Load data and create an exreport experiment
expCreate(data, methods = "method", problems = "problem", parameters = c(), respectOrder = FALSE, name, tol = 1e-09)
- A data.frame object satisfying the experiment format
- The name of the variable which contains the methods, by default is searches for a column named "method".
- The name of the variable which contains the problems, by default is searches for a column named "problem".
- A list of the columns names to be identified as parameters. By default the remaining categorical columns are identified as parameters, so this list is useful only to identify numeric columns.
- A logical parameter which indicates if the order of the elements of the method and problem columns must be respected by appearance or ordered alphabeticaly. It affects to the look of data representations.
- A string which will identify the experiment in the report.
- Tolerance factor to identify repeated experiments for duplicated rows.
A new exreport experiment object.
This function loads a data.frame, checks its properties and formats an exreport experiment object. The columns of an experiments must contain at least two categorical columns to be identified as the method and problem variables and a thrid numerical column to be identified as an output variable. Additional columns can be added as parameters or additional outputs.
# Creates experiment specifying column names and the numerical variables that # are parameters expCreate(wekaExperiment, methods="method", problems="problem", parameters="fold", name="Test Experiment")#Experiment name: Test Experiment #method: J48, NaiveBayes, OneR, RandomForest #problem: anneal, audiology, balance-scale, car, glass, horsecolic, hypothyroid, ionosphere, liver-disorders, lymph, primary-tumor, soybean, vehicle, vote, vowel #parameters: *) featureSelection [no,yes] *) fold [0,1,2,3,4,5,6,7,8,9] #outputs: accuracy, trainingTime