1.4.0rc4
Pre-release
Pre-release
First release in the 1.4.x series. This release includes enhancements and bug-fixes towards the release of the first beta version of *dMRIPrep*. It also contains new features that are necessary for the API overhaul that has happened within the new *SDCFlows 2.x* series. The new series include a cross-cutting (modalities, species) workflow to generate EPI references. Finally other *NiPreps* will also have a first release with specific support for them: *NiRodents* and *NiBabies* (and their corresponding *fMRIPrep* extensions). A list of prominent changes follows: * FIX: Address issues with ``RobustAverage`` global signal measurement (#607) * FIX: ``NonsteadyStatesDetector`` wrongly using Nipype's ``is_outlier`` (#605) * FIX: Change in *svgutils*' API on 0.3.2 breaks reportlets (#599) * FIX: Check for ``in_segm`` input, not ``seg_file`` (typo) (#592) * FIX: Use the mask to calculate FOV rather than the fixed image in ``GenerateSamplingReference`` (#583) * FIX: Allow omission of ``<res>`` for template normalization (#582) * FIX: Include ``_T2starw`` ``_MTw`` and ``_TSE``-suffixes in ``build_path`` options (#584) * FIX: ``DerivativesDataSink`` warning when it has multiple source files (#573) * ENH: Add an inversion operation to ``IntensityClip`` (#616) * ENH: Cross-cutting (modalities, species) workflow to generate EPI references (#610) * ENH: Add a ``RobustAverage`` interface and split volume selection (#602) * ENH: Revise some patterns of the ``BIDSLayout`` config to aid *SDCFlows* new API (#585) * ENH: Upstream *fMRIPrep*'s ``init_bbreg_wf`` to integrate it in *dMRIPrep* (#586) * MAINT: Add DS030 dataset, with clipped (55 timepoints) BOLD data (#609) * MAINT: Migrate ``ds003_downsampled`` to ``nipreps-data`` (#608) * MAINT: Move mask-regressions test-data to datalad + nipreps-data (#606) * MAINT: Refactor structure of interfaces (#603) * MAINT: Drop Python 3.6, test setuptools builds, pip installations, and revise Docker pinned versions (#593) * MAINT: CircleCI housekeeping (#580)