plotStatCurves: Make plots for MCC, ROC, and Prec-Rec

In PriceLab/FPML: Footprint Machine Learning

Description

Given the stats dataframes for boosted and linear models, plus a flag denoting the data type, plot the 3 relevant curves: the Matthews Correlation Coefficient curve, the ROC curve, and the Precision-Recall curve. Also requires specification of file paths to store the .PNG files for these curves

Usage

plotStatCurves(boostedStats, linearStats, dataType, filePaths)

Arguments

boostedStats	The dataframe containing statistics from the boosted model. Generally, these are generated using the buildBoostedModel function
dataType	A string that must be one of c("seed16","seed20","both","motifsOnly"). This string denotes what labels will be given for the plots within the figures themselves
filePaths	A character string of length = 3, giving the 3 filepaths to which the function will save the curves. They should be in the following format: c(MCC_path, ROC_path, PrecRec_path)
linearStatList	The list containing 2 dataframes, one with the statistics from the full linear model, the other with statistics from the individual linear models. Generally, these are generated using the buildLinearModels function

Value

3 .PNG files, each containing one of the 3 figures for the given data.

PriceLab/FPML documentation built on May 28, 2019, 2:25 p.m.