This mode calculates performance statistics from a table of predictions generated by Predict Mode.
It is implemented by org.bdval.StatsMode.java.
Mode Parameters
The following options are available in this mode
| Flag | Arguments | Required | Description |
|---|---|---|---|
| result-directory1 result-directory2 … result-directoryN | n/a | no | Directory that contains results written by the bdval ant scripts. |
(-c|--model-conditions) |
model-conditions | no | The model condition file where models which are be processed are described. |
--survival |
survival | no | Survival filename. This file contains survival data in tab delimited format; 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 numerical covariates that will be included in the regression model |
--aggregation-method |
aggregation-method | no | Type of aggregation method. Predictions can be aggregated by repeat of cross-validation (per-repeat, default value of this option), or by test set (per-test-set). (default: per-test-set) |
--submission-file |
submission-file | no | The 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-measures |
other-measures | no | A 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: ) |
--label |
label | yes | A string that the type of model construction process used to generate the models. (default: auto) |
--binary |
binary | no | 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) |

Leave a Comment