col = FALSE
)

## S4 method for signature 'BIOMOD.models.out'
get_built_models(obj, full.name = NULL, PA = NULL, run = NULL, algo = NULL)

## S4 method for signature 'BIOMOD.models.out'
get_evaluations(
  obj,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  metric.eval = NULL
)

## S4 method for signature 'BIOMOD.models.out'
get_variables_importance(
  obj,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  expl.var = NULL
)

## S4 method for signature 'BIOMOD.projection.out'
get_projected_models(
  obj,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL
)

## S4 method for signature 'BIOMOD.projection.out'
free(obj)

## S4 method for signature 'BIOMOD.projection.out'
get_predictions(
  obj,
  metric.binary = NULL,
  metric.filter = NULL,
  full.name = NULL,
  PA = NULL,
  run = NULL,
  algo = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL,
  model.as.col = FALSE,
  ...
)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_formal_data(obj, subinfo = NULL)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_built_models(
  obj,
  full.name = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL,
  algo = NULL
)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_kept_models(obj)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_predictions(
  obj,
  evaluation = FALSE,
  full.name = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL,
  algo = NULL,
  model.as.col = FALSE
)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_evaluations(
  obj,
  full.name = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL,
  algo = NULL,
  metric.eval = NULL
)

## S4 method for signature 'BIOMOD.ensemble.models.out'
get_variables_importance(
  obj,
  full.name = NULL,
  merged.by.algo = NULL,
  merged.by.run = NULL,
  merged.by.PA = NULL,
  filtered.by = NULL,
  algo = NULL,
  expl.var = NULL
)

Arguments

`obj`	a `BIOMOD.formated.data`, `BIOMOD.formated.data.PA`, `BIOMOD.models.out`, `BIOMOD.projection.out` or `BIOMOD.ensemble.models.out` object
`as.data.frame`	a `logical` defining whether output should be returned as `data.frame` or `array` object
`PA`	(optional, default `NULL`) A `vector` containing pseudo-absence set to be loaded, must be among `PA1`, `PA2`, `...`, `allData`
`run`	(optional, default `NULL`) A `vector` containing repetition set to be loaded, must be among `RUN1`, `RUN2`, `...`, `allRun`
`subinfo`	a `character` corresponding to the information to be extracted, must be among `NULL`, `expl.var.names`, `resp.var`, `expl.var`, `MinMax`, `eval.resp.var`, `eval.expl.var` (see Details)
`evaluation`	a `logical` defining whether evaluation data should be used or not
`full.name`	(optional, default `NULL`) A `vector` containing model names to be kept, must be either `all` or a sub-selection of model names that can be obtained with the `get_built_models` function
`algo`	(optional, default `NULL`) A `character` containing algorithm to be loaded, must be either `ANN`, `CTA`, `FDA`, `GAM`, `GBM`, `GLM`, `MARS`, `MAXENT`, `MAXNET`, `RF`, `SRE`, `XGBOOST`
`model.as.col`	(optional, default `FALSE`) A `boolean` given to `get_predictions`. If `TRUE` prediction are returned as a wide `data.frame` with each column containing predictions for a single model and corresponding to the old output given by biomod2 in version < 4.2-2. If `FALSE` predictions are returned as a long `data.frame` with many additional informations readily available.
`metric.eval`	(optional, default `NULL`) A `vector` containing evaluation metric to be kept, must be among `POD`, `FAR`, `POFD`, `SR`, `ACCURACY`, `BIAS`, `ROC`, `TSS`, `KAPPA`, `OR`, `ORSS`, `CSI`, `ETS`, `BOYCE`, `MPA`
`expl.var`	(optional, default `NULL`) A `vector` containing explanatory variables to be kept, that can be obtained with the `get_formal_data(obj, subinfo = 'expl.var.names')` function
`merged.by.algo`	(optional, default `NULL`) A `character` containing merged algorithm to be loaded, must be among `ANN`, `CTA`, `FDA`, `GAM`, `GBM`, `GLM`, `MARS`, `MAXENT`, `MAXNET`, `RF`, `SRE`, `XGBOOST`, `mergedAlgo`
`merged.by.run`	(optional, default `NULL`) A `vector` containing merged repetition set to be loaded, must be among `RUN1`, `RUN2`, `...`, `mergedRun`
`merged.by.PA`	(optional, default `NULL`) A `vector` containing merged pseudo-absence set to be loaded, must be among `PA1`, `PA2`, `...`, `mergedData`
`filtered.by`	(optional, default `NULL`) A `vector` containing evaluation metric selected to filter single models to build the ensemble models, must be among `POD`, `FAR`, `POFD`, `SR`, `ACCURACY`, `BIAS`, `ROC`, `TSS`, `KAPPA`, `OR`, `ORSS`, `CSI`, `ETS`, `BOYCE`, `MPA`
`metric.binary`	(optional, default `NULL`) A `vector` containing evaluation metric selected to transform predictions into binary values, must be among `POD`, `FAR`, `POFD`, `SR`, `ACCURACY`, `BIAS`, `ROC`, `TSS`, `KAPPA`, `OR`, `ORSS`, `CSI`, `ETS`, `BOYCE`, `MPA`
`metric.filter`	(optional, default `NULL`) A `vector` containing evaluation metric to filter predictions, must be among `POD`, `FAR`, `POFD`, `SR`, `ACCURACY`, `BIAS`, `ROC`, `TSS`, `KAPPA`, `OR`, `ORSS`, `CSI`, `ETS`, `BOYCE`, `MPA`
`...`	(optional, one or several of the following arguments depending on the selected function)

