| Title: | piamModelTests Tools |
|---|---|
| Description: | A collection of R tools provided by the Integrated Assessment Modeling Consortium (IAMC) for data analysis and diagnostics. |
| Authors: | Jan Philipp Dietrich [aut, cre], Cornelia Auer [aut], Anastasis Giannousakis [aut], Christoph Bertram [aut], Falk Benke [aut], Florian Humpenoeder [aut], Lavinia Baumstark [aut] |
| Maintainer: | Jan Philipp Dietrich <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.31.5 |
| Built: | 2024-11-06 05:58:54 UTC |
| Source: | https://github.com/pik-piam/piamModelTests |
A collection of R tools provided by the Integrated Assessment Modeling Consortium (IAMC) for data analysis and diagnostics.
Maintainer: Jan Philipp Dietrich [email protected]
Authors:
Cornelia Auer
Anastasis Giannousakis
Christoph Bertram
Falk Benke
Florian Humpenoeder
Lavinia Baumstark
Encodes data as integer values(corresponding to their factor representatives). NA values are exchanged by 0.
codedAsInteger(input)codedAsInteger(input)
input |
Input data that should be checked, provided as a object related to a data frame. |
values encoded as integers without NAs.
Collects functions which follow a given name pattern and compares their arguments against a list of allowed arguments
collectFunctions( pattern = "^check", globalenv = FALSE, allowed_args = c("x", "cfg") )collectFunctions( pattern = "^check", globalenv = FALSE, allowed_args = c("x", "cfg") )
pattern |
Name pattern the function name should match. Default is to collect functions starting with "check" |
globalenv |
Boolean deciding whether functions in the global environment should be considered or not. |
allowed_args |
Vector of allowed arguments. If a function contains an argument not listed here it will be ignored and a warning will be returned |
a character vector of function calls fulfilling all requirements
Jan Philipp Dietrich
collectFunctions()collectFunctions()
Takes given data and provides it in the required data format (magpie)
createInputData(x, cfg = "CDLINKS", ref = "IAMC", verbose = TRUE, ...)createInputData(x, cfg = "CDLINKS", ref = "IAMC", verbose = TRUE, ...)
x |
Input data that should be checked, provided as a file path to a reporting file,
a quitte object or an object which can be converted to quitte using |
cfg |
Project configuration that should be used (currently available: "CDLINKS"). Either a project name, a path to a
config file or a data frame specifying available variables and corresponding properties as returned by
|
ref |
Reference data for comparison. Either a project name (currently available: "IAMC"), a path to a mif file or a quitte object containing the data. |
verbose |
Boolean influencing the degree of information returned by the function. |
... |
additional data objects which are forwarded to the check functions |
Input named list with elements available for check functions
Cornelia Auer
Example output from the REMIND model
data generated from the REMIND model
Cornelia Auer
Filters given data (input) according to available default variables (cfg). Pre-checks are performed about the consistency of the input data (e.g. illegal or missing variables) and the status (out) returned.
filterInputData(input, cfg = "CDLINKS", globalenv = FALSE, out = NULL)filterInputData(input, cfg = "CDLINKS", globalenv = FALSE, out = NULL)
input |
Named list with elements available for check functions |
cfg |
Project configuration that should be used. Either a project name (currently available: "CDLINKS"), a path to a
config file or a data frame specifying available variables and corresponding properties as returned by
|
globalenv |
Boolean deciding whether functions in the global environment should be considered |
out |
List with status from pre-checks, e.g. illegal or missing variables. |
List with 1) filtered input and 2) status-output about the consistency of the input data
Cornelia Auer
Method: Essentially computes order by the power of ten from a given positive number The order value is then used to compute a shift factor, which is needed to uniquely represent the data values as integer in preCheckDuplicates.
getShiftFactor(positiveNumber)getShiftFactor(positiveNumber)
positiveNumber |
number >= 0 |
order of power of ten for the input value
Runs various diagnostics over a provided data set checking whether the provided data is in line with IAMC database guidelines.
