- SciPy 0.12.0 Release Notes
- New features
- scipy.spatial improvements
- Spectral estimators (scipy.signal)
- scipy.optimize improvements
- scipy.special improvements
- scipy.sparse.linalg features
- Listing Matlab(R) file contents in scipy.io
- Documented BLAS and LAPACK low-level interfaces (scipy.linalg)
- Polynomial interpolation improvements (scipy.interpolate)
- Deprecated features
- Backwards incompatible changes
- New features
SciPy 0.12.0 is the culmination of 7 months of hard work. It contains many new features, numerous bug-fixes, improved test coverage and better documentation. There have been a number of deprecations and API changes in this release, which are documented below. All users are encouraged to upgrade to this release, as there are a large number of bug-fixes and optimizations. Moreover, our development attention will now shift to bug-fix releases on the 0.12.x branch, and on adding new features on the master branch.
Some of the highlights of this release are:
- Completed QHull wrappers in scipy.spatial.
- cKDTree now a drop-in replacement for KDTree.
- A new global optimizer, basinhopping.
- Support for Python 2 and Python 3 from the same code base (no more 2to3).
This release requires Python 2.6, 2.7 or 3.1-3.3 and NumPy 1.5.1 or greater. Support for Python 2.4 and 2.5 has been dropped as of this release.
Cython version of KDTree, cKDTree, is now feature-complete. Most operations (construction, query, query_ball_point, query_pairs, count_neighbors and sparse_distance_matrix) are between 200 and 1000 times faster in cKDTree than in KDTree. With very minor caveats, cKDTree has exactly the same interface as KDTree, and can be used as a drop-in replacement.
scipy.spatial now contains functionality for computing Voronoi diagrams and convex hulls using the Qhull library. (Delaunay triangulation was available since Scipy 0.9.0.)
The functions scipy.signal.periodogram and scipy.signal.welch were added, providing DFT-based spectral estimators.
A callback mechanism was added to L-BFGS-B and TNC minimization solvers.
A new global optimization algorithm. Basinhopping is designed to efficiently find the global minimum of a smooth function.
The computation of special functions related to the error function now uses a new Faddeeva library from MIT which increases their numerical precision. The scaled and imaginary error functions erfcx and erfi were also added, and the Dawson integral dawsn can now be evaluated for a complex argument.
- In scipy.sparse.linalg.spsolve, the b argument can now be either a vector or a matrix.
- scipy.sparse.linalg.inv was added. This uses spsolve to compute a sparse matrix inverse.
- scipy.sparse.linalg.expm was added. This computes the exponential of a sparse matrix using a similar algorithm to the existing dense array implementation in scipy.linalg.expm.
A new function whosmat is available in scipy.io for inspecting contents of MAT files without reading them to memory.
The barycentric, Krogh, piecewise and pchip polynomial interpolators in scipy.interpolate accept now an axis argument.
The module scipy.lib.lapack is deprecated. You can use scipy.linalg.lapack instead. The module scipy.lib.blas was deprecated earlier in Scipy 0.10.0.
The function scipy.io.save_as_module was deprecated in Scipy 0.11.0, and is now removed.
Its private support modules scipy.io.dumbdbm_patched and scipy.io.dumb_shelve are also removed.
The function scipy.stats.scoreatpercentile has been given an axis argument. The default argument is axis=None, which means the calculation is done on the flattened array. Before this change, scoreatpercentile would act as if axis=0 had been given. Code using scoreatpercentile with a multidimensional array will need to add axis=0 to the function call to preserve the old behavior. (This API change was not noticed until long after the release of 0.12.0.)