This mode is used to make predictions for unlabeled samples. It can be used to evaluate predictions on an independent test set. True labels can be provided for the  purpose of evaluation in the form of a cids file.

It is implemented by org.bdval.Predict.java.

Mode Parameters

The following options are available in this mode

Flag Arguments Required Description
--model modelyesModel filename prefix. Models have several files named with a common prefix. The model that will be used to predict the label of the samples in the input file.
--test-samplestest-samplesnoSurvival filename.Filename for list of test sample ids. Path to a file with one line per test sample id. The input dataset will be filtered to keep only those samples in the list for prediction and performance calculation. will be included in the regression model
--print-statsn/anoPrint statistics instead of detailed result table.
--estimate-with-replacement estimate-with-replacementnoEstimate performance measure as an average over a number of test set samples constructed by sampling the fixed test set with replacement. A thousand samplings are considered. This makes it possible to estimate standard deviation of each measure on the test set and acknowledges that the test set is just another sample of a very large population. Default is false. (default: false)
--true-labelstrue-labelsnoTrue labels for this dataset, in the cids format. Providing true labels makes it possible to report evaluation measures on the test set.
--survivalsurvivalnoSurvival filename. This file contains survival data in tab delimited table; column 1: chipID has to match cids and tmm, column 2: time to event, column 3 censor with 1 as event 0 as censor, column 4 and beyond are all numerical covariates that will be included in the regression model
--submission-filesubmission-filenoThe MAQC-II submission file to create. Please note that this file lacks some columns required for MAQCII submission. These columns must be created manually in excel. (default: -)
--other-measuresother-measuresnoA list of performance measures to evaluate and report in the maqcii file. These measures will be appended at the end of the columns, after the official maqcii submission columns. (default: )
--labellabelyesA string that the type of model construction process used to generate the models. (default: auto)
--binarybinaryno

Indicates that binary decision values (-1/+1) should be used to evaluate the binary flavor of evaluation measures in addition to the traditional evaluation measures. (default: false)