iamCheck( x, pdf = NULL, cfg = "CDLINKS", refData = "IAMC", verbose = FALSE, globalenv = FALSE, pdfStyle = NULL, ... )iamCheck( x, pdf = NULL, cfg = "CDLINKS", refData = "IAMC", verbose = FALSE, globalenv = FALSE, pdfStyle = NULL, ... )
x |
Input data that should be checked, provided as a file path to a reporting file,
a quitte object or an object which can be converted to quitte using |
pdf |
File name used for a PDF containing diagnostic results of the check. If set to NULL no pdf will be written. |
cfg |
Project configuration that should be used. Either a project name (currently available: "CDLINKS"), a path to a
config file or a data frame specifying available variables and corresponding properties as returned by
|
refData |
Reference data for comparison. Either a project name (currently available: "IAMC"), a path to a mif file or a quitte object containing the data. |
verbose |
Boolean influencing the degree of information returned by the function. |
globalenv |
Boolean deciding whether functions in the global environment should be considered or not. |
pdfStyle |
list of style-options for the pdf |
... |
additional data objects which are forwarded to the check functions |
List of all inputs and outputs created by the performed checks (invisible)
Jan Philipp Dietrich
iamProjectConfig, as.quitte, is.quitte
# run check with example data iamCheck(example_REMIND, cfg="CDLINKS")# run check with example data iamCheck(example_REMIND, cfg="CDLINKS")
Function to return available variables and corresponding properties for a given project configuration.
iamProjectConfig(cfg = "CDLINKS")iamProjectConfig(cfg = "CDLINKS")
cfg |
Project configuration that should be used. Either a project name (currently available: "CDLINKS"), a path to a config file or a data frame specifying available variables and corresponding properties |
Data frame containing avaible variables and corresponding properties
Jan Philipp Dietrich
iamProjectConfig()iamProjectConfig()
Function to return historical reference data for validation.
iamReferenceData(ref = "IAMC")iamReferenceData(ref = "IAMC")
ref |
Reference data for comparison. Either a project name (currently available: "IAMC"), a path to a file or a data frame containing the data. |
Quitte object containing the reference data set
Jan Philipp Dietrich
iamReferenceData()iamReferenceData()
Creates a PDF summarizing check results and adding additional results for a comparison to reference data (e.g. matching to historical data)
iamSummaryPDF( input, check_results = NULL, file = "summary.pdf", maxLinesOutput = 200, pdfStyle = NULL, ... )iamSummaryPDF( input, check_results = NULL, file = "summary.pdf", maxLinesOutput = 200, pdfStyle = NULL, ... )
input |
named list with elements available for check functions |
check_results |
list with check results as returned by |
file |
File name the summary should be written to or a Sweave object. If a sweave object is provided the function will return the updated object, otherwise it will write its content to the file. |
maxLinesOutput |
maximum number of lines that should be output in the pdf |
pdfStyle |
list of style-options for the pdf |
... |
additional arguments sent to |
Jan Philipp Dietrich
## Not run: input <- list(x=example_REMIND, ref=iamReferenceData()) check_results <- iamCheck(example_REMIND) iamSummaryPDF(check_results$input, check_results$out) ## End(Not run)## Not run: input <- list(x=example_REMIND, ref=iamReferenceData()) check_results <- iamCheck(example_REMIND) iamSummaryPDF(check_results$input, check_results$out) ## End(Not run)
Checks for duplicate entries in the data. To optimize performance the single data values are encoded into integer values (corresponding to their factor representatives). The encoded integer values are then compared if duplicates exist.
preCheckDuplicates(x)preCheckDuplicates(x)
x |
Input data that should be checked, provided as a quitte object |
Two-dimensional list. 1) message with how many dpulicates 2) list of duplicates (in their integer representation)
Cornelia Auer
Processes a check, which means that it returns a note if the check fails, returns a check summary based on the chosen verbosity and returns the check results in a structured way. All checks should be run within processCheck. Requirement for a check function is that it returns a list with 2 elements: "message" which contains the standard message that should show up for the test and "failed" which is a vector of names for which the corresponding test failed.
