Using Python for large scale linear algebra applications

Python for Unified Research in Econometrics and Statistics

“…Python 2.6.5 includes the core language and standard libraries, and is installed automatically from www.python.org/download, by following the download instructions. 4 After installation, the Python Integrated Development Environment (IDLE), or 'shell window' (interactive interpreter) becomes available (see Figure 1). The compatible Pywin32 should also be installed.…”

“…This provides extension modules for access to many Windows API (Application Programming Interface) functions, and is available from sourceforge.net. 4 The Windows Installer file for Python 2.6.5 is python-2.6.5.msi.…”

“…>>> squares_cubes=[[x**2,x**3] for x in range(6) if x**2<=16]; print squares_cubes \ # list comprehension [[0, 0], [1,1], [4,8], [9,27], [16, 64]] >>> for x,y in enumerate(squares_cubes): print x,y, # the enumerate() function [4,8] 3 [9,27] 4 [16,64] >>> print zip(range (5),squares_cubes) # the zip(function) (2, [4, 8]), (3, [9, 27]), (4, [16, 64]…”

“…In industry and academia, Python has become an alternative to low-level compiled languages such as C++. Recent examples in large-scale computational applications include [4], [16], [17], [19] and [20, who explicitly refers to faster development times], and indicate comparable run times with C++ implementations in some situations (although we would generally expect some overhead from using an interpreted language). The Python Wiki lists Google, Industrial Light and Magic, and Yahoo!…”

Python for Unified Research in Econometrics and Statistics

“…In industry and academia, Python has become an alternative to low-level compiled languages such as C++. Recent examples in large-scale computational applications include Bröker et al (2005), Lucks (2008), Meinke et al (2008), Nilsen (2007a), and Nilsen (2007b)-the latter explicitly referring to faster development times-and indicate comparable run times with C++ implementations in some situations (although we would generally expect some overhead from using an interpreted language). The Python Wiki lists Google, Industrial Light and Magic, and Yahoo!…”