<br><br><div class="gmail_quote">On Sun, Nov 13, 2011 at 12:19 PM, Ralf Gommers <span dir="ltr"><<a href="mailto:ralf.gommers@googlemail.com">ralf.gommers@googlemail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px">Hi all,<br><br>I am pleased to announce the availability of <span><span><span style="background-color:rgb(251, 238, 153);color:rgb(34, 34, 34);background-repeat:initial initial">SciPy</span></span></span> <span><span style="background-color:rgb(251, 238, 153);color:rgb(34, 34, 34);background-repeat:initial initial">0.10</span></span>.<span style="background-color:rgb(251, 238, 153);color:rgb(34, 34, 34);background-repeat:initial initial">0</span>. For this release over a 100 tickets and pull requests have been closed, and many new features have been added. Some of the highlights are:<br>
<br> - support for Bento as a build system for <span><span style="background-color:rgb(251, 238, 153);color:rgb(34, 34, 34);background-repeat:initial initial">scipy</span></span><br>
- generalized and shift-invert eigenvalue problems in sparse.linalg<br> - addition of discrete-time linear systems in the signal module<br><br>Sources and binaries can be found at , release notes are copied below.<br>
<br>
Enjoy,<font color="#888888"><br></font></span><div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px">The SciPy developers</span></div>
<div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"><br></span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"><br>
</span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"><br></span></div><div><span style="border-collapse:collapse;font-family:arial, sans-serif;font-size:13px"><div>
==========================</div><div>SciPy 0.10.0 Release Notes</div><div>==========================</div><div><br></div><div>.. contents::</div><div><br></div><div>SciPy 0.10.0 is the culmination of 8 months of hard work. It contains</div>
<div>many new features, numerous bug-fixes, improved test coverage and</div><div>better documentation. There have been a limited number of deprecations</div><div>and backwards-incompatible changes in this release, which are documented</div>
<div>below. All users are encouraged to upgrade to this release, as there</div><div>are a large number of bug-fixes and optimizations. Moreover, our </div><div>development attention will now shift to bug-fix releases on the 0.10.x </div>
<div>branch, and on adding new features on the development master branch.</div><div><br></div><div>Release highlights:</div><div><br></div><div> - Support for Bento as optional build system.</div><div> - Support for generalized eigenvalue problems, and all shift-invert modes</div>
<div> available in ARPACK.</div><div><br></div><div>This release requires Python 2.4-2.7 or 3.1- and NumPy 1.5 or greater.</div><div><br></div><div><br></div><div>New features</div><div>============</div><div><br></div>
<div>Bento: new optional build system</div><div>--------------------------------</div><div><br></div><div>Scipy can now be built with `Bento <<a href="http://cournape.github.com/Bento/" target="_blank">http://cournape.github.com/Bento/</a>>`_.</div>
<div>Bento has some nice features like parallel builds and partial rebuilds, that</div><div>are not possible with the default build system (distutils). For usage</div><div>instructions see BENTO_BUILD.txt in the scipy top-level directory.</div>
<div><br></div><div>Currently Scipy has three build systems, distutils, numscons and bento.</div><div>Numscons is deprecated and is planned and will likely be removed in the next</div><div>release.</div><div><br></div><div>
<br></div><div>Generalized and shift-invert eigenvalue problems in ``scipy.sparse.linalg``</div><div>---------------------------------------------------------------------------</div><div><br></div><div>The sparse eigenvalue problem solver functions</div>
<div>``scipy.sparse.eigs/eigh`` now support generalized eigenvalue</div><div>problems, and all shift-invert modes available in ARPACK.</div><div><br></div><div><br></div><div>Discrete-Time Linear Systems (``scipy.signal``)</div>
<div>-----------------------------------------------</div><div><br></div><div>Support for simulating discrete-time linear systems, including</div><div>``scipy.signal.dlsim``, ``scipy.signal.dimpulse``, and ``scipy.signal.dstep``,</div>
<div>has been added to SciPy. Conversion of linear systems from continuous-time to</div><div>discrete-time representations is also present via the</div><div>``scipy.signal.cont2discrete`` function.</div><div><br></div><div>
<br></div><div>Enhancements to ``scipy.signal``</div><div>--------------------------------</div><div><br></div><div>A Lomb-Scargle periodogram can now be computed with the new function</div><div>``scipy.signal.lombscargle``.</div>
<div><br></div><div>The forward-backward filter function ``scipy.signal.filtfilt`` can now</div><div>filter the data in a given axis of an n-dimensional numpy array.</div><div>(Previously it only handled a 1-dimensional array.) Options have been</div>
<div>added to allow more control over how the data is extended before filtering.</div><div><br></div><div>FIR filter design with ``scipy.signal.firwin2`` now has options to create</div><div>filters of type III (zero at zero and Nyquist frequencies) and IV (zero at zero</div>
<div>frequency).</div><div><br></div><div><br></div><div>Additional decomposition options (``scipy.linalg``)</div><div>---------------------------------------------------</div><div><br></div><div>A sort keyword has been added to the Schur decomposition routine </div>
