SnPM 13.1.09

• fix: Typo fix in design reporting.
• enh: Added function snpm_update to allow users to check for new versions of the code (#52, #75).
• enh: Added CITATION.cff file (#65).
• fix: Broken links in README.md fixed (#74).
• fix: Updated documentation for updating SnPM in README.md (#73).

SnPM 13.1.08

• fix: fix error where large ANOVA-within-subject would fail with "SnPM:InvalidPiCond" (#61, cmaumet#1)
• fix: dimensions when 'low mem' cluster inference done in 2D (#69)
• enh: Table formatting (#51)
• enh: Updated cluster forming threshold default to 2.3 (#59)
• enh: allow for covariates in ANOVA within- and between-subject (#61, #60)
• enh: Adding Zenodo's badge to the README (with doi) (#62)
• dev: Add template for pull requests (#72)

SnPM 13.1.07

• fix: Add GPL licence. cf. #29
• fix: decrease lower bound on number of subject for approximate perms. cf #7
• enh: Table: Create TabDat structure expected by spm_list, allowing access to all peak/cluster results, even if there are too many for the SnPM results page. The TabDat structure is (inelegantly) saved to a mat file. cf. #30 for more details.
• enh: Apply mlint suggestions for better compatibility with MATLAB Compiler and Octave. cf #47.
• test: use latest available ground truth if perfect match for matlab version is not available. cf #41.

SnPM 13.1.06

• fix: ANOVA between group "all zero"/"all equal" options were flipped in the menu of the matlabbatch.
• fix: snpm_cp can now handle XYZ locations in a 1D space.
• Version 13.1.06 (snpm)

SnPM 13.1.05

• fix: two-sample t-test with nscans>12 was errored due to call to undefined variable nPerm (bug introduced in previous release: 13.1.4)
• fix: warning SnPM_ST.mat is too big was never raised.
• Add seed shuffling for paired two-sample test modules.

SnPM 13.1.04

• Fix: implicit masking for non-float datatypes (cf. #2)
• Fix: Do not assume statistic is T when computing cluster forming threshold from P-value (cf. #14)
• Update syntax to set the seed of the random number generator (rng)
• New SnPM global parameter shuffle_seed. If true, the rng uses a 'shuffle' seed providing different results for each run. If false Matlab's 'default' seed or any seed defined by the user will be used (false will be useful for HCP computing, replication, testing (cf. #3)
• Allows MC permutation (with reptitions) for large enough nPerms (cf. #12)
• Do not display figures in command-line mode (cf. #6)
• Add identifiers to errors and warnings (fixes #13)

SnPM13.1.03

Updates to paired T test plugin snpm_pi_PairT.m, allowing it to work with more than 52 subjects.