final touches for release

contrasts.m

@@ -14,7 +14,10 @@
 % contrasts are not orthonormalized!
 %
 %
-% Copyright (C) 2013 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 % number of factors
@@ -61,4 +64,3 @@
 % it will be useful, but without any warranty; without even the implied
 % warranty of merchantability or fitness for a particular purpose. See the
 % GNU General Public License <http://www.gnu.org/licenses/> for more details.
-

cvManovaCore.m

@@ -33,6 +33,9 @@
 % See also cvManovaSearchlight, cvManovaRegion
 %
 %
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
 % Copyright (C) 2016 Carsten Allefeld
 
 

cvManovaSearchlight.m

@@ -6,20 +6,23 @@ function cvManovaSearchlight(dirName, slRadius, Cs, permute, lambda)
 %
 % dirName:   directory where the SPM.mat file referring to an estimated
 %            model is located
-% slRadius:  radius of the searchlight sphere, in voxels
+% slRadius:  radius of the searchlight sphere in voxels
 % Cs:        cell array of contrast vectors or matrices
 % permute:   whether to compute permutation values
 % lambda:    regularization parameter (0–1)
 %
 % Output files are written to the same directory:
-% spmD_C####_P####.nii:  images of the pattern discriminability D
-%                        contrast and permutation are identified by numbers
-% spmD_C####_P####.nii:  images of standardized pattern discriminability D_s
-% VPSL.nii:              image of the number of voxels for each searchlight
-% cmsParameters.mat:     record of the analysis parameters
+% spmD_C####_P####.nii:   images of the pattern discriminability D
+%                         contrast and permutation are identified by numbers
+% spmDs_C####_P####.nii:  images of standardized pattern discriminability D_s
+% VPSL.nii:               image of the number of voxels for each searchlight
+% cmsParameters.mat:      record of the analysis parameters
 %
 %
-% Copyright (C) 2013-2016 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 fprintf('\n\ncvManovaSearchlight\n\n')

fletcher16.m

@@ -7,11 +7,14 @@
 % data:     data to be checksummed,
 %           array of integers in the range 0 to 255 (0xFF)
 % checksum: Fletcher-16 checksum,
-%           integer where each of the bytes is in the range 0 to 254 (0xFE)
+%           16-bit integer where each of the bytes is in the range 0 to 254 (0xFE)
 %
 % https://en.wikipedia.org/wiki/Fletcher%27s_checksum
 %
 %
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
 % Copyright (C) 2016 Carsten Allefeld
 
 

inestimability.m

@@ -1,6 +1,6 @@
 function ie = inestimability(C, X)
 
-% degree of inestimability of a contrast wrt a design matrix
+% degree of inestimability of a contrast w.r.t. a design matrix
 %
 % ie = inestimability(C, X)
 %
@@ -12,7 +12,10 @@
 % part. For a completely inestimable contrast, ie = 1.
 %
 %
-% Copyright (C) 2013 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 % maximum "0" observed so far: 4.79*eps
@@ -46,4 +49,3 @@
 % it will be useful, but without any warranty; without even the implied
 % warranty of merchantability or fitness for a particular purpose. See the
 % GNU General Public License <http://www.gnu.org/licenses/> for more details.
-

loadDataSPM.m

@@ -7,20 +7,25 @@
 % dirName:  name of directory that contains SPM.mat
 % regions:  optional additional region mask(s),
 %           (cell array of) logical 3D volume(s) or filename(s)
-% Ys:       MR data (within mask) for each session, scans x voxels
-% Xs:       design matrix for each session, scans x regressors
+% Ys:       MR data (within mask), cell array with one element for each
+%           session, containing an array of size scans × voxels
+% Xs:       design matrix for each session, cell array with one element
+%           for each session, containing an array of size scans × regressors
 % mask:     analysis brain mask, logical 3D volume;
-%           possibly combined with region mask
+%           possibly combined with union of region masks
 % misc:     struct with additional data:
 %     mat   voxels to mm transformation matrix
-%     fE    residual degrees of freedom per session
+%     fE    residual degrees of freedom for each session
 %     rmvi  cell array of mask voxel indices for each region
 %
 % Y & X and are high-pass filtered and whitened.
 % Y includes only those voxels selected through mask.
 %
 %
-% Copyright (C) 2013-2016 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 % default argument values