processCheck(check, input)processCheck(check, input)
check |
Function call as character that should be run |
input |
Named list with elements available for check functions |
List containing check results
Jan Philipp Dietrich
iamProjectConfig, as.quitte, is.quitte
out <- processCheck(check = "checkMin(x, cfg)", input = list(x=example_REMIND, cfg=iamProjectConfig()))out <- processCheck(check = "checkMin(x, cfg)", input = list(x=example_REMIND, cfg=iamProjectConfig()))
Rename and aggregate data using a mapping
Reads in a substitutes names of variables according to the mapping, multiplies reported values by an optional factor in a column named "factor" of the mapping, and saves the output in a new *.mif
RenameAndAggregate(data, mapping, missing_log = NULL)RenameAndAggregate(data, mapping, missing_log = NULL)
data |
Lists with list of magpie-objects (a magpie-object as created by read.report), first list containts scenarios, second list the models |
mapping |
mapping of the variable names of the read-in mif. The header is used for naming. The format of the mapping should be: 1st column the standard naming in PIK mif format. X further columns that contain the indicator names in the reporting format. Can also contain several indicator columns (e.g Variable and Item). Optional columns with reserved names are unit, weight, factor, and spatial. Factor is a number that the results will be multiplied with (e.g. to transform CO2 into C). Weight is needed if several mif indicators shall be aggregated to one reporting indicator. You always need a weight column if you have multiple mif to one reporting mappings. If you have a weight column, you have to have values in it for all indicators. If NULL, the results are added up; if you provide an indicator name (of a mif indicator), this indicator will be used for the weighting of a weighted mean. Spatial should be set to "glo" for mif indicators that shall only be reported globally and "reg" for only reporting locally. The default is "reg+glo", which implies reporting on global and local level. In the case of aggregation, contradicting entries in spatial column for the same reporting indicator will throw an error. Unit is a name of the unit without () Example: "magpie";"agmip";"item";"unit";"weight";"factor" "Nutrition|+|Calorie Supply (kcal/capita/day)";"CALO";"AGR";"kcal/capita/day";"NULL";1 |
missing_log |
name of logfile to record variables which are present in the mapping but missing in the mif file. By default, no logfile is produced |
Christoph Bertram, Lavinia Baumstark, Anastasis Giannousakis, Florian Humpenoeder, Falk Benke, Benjamin Leon Bodirsky
## Not run: RenameAndAggregate(list(model=list(scenario=population_magpie)),"Mapping_generic_ADVANCE.csv") ## End(Not run)## Not run: RenameAndAggregate(list(model=list(scenario=population_magpie)),"Mapping_generic_ADVANCE.csv") ## End(Not run)
Reads in a reporting.mif or uses a magpie object based on a read-in reporting.mif, substitutes names of variables according to the mapping, multiplies reported values by an optional factor in a column named "factor" of the mapping, and saves the output in a new *.mif
write.reportProject( mif, mapping, file = NULL, max_file_size = NULL, format = "default", append = FALSE, missing_log = NULL, ... )write.reportProject( mif, mapping, file = NULL, max_file_size = NULL, format = "default", append = FALSE, missing_log = NULL, ... )
mif |
Lists with magpie-objects or a magpie-object as created by read.report or a path to a report.mif |
mapping |
mapping of the variable names of the read-in mif. The header is used for naming. The format of the mapping should be: 1st column the standard naming in PIK mif format. X further columns that contain the indicator names in the reporting format. Can also contain several indicator columns (e.g Variable and Item). Optional columns with reserved names are unit, weight, factor, and spatial. Factor is a number that the results will be multiplied with (e.g. to transform CO2 into C). Weight is needed if several mif indicators shall be aggregated to one reporting indicator. You always need a weight column if you have multiple mif to one reporting mappings. If you have a weight column, you have to have values in it for all indicators. If NULL, the results are added up; if you provide an indicator name (of a mif indicator), this indicator will be used for the weighting of a weighted mean. Spatial should be set to "glo" for mif indicators that shall only be reported globally and "reg" for only reporting locally. The default is "reg+glo", which implies reporting on global and local level. In the case of aggregation, contradicting entries in spatial column for the same reporting indicator will throw an error. Unit is a name of the unit without () Example: "magpie";"agmip";"item";"unit";"weight";"factor" "Nutrition|+|Calorie Supply (kcal/capita/day)";"CALO";"AGR";"kcal/capita/day";"NULL";1 |
file |
name of the output file, default=NULL returns the output object |
max_file_size |
maximum file size in MB; if size of file exceeds max_file_size reporting is split into multiple files |
format |
available reporting formats: "default", "IAMC" and "AgMIP". "default" and "IAMC" are very similar (wide format for year) and differ only in the use of semi-colon (default) and comma (IAMC) as seperator. "AgMIP" is in long format. |
append |
Logical which decides whether data should be added to an existing file or an existing file should be overwritten |
missing_log |
name of logfile to record variables which are present in the mapping but missing in the mif file. By default, no logfile is produced |
... |
arguments passed to write.report and write.report2 |
Christoph Bertram, Lavinia Baumstark, Anastasis Giannousakis, Florian Humpenoeder, Falk Benke, Benjamin Leon Bodirsky
write.report, RenameAndAggregate
## Not run: write.reportProject("REMIND_generic_test.mif","Mapping_generic_ADVANCE.csv") ## End(Not run)## Not run: write.reportProject("REMIND_generic_test.mif","Mapping_generic_ADVANCE.csv") ## End(Not run)