Mailman 3 December 2009 - SciPy-User

Functional nonlinear equation
by Dorian Dec. 27, 2009

Dec. 27, 2009

Hi all, How do I solve the following functional nonlinear equation f(x,y)= (-1/a^2)*(x*y)^(1/a-1)+(1/a)*x^(1/a-1)+(1/a)*y^(1/a-1) =0 where "a" is positive parameter. If there is a way to plot f(x,y) for different values of "a" it will be very appreciated. Thanking you in advance Dorian

6 7

Re: [SciPy-User] Parallelizing a for loop
by Scott Askey Dec. 26, 2009

Dec. 26, 2009

Anne: A pointer to the tools in numpy to apply array operation to my function would be appreciated. I found found the scipy.vectorize but it did not seem applicable to my 42 input As best I could tell I seemed that the time required to calculate two elements (2 x dof in residual 500 times) was the same as calculating 1 x dof residual 1000 times in a for loop. Chuck: It is a time marching structural dynamics 1-d beam problem. The 100 or so parameters are the initial conditions, geometric and material properties. The 42 dof are the configuration of the element at the end of the time step. Given the array of (42 * elements) initial conditions the residuals for each element may be independently calculated. Currently I calculate the 42*elements residual by running "for i in range(element)" and getting 42 residuals in each iteration. My element residual function is at most at worst cubic polynomials. The element wise residual and fprime functions are lambdafiedsympy functions. Cheers, Scott

2 1

Re: [SciPy-User] Parallelizing a for loop
by Askey, Scott A Capt USAF AETC AFIT/ENY Dec. 26, 2009

Dec. 26, 2009

Chuck For a "small" problem 10 elements (420 dof total) runs in 2.2 second when fsolve is is given a jacobian function the problem solves in 2.2 seconds jerval=3 and feval =23 , without the jacobian it takes 20+ seconds. Scott - The reason I asked is that there are a couple of finite element packages available for python. Two of them are SfePy <http://sfepy.org/> and FiPy<http://www.ctcms.nist.gov/fipy/>. I don't know if either is directly applicable to your problem. As to fsolve, are you directly computing the Jacobian or do you let the routine compute it numerically? Chuck

1 0

[ANN] pandas 0.1, a new NumPy-based data analysis library
by Wes McKinney Dec. 26, 2009

Dec. 26, 2009

Hello all, I'm very happy to announce the release of a new data analysis library that many of you will hopefully find useful. This release is the product of a long period of development and use; hence, despite the low version number, it is quite suitable for general use. The documentation is still a bit sparse but will become much more complete in the coming weeks and months. Info / Documentation: http://pandas.sourceforge.net/ Overview slides: http://pandas.googlecode.com/files/nyfpug.pdf What it is ========== pandas is a library for pan-el da-ta analysis, i.e. multidimensional time series and cross-sectional data sets commonly found in statistics, econometrics, or finance. It provides convenient and easy-to-understand NumPy-based data structures for generic labeled data, with focus on automatically aligning data based on its label(s) and handling missing observations. One major goal of the library is to simplify the implementation of statistical models on unreliable data. Main Features ============= * Data structures: for 1, 2, and 3 dimensional labeled data sets. Some of their main features include: * Automatically aligning data * Handling missing observations in calculations * Convenient slicing and reshaping ("reindexing") functions * Provide 'group by' aggregation or transformation functionality * Tools for merging / joining together data sets * Simple matplotlib integration for plotting * Date tools: objects for expressing date offsets or generating date ranges; some functionality similar to scikits.timeseries * Statistical models: convenient ordinary least squares and panel OLS implementations for in-sample or rolling time series / cross-sectional regressions. These will hopefully be the starting point for implementing other models pandas is not necessarily intended as a standalone library but rather as something which can be used in tandem with other NumPy-based packages like scikits.statsmodels. Where possible wheel-reinvention has largely been avoided. Also, its time series manipulation capability is not as extensive as scikits.timeseries; pandas does have its own time series object which fits into the unified data model. Some other useful tools for time series data (moving average, standard deviation, etc.) are available in the codebase but do not yet have a convenient interface. These will be highlighted in a future release. Where to get it =============== The source code is currently hosted on googlecode at: http://pandas.googlecode.com Releases can be downloaded currently on the Python package index or using easy_install PyPi: http://pypi.python.org/pypi/pandas/ License ======= BSD Documentation ============= The official documentation is hosted on SourceForge. http://pandas.sourceforge.net/ The sphinx documentation is still in an incomplete state, but it should provide a good starting point for learning how to use the library. Expect the docs to continue to expand as time goes on. Background ========== Work on pandas started at AQR (a quantitative hedge fund) in 2008 and has been under active development since then. Discussion and Development ========================== Since pandas development is related to a number of other scientific Python projects, questions are welcome on the scipy-user mailing list. Specialized discussions or design issues should take place on the pystatsmodels mailing list / google group, where scikits.statsmodels and other libraries will also be discussed: http://groups.google.com/group/pystatsmodels Best regards, Wes McKinney

