[pymvpa] Fresh release (2.4.2) up for grabbers
Yaroslav Halchenko
debian at onerussian.com
Tue Mar 8 16:58:41 UTC 2016
Dear PyMVPA users,
We have released 2.4.2! Thanks go to everyone who contributed in code or
discussions and help on the mailing list and/or github issues.
More detailed changelog is below -- this release carries quite a few of
bugfixes (reverse mapping, OpenfMRI datasets compatibility, etc),
refactorings and some new features which might be of interest for you (fast
m1nn searchlight, SplitRFE on static measures etc)
One of interesting experimental features (actually was already in 2.4.1)
is "integration" with DueCredit (http://duecredit.org), so please try
installing duecredit (apt-get install python-duecredit on debian based systems,
pip install duecredit otherwise) and running your analysis with
DUECREDIT_ENABLE=1 environment variable and see what you would get? ;)
Release was uploaded to Debian proper sid, and all backports for NeuroDebian.
It will eventually come to pypi and other distributions. pymvpa.org
website eventually will be updated ;)
Enjoy and happy 8th of March to those who are celebrating ;)
* 2.4.2 (Tue, 8 Mar 2016)
* Fixes
- *Important:*
Reverse mapping of some chained Flatten/StaticSelection mappers did not
work correctly e.g. if you selected some features from already masked
fmri_dataset. This could have resulted in incorrect cluster counts by
:class:`~mvpa2.algorithms.group_clusterthr.GroupClusterThreshold`. Please
recreate your datasets and re-estimate GroupClusterThreshold if that was
the case for you
- ad-hoc searchlights (gnb, m1nn) can now operate with partitioners which
leave some samples out of training and testing sets. Also `splitter` argument
was added to them for greater flexibility
- Due to the bug in OpenfMRI datasets' TR within NIfTIs being hardcoded to
wrong 1.0, so `scan_key.txt` will now be consulted if TR is 1 in the .nii*
- Compatibility with :mod:`~numpy` 1.10 fixes
- :class:`CachedQueryEngine` acquired .ids making it compatible with some
ad-hoc searchlights
- `FeatureSelection` acquired `__iadd__` fixing the incorrect behavior upon
reverse after a sequence of feature selections
* Deprecations/removal
- `Hamster` is gone.
* Enhancements
- Bundled version of libsvm updated to 3.12. Now includes maxiter
setting which prevents infinite looping which can happen in some rare cases
- A swarm of stylistic improvements ("is not", PEP8, etc) which should not
affect functionality but could result in more robust operation
- `CrossValidation` can now operate with a None generator (i.e. partitioner) using
solely `Splitter` to generate a single split on original dataset. Provides easier
means for "cross-classification"
- :class:`~mvpa2.measures.nnsearchlight.M1NNSearchlight` can now do classification
based on correlation distance (just provide `dfx=one_minus_correlation` to kNN)
- libsvm bindings for SVM were refactored to interface via svmc not _svmc interface,
which made them also compatible with swig 3.x
- :meth:`~mvpa2.base.dataset.AttrDataset.to_npz` and :meth:`~mvpa2.base.dataset.AttrDataset.from_npz`
to interface Datasets through NumPy's npz files
- Variety of PEP8 and other tune ups for more readable code
- :class:`~mvpa.featsel.rfe.SplitRFE` can now work with static measures (e.g.
`OneWayAnova`) and `BinaryFxFeaturewiseMeasure`. So do feature selection
with nested cross-validation without double-dipping!
* New functionality
- :class:`~mvpa2.generators.partition.FactorialPartitioner` for factorial designs
to cross-validate across sub-ordinate category samples (more efficient/avoids
previously recommend ChainMapper of NFoldPartitioner and Sifter)
--
Yaroslav O. Halchenko
Center for Open Neuroscience http://centerforopenneuroscience.org
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834 Fax: +1 (603) 646-1419
WWW: http://www.linkedin.com/in/yarik
More information about the Pkg-ExpPsy-PyMVPA
mailing list