runSearchlight.m

@@ -6,11 +6,11 @@
 % res = runSearchlight(checkpoint, slRadius, mask, fun, ...)
 %
 % checkpoint:   name of checkpoint file ([] to disable checkpointing)
-% slRadius:     radius of searchlight, in voxels
+% slRadius:     radius of searchlight in voxels
 % mask:         3-dimensional logical array indicating which voxels to use
 % fun:          function to call with voxel indices within window
 % ...:          additional arguments are passed through to the function
-% res:          results, 2-dimensional matrix, voxels x output values
+% res:          results, array of size voxels × output values
 % p:            number of voxels in each searchlight, column vector
 %
 % The function has to be of the form r = fun(mvi, ...)
@@ -21,17 +21,20 @@
 %
 % A voxel is included in the searchlight if its distance from the center is
 % *smaller than or equal to* the radius. Note that fractional values are
-% possible. Run slSizes for a table of meaningful values and resulting
+% possible. Run slSize for a table of meaningful values and resulting
 % searchlight sizes
 %
 % Intermediate results are saved at regular intervals to the checkpoint
 % file, if given. On a subsequent run, if the checkpoint file exists, its
 % contents are loaded and the computation continues from that point. 
 %
-% See also slSizes
+% See also slSize
 %
 %
-% Copyright (C) 2013-2016 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 % normalize checkpoint file name, preserving emptiness

signPermutations.m

@@ -14,7 +14,10 @@
 % permutation, ones(n, 1).
 %
 %
-% Copyright (C) 2013-2016 Carsten Allefeld
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
+% Copyright (C) 2013–2016 Carsten Allefeld
 
 
 if nargin < 2
@@ -42,4 +45,3 @@
 % it will be useful, but without any warranty; without even the implied
 % warranty of merchantability or fitness for a particular purpose. See the
 % GNU General Public License <http://www.gnu.org/licenses/> for more details.
-

spmReadVolMatched.m

@@ -26,9 +26,12 @@
 % consistent. Only the fields mat (4 × 4 matrix) and dim (1 × 3 vector)
 % have to be present.
 %
-% See also spmCoords, spm_write_sn, spm_slice_vol
+% See also spm_write_sn, spm_slice_vol
 %
 %
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
 % Copyright (C) 2016 Carsten Allefeld
 % adapted from spm_read_vols and John Ashburner's reslice.m, v1.42
 % see https://www2.warwick.ac.uk/fac/sci/statistics/staff/academic-research/nichols/scripts/spm/johnsgems/#Gem8

spmReadVolsMasked.m

@@ -14,9 +14,12 @@
 % memory spikes and to gain speed. This is achieved by directly accessing
 % memory-mapped data via the volume structs' file_array object V.private.raw. 
 %
-% See also spmUnMask, spmReadMatched, spmCoords
+% See also spmUnMask, spmReadVolMatched, spmCoords
 %
 %
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
 % Copyright (C) 2016 Carsten Allefeld
 
 % This function is about 2–3 times faster than spm_read_vols on a local

spmWriteImage.m

@@ -38,19 +38,22 @@
 % For background information, see
 %   http://nifti.nimh.nih.gov/nifti-1/documentation/nifti1fields/
 %
-% It is however possible to store floating-point data that only attain
-% discrete and linearly-spaced values (e.g. ratios) using an integer data
-% type using the linear transformation feature of the NIfTI format,
+% It is possible to store floating-point data that only attain discrete and
+% linearly-spaced values (e.g. ratios) using an integer data type using the
+% linear transformation feature of the NIfTI format,
 %   spmWriteImage(int, fname, mat, 'scl_slope', slope, 'scl_inter', inter)
 % where int is of class uint8, uint16, or uint32. Upon reading the file,
 % the integer data are automatically transformed according to
 %   Y = double(int) * slope + inter;
 % This way the image file may be kept substantially smaller. Note however
 % that 'scl_slope' and 'scl_inter' are stored only with single precision.
 %
-% See also spmCoords, nifti/Contents
+% See also nifti/Contents
 %
 %
+% This file is part of v3 of cvmanova, see
+% https://github.com/allefeld/cvmanova/releases
+%
 % Copyright (C) 2016 Carsten Allefeld
 
 
@@ -139,4 +142,3 @@
 % it will be useful, but without any warranty; without even the implied
 % warranty of merchantability or fitness for a particular purpose. See the
 % GNU General Public License <http://www.gnu.org/licenses/> for more details.
-