2 1

Re: [SciPy-User] Parallelizing a for loop
by Scott Askey Dec. 24, 2009

Dec. 24, 2009

Also, where might I find guidance for how to get rid of for loops/ ----- Original Message ---- From: Scott Askey <scotta_2002(a)yahoo.com> To: scipy-user(a)scipy.org Sent: Thu, December 24, 2009 9:50:13 AM Subject: Re: Parallelizing a for loop Anne: A pointer to the tools in numpy to apply array operation to my function would be appreciated. I found found the scipy.vectorize but it did not seem applicable to my 42 input As best I could tell I seemed that the time required to calculate two elements (2 x dof in residual 500 times) was the same as calculating 1 x dof residual 1000 times in a for loop. Chuck: It is a time marching structural dynamics 1-d beam problem. The 100 or so parameters are the initial conditions, geometric and material properties. The 42 dof are the configuration of the element at the end of the time step. Given the array of (42 * elements) initial conditions the residuals for each element may be independently calculated. Currently I calculate the 42*elements residual by running "for i in range(element)" and getting 42 residuals in each iteration. My element residual function is at most at worst cubic polynomials. The element wise residual and fprime functions are lambdafiedsympy functions. Cheers, Scott

2 1

Parallelizing a for loop
by Askey, Scott A Capt USAF AETC AFIT/ENY Dec. 23, 2009

Dec. 23, 2009

I am looking for a good way to speed the calculation of the of the residuals and and the Jacobian of for fsolve. For a small problem with 10 elements (420x420) the Jacobians .04 seconds to calculate and the residuals 420x1 take .007 seconds. The residual and Jacobian elements are independent and calculated in a for loop. Is threading or multiprocessing more appropriate. Cheers, Scott

3 2

scipy unit test failures
by Keith Goodman Dec. 22, 2009

Dec. 22, 2009

I built scipy for the first time (0.7.1). No different than building numpy, it just takes longer. Somebody did something right when designing all this. I get 16 failures. They all look similar to this: ====================================================================== FAIL: test_identity (test_construct.TestConstructUtils) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/local/lib/python2.6/site-packages/scipy/sparse/tests/test_construct.py", line 70, in test_identity assert_equal( I.dtype, 'int8' ) File "/usr/local/lib/python2.6/site-packages/numpy/testing/utils.py", line 284, in assert_equal raise AssertionError(msg) AssertionError: Items are not equal: ACTUAL: dtype('int8') DESIRED: 'int8' Is this anything to worry about? Or it is just a missing .type in the unit tests?

1 0

Re: [SciPy-User] SciPy-User Digest, Vol 76, Issue 38
by Massimo Di Stefano Dec. 22, 2009

Dec. 22, 2009

Guys, thanks to All ! removing the old scipy.egg (from the superpack installation) all works fine no more segfault. i also tried the test getting this results : ---------------------------------------------------------------------- Ran 4871 tests in 937.000s FAILED (KNOWNFAIL=8, SKIP=38, errors=5, failures=4) Out[4]: <nose.result.TextTestResult run=4871 errors=5 failures=4> thanks for the wonderful link too! in it there are a lot of usefull information. thanks again, happy end-year ti All! Massimo. Il giorno 22/dic/2009, alle ore 05.39, scipy-user-request(a)scipy.org ha scritto: >> Hi, >> >> i'm tring to install scipy from svn on mac osx 10.6

