Array Statistics Project Definition

Restrictions, Exclusions, Limitations

People Affected

Person	Initials	Notes
Ryan Cabeen	RC
Craig Schwartz	CS
Roger Woods	RW

Version

1.0 draft

Description

This program will read a collection of array data and calculate the mean and variance of the collection.

Input formats

The data samples may be in any of the following formats. A default format is assumed. If this is incorrect the input file format must be specified by the user via the appropriate command line argument.

• an ASCII file with space-delimited and line-delimited floating point values.
• a LONI FPIC UIF file.

Input Modes

The input data filenames may be specified in the following ways. A default input mode is assumed.

• the filenames as command line arguments
• the filename of an ascii list of input data filenames

Output Formats

The output file formats will be the same as the input file format.

Requirements

Requirements Change Procedure

• Changes to be mutually agreed upon by Craig and Ryan.
• Changes and their date will be entered in a 'requirements change' list.

Requirements Changes

Command Line Arguments

Note: this list is provisional

the set of command line arguments will be

• -fileList filePath : a text file containing list of paths to the input files. If not present, files will be read from the command line.
• -claList : the input filenames will be parsed from the command line arguments
• -uif : the input files will be read as UIFs
• -ascii : the input files will be read as ASCII arrays
• -var fileName : the name of the output file to hold the calculated sample variance (optional).
• -mean fileName : the name of the output file to hold the calculated sample mean (optional).
• -verbose : enables runtime status messages.
• -usage : prints a summary of how to run this program.
• -version : prints the program version number.
• -license : prints license status, if appropriate.
• -delimiter char - sets the sample value delimiter.

The output file format will match the input file format.

Run-time errors

• if the input file formats do not agree
• if the input file dimensions disagree
• if no output statistic is selected.

Products

This project will result in the following products:

• A Java application jar that will create a file storing the Jacobian determinant.
• A Loni Pipeline descriptor definition.
• A CVS repository containing the Java source code of this application.
• Examples illustrating use of the application.
• Documentation describing the algorithms and limitations of the calculations.

Lessons Learned