Value

get_species_data

a data.frame combining data.species, coord, data.env.var (and PA.table) slots of BIOMOD.formated.data (or BIOMOD.formated.data.PA) object

get_eval_data

a data.frame combining eval.data.species, eval.coord, eval.data.env.var slots of BIOMOD.formated.data or BIOMOD.formated.data.PA object

get_options

a BIOMOD.stored.options-class object from the models.options slot of a BIOMOD.models.out-class object

get_calib_lines

a BIOMOD.stored.data.frame-class object from the calib.lines slot of a BIOMOD.models.out object

get_projected_models

a vector from the models.projected slot of a BIOMOD.projection.out object

get_predictions

a BIOMOD.stored.data object from the proj.out slot of a BIOMOD.models.out, BIOMOD.projection.out or BIOMOD.ensemble.models.out object

get_kept_models

a vector containing names of the kept models of a BIOMOD.ensemble.models.out object

get_formal_data

depending on the subinfo parameter :

NULL: a BIOMOD.stored.formated.data-class (or BIOMOD.stored.models.out-class) object from the formated.input.data (or models.out) slot of a BIOMOD.models.out (or BIOMOD.ensemble.models.out) object
expl.var.names: a vector from the expl.var.names slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object
resp.var: a vector from the data.species slot of the formated.input.data slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object
expl.var: a data.frame from the data.env.var slot of the formated.input.data slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object
MinMax: a list of minimum and maximum values (or levels if factorial) of variable contained in the data.env.var slot of the formated.input.data slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object
eval.resp.var: a vector from the eval.data.species slot of the formated.input.data slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object
eval.expl.var: a data.frame from the eval.data.env.var slot of the formated.input.data slot of a BIOMOD.models.out or BIOMOD.ensemble.models.out object

get_built_models

a vector from the models.computed slot (or em.computed) of a BIOMOD.models.out (or BIOMOD.ensemble.models.out) object

get_evaluations

a data.frame from the models.evaluation slot (or model_evaluation of each model in em.computed) of a BIOMOD.models.out (or BIOMOD.ensemble.models.out) object. Contains evaluation metric for different models and dataset. Evaluation metric are calculated on the calibrating data (column calibration), on the cross-validation data (column validation) or on the evaluation data (column evaluation).
For cross-validation data, see CV.[...] parameters in BIOMOD_Modeling function ; for evaluation data, see eval.[...] parameters in BIOMOD_FormatingData.

get_variables_importance

a BIOMOD.stored.data.frame-class from the variables.importance slot (or model_variables_importance of each model in em.models) of a BIOMOD.models.out (or BIOMOD.ensemble.models.out) object

Author(s)

Damien Georges