Experimental Data Manipulation
This functions manipulate the experiment objects, implementing common data manipulation operations, such as subseting and joining. In addition, there are a number of available functions to sanitize and preprecessing the experimental data such as looking for duplicated entries, reducing parameters and instantiating the methods with the parameters.
Load data and create an exreport experiment
Create an exreport experiment from a tabular representation
Combine two experiments with different outputs
Concatenate rows of matching experiments
Extend an experiment by adding new parameters
Extract statistically equivalent methods from a multiple comparison test
Create a new experiment with only the duplicated rows
Instatiate the methods in the experiment for each one of the different parameter configurations.
Reduce a parameter by a function for each method, problem and remaining parameter configuration interaction
Remove duplicated rows from an experiment
Change the name of elements that an experiment contains
Change the order of elements that an experiment contains
Obtains a subset of an experiment matching the given conditions
This functions implement several statistical test to compare the methods of the experiment.
Paired Wilcoxon statistical test
Multiple Comparison Statistical Test (Friedman + Control Holm PostHoc)
Multiple Comparison Statistical Test (Friedman + Pairwise Shaffer PostHoc)
Tabular Data Generation
This functions generate tables summarizing the information of an experiment or a text to be printed to pdf or web reports.
Summarize the experiment with a table for given outputs
Display pairwise information about a multiple test between the methods
Summarize the result of a multiple comparison statistical test in a table
This functions generate plots summarizing the information of an experiment or a text to be printed to pdf or web reports.
Area plot for the rank distribution from a multiple test
Barplot for summarizing an experiment output variable
Boxplot for the ranks distribution and control hypotheses from multiple test
These are the main functions used to generate and render the reports
Create a new exreport document
Add elements to an existing exreport document
Render an exreport document
Example problems for the examples and documentations
Problem: Comparison between several Machine Learning algorithms from the Weka library