<div>(``scipy.linalg.schur``) to allow the sorting of eigenvalues in</div><div>the resultant Schur form.</div><div><br></div><div>Additional special matrices (``scipy.linalg``)</div><div>----------------------------------------------</div>
<div><br></div><div>The functions ``hilbert`` and ``invhilbert`` were added to ``scipy.linalg``.</div><div><br></div><div><br></div><div>Enhancements to ``scipy.stats``</div><div>-------------------------------</div><div>
<br></div><div>* The *one-sided form* of Fisher's exact test is now also implemented in</div><div> ``stats.fisher_exact``. </div><div>* The function ``stats.chi2_contingency`` for computing the chi-square test of</div>
<div> independence of factors in a contingency table has been added, along with</div><div> the related utility functions ``stats.contingency.margins`` and</div><div> ``stats.contingency.expected_freq``.</div><div><br>
</div>
<div><br></div><div>Basic support for Harwell-Boeing file format for sparse matrices</div><div>----------------------------------------------------------------</div><div><br></div><div>Both read and write are support through a simple function-based API, as well as</div>
<div>a more complete API to control number format. The functions may be found in</div><div><a href="http://scipy.sparse.io" target="_blank">scipy.sparse.io</a>.</div><div><br></div><div>The following features are supported:</div>
<div><br>
</div><div> * Read and write sparse matrices in the CSC format</div><div> * Only real, symmetric, assembled matrix are supported (RUA format)</div><div><br></div><div><br></div><div>Deprecated features</div><div>===================</div>
<div><br></div><div>``scipy.maxentropy``</div><div>--------------------</div><div><br></div><div>The maxentropy module is unmaintained, rarely used and has not been functioning</div><div>well for several releases. Therefore it has been deprecated for this release,</div>
<div>and will be removed for scipy 0.11. Logistic regression in scikits.learn is a</div><div>good alternative for this functionality. The ``scipy.maxentropy.logsumexp``</div><div>function has been moved to ``scipy.misc``.</div>
<div><br></div><div><br></div><div>``scipy.lib.blas``</div><div>------------------</div><div><br></div><div>There are similar BLAS wrappers in ``scipy.linalg`` and ``scipy.lib``. These</div><div>have now been consolidated as ``scipy.linalg.blas``, and ``scipy.lib.blas`` is</div>
<div>deprecated.</div><div><br></div><div><br></div><div>Numscons build system</div><div>---------------------</div><div><br></div><div>The numscons build system is being replaced by Bento, and will be removed in</div><div>
one of the next scipy releases.</div><div><br></div><div><br></div><div>Backwards-incompatible changes</div><div>==============================</div><div><br></div><div>The deprecated name `invnorm` was removed from ``scipy.stats.distributions``,</div>
<div>this distribution is available as `invgauss`.</div><div><br></div><div>The following deprecated nonlinear solvers from ``scipy.optimize`` have been</div><div>removed::</div><div><br></div><div> - ``broyden_modified`` (bad performance)</div>
<div> - ``broyden1_modified`` (bad performance)</div><div> - ``broyden_generalized`` (equivalent to ``anderson``)</div><div> - ``anderson2`` (equivalent to ``anderson``)</div><div> - ``broyden3`` (obsoleted by new limited-memory broyden methods)</div>
<div> - ``vackar`` (renamed to ``diagbroyden``)</div><div> </div><div><br></div><div>Other changes</div><div>=============</div><div><br></div><div>``scipy.constants`` has been updated with the CODATA 2010 constants.</div>
<div><br></div><div>``__all__`` dicts have been added to all modules, which has cleaned up the</div><div>namespaces (particularly useful for interactive work).</div><div><br></div><div>An API section has been added to the documentation, giving recommended import</div>
<div>guidelines and specifying which submodules are public and which aren't.</div><div><br></div><div><br></div><div>Authors</div><div>=======</div><div><br></div><div>This release contains work by the following people (contributed at least</div>
<div>one patch to this release, names in alphabetical order):</div><div><br></div><div>* Jeff Armstrong +</div><div>* Matthew Brett</div><div>* Lars Buitinck +</div><div>* David Cournapeau</div><div>* FI$H 2000 +</div><div>
* Michael McNeil Forbes +</div><div>* Matty G +</div><div>* Christoph Gohlke</div><div>* Ralf Gommers</div><div>* Yaroslav Halchenko</div><div>* Charles Harris</div><div>* Thouis (Ray) Jones +</div><div>* Chris Jordan-Squire +</div>
<div>* Robert Kern</div><div>* Chris Lasher +</div><div>* Wes McKinney +</div><div>* Travis Oliphant</div><div>* Fabian Pedregosa</div><div>* Josef Perktold</div><div>* Thomas Robitaille +</div><div>* Pim Schellart +</div>
<div>* Anthony Scopatz +</div><div>* Skipper Seabold +</div><div>* Fazlul Shahriar +</div><div>* David Simcha +</div><div>* Scott Sinclair +</div><div>* Andrey Smirnov +</div><div>* Collin RM Stocks +</div><div>* Martin Teichmann +</div>
<div>* Jake Vanderplas +</div><div>* Gaël Varoquaux +</div><div>* Pauli Virtanen</div><div>* Stefan van der Walt</div><div>* Warren Weckesser</div><div>* Mark Wiebe +</div><div><br></div><div>A total of 35 people contributed to this release.</div>
<div>People with a "+" by their names contributed a patch for the first time.</div><div><br></div></span></div>
<br></blockquote><div><br>Congratulations to all. This looks like a nice release.<br><br>Chuck <br></div><br></div>