pymvpa21

Usage

Overview

This is the command line interface of the PyMVPA framework. Specific functionality is provided through a collection of sub-commands. Available sub-commands are:

info: Query various information about a PyMVPA installation.

mkds: Create a PyMVPA dataset from various sources.

mkevds: Extract (multi-sample) events from a dataset

describe: Describe a dataset's content

dump: Dump dataset components in various formats

preproc: Preprocess a PyMVPA dataset.

crossval: Cross-validation of a learner's performance

searchlight: Traveling ROI analysis

select: Select a subset of samples and/or features from a dataset

atlaslabeler: Query atlases for anatomical labels of voxel coordinates, or their

statistics

exec: Execute arbitrary Python expressions (on datasets)

(scatter): not available in this installation

plotmotionqc: Generate a QC plot for (BOLD fMRI) motion estimates of multiple segments.

ttest: t-test across multiple volumes against some chance level

Detailed usage information for individual commands is available via command-specific help options, i.e.: bin/pymvpa2 <command> --help

Options

-h, --help, --help-np

show this help message and exit. --help-np forcefully disables the use of a pager for displaying the help.

--version

show program's version and license information and exit

--dbg

invoke a debugger when a crash occurs

--dbg-channel COMMON_DEBUG

enable debug channel (see 'info' command for available channels)

--preload PRELOAD

filename of a custom Python script that is executed prior to, and in the same session as the actual command. This can be used to modify the execution environment, for example creating a custom classifier instance. This option can be given multiple times, and the associated scripts are ran in order of their appearance on the command line.

--verbose [COMMON_VERBOSE]

output verbosity level