1 0

equivalent of tolist().index(entry) for numpy 1d array of strings
by Ryan Krauss Dec. 22, 2009

Dec. 22, 2009

I wrote some code to work with csv spreadsheet files by reading the columns into lists, but I need to rework the code to work with numpy 1d arrays of strings rather than lists. I need to search one of these columns/arrays. What is the best way to find the index for the element that matches a certain string (or maybe just the first element to match such a string)? With the columns as lists, I was doing index = mylist.index(entry) So, I could obviously do index = mylist.tolist().index(entry) but I don't know if that would be slower or clumsier than something like bool_vect = where(mylist==entry)[0] index = bool_vect[0] or just index = where(mylist==entry)[0][0] Any thoughts? Is there an easier way? Thanks, Ryan

3 4

Numpy Building(AMD64)
by Ewan Maxwell Dec. 21, 2009

Dec. 21, 2009

Hi! I cannot [Build Numpy] which I want for [PyCuda] I have [Windows 7(x64)] and PSDK 6.1(tried 7, didn't work either) Compilers: VS 2008(Not Express) gfortran(http://gcc.gnu.org/wiki/GFortranBinaries; http://quatramaran.ens.fr/~coudert/gfortran/gfortran-windows.exe) Python 2.6.4(AMD64) I downloaded the numpy-1.3.0 source I try python.exe setup.py install I get this C:\Program Files (x86)\Microsoft Visual Studio 9.0\VC\BIN\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:C:\Apps\Python26\libs /LIBPATH:C:\Apps\Python26\PCbuil \amd64 /LIBPATH:build\temp.win-amd64-2.6 /EXPORT:init_sort build\temp.win-amd64 2.6\Release\build\src.win-amd64-2.6\numpy\core\src\_sortmodule.obj /OUT:build\l b.win-amd64-2.6\numpy\core\_sort.pyd /IMPLIB:build\temp.win-amd64-2.6\Release\b ild\src.win-amd64-2.6\numpy\core\src\_sort.lib /MANIFESTFILE:build\temp.win-amd 4-2.6\Release\build\src.win-amd64-2.6\numpy\core\src\_sort.pyd.manifest Creating library build\temp.win-amd64-2.6\Release\build\src.win-amd64-2.6\nu py\core\src\_sort.lib and object build\temp.win-amd64-2.6\Release\build\src.win amd64-2.6\numpy\core\src\_sort.exp _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyErr_Format referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyExc_Runtim Error referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyCObject_As oidPtr referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyCObject_Ty e referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyObject_Get ttrString referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyImport_Imp rtModule referenced in function __import_array _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyErr_NoMemo y referenced in function _BOOL_mergesort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyMem_Free r ferenced in function _BOOL_amergesort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyMem_Malloc referenced in function _BOOL_amergesort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyErr_SetStr ng referenced in function _init_sort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyExc_Import rror referenced in function _init_sort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__PyErr_Print eferenced in function _init_sort _sortmodule.obj : error LNK2019: unresolved external symbol __imp__Py_InitModul 4 referenced in function _init_sort build\lib.win-amd64-2.6\numpy\core\_sort.pyd : fatal error LNK1120: 13 unresolv d externals error: Command "C:\Program Files (x86)\Microsoft Visual Studio 9.0\VC\BIN\link. xe /DLL /nologo /INCREMENTAL:NO /LIBPATH:C:\Apps\Python26\libs /LIBPATH:C:\Apps Python26\PCbuild\amd64 /LIBPATH:build\temp.win-amd64-2.6 /EXPORT:init_sort buil \temp.win-amd64-2.6\Release\build\src.win-amd64-2.6\numpy\core\src\_sortmodule. bj /OUT:build\lib.win-amd64-2.6\numpy\core\_sort.pyd /IMPLIB:build\temp.win-amd 4-2.6\Release\build\src.win-amd64-2.6\numpy\core\src\_sort.lib /MANIFESTFILE:bu ld\temp.win-amd64-2.6\Release\build\src.win-amd64-2.6\numpy\core\src\_sort.pyd. anifest" failed with exit status 1120 What I'm I doing wrong? What can I do better?

2 2