NumPy, SciPy and SciKits - Revision history

WikiSysop at 03:02, 10 April 2018

2018-04-10T03:02:05Z

← Older revision		Revision as of 03:02, 10 April 2018
Line 210:		Line 210:
	array([0, 1, 2, 3, 4])		array([0, 1, 2, 3, 4])

	=== Functions ~~broadcasting over~~ an ~~array~~ ===		=== Functions Broadcasting Over an Array ===

	The term ''broadcasting'' in NumPy refers to applying an operation to every element of an array. Consider this simple example		The term ''broadcasting'' in NumPy refers to applying an operation to every element of an array. Consider this simple example
Line 289:		Line 289:
	Printing index will show an array of True and False that mask the original array. The array b will only have those elements that satisified the condition used to make the index. The index is used again, to put modified elements back into the original array.		Printing index will show an array of True and False that mask the original array. The array b will only have those elements that satisified the condition used to make the index. The index is used again, to put modified elements back into the original array.

	=== Matrix and ~~vector math~~ in NumPy ===		=== Matrix and Vector Math in NumPy ===

	We have seen that NumPy has its own set of math functions that broadcast over arrays. If you define an array		We have seen that NumPy has its own set of math functions that broadcast over arrays. If you define an array

WikiSysop at 04:47, 28 January 2017

2017-01-28T04:47:42Z

← Older revision		Revision as of 04:47, 28 January 2017
Line 761:		Line 761:
	=== SciKits Packages ===		=== SciKits Packages ===

	We mention SciKits in the context of SciPy because it is a collection of SciPy applications that extend into many different domains of physics, astronomy, and engineering. The best introduction is to browse the [http://scikits.appspot.com/scikits SciKits website] and explore the variety there. In addition to image processing, you'll find speech processing, light scattering from non-spherical particles, sparse matrices, machine learning, vector field plotting, and more. Some are fully developed, others are works in progress. One of the most promising is the [http://scikits.appspot.com/cuda scikits.cuda] package, which interfaces Python to CUDA, NVidia's parallel processing through graphical processing units (GPU). The most recent hardware at this date (February ~~2015~~), is the		We mention SciKits in the context of SciPy because it is a collection of SciPy applications that extend into many different domains of physics, astronomy, and engineering. The best introduction is to browse the [http://scikits.appspot.com/scikits SciKits website] and explore the variety there. In addition to image processing, you'll find speech processing, light scattering from non-spherical particles, sparse matrices, machine learning, vector field plotting, and more. Some are fully developed, others are works in progress. One of the most promising is the [http://scikits.appspot.com/cuda scikits.cuda] package, which interfaces Python to CUDA, NVidia's parallel processing through graphical processing units (GPU). The most recent hardware at this date (February 2017), is the
	the [http://www.nvidia.com/object/tesla-servers.html Tesla ~~K80~~] for servers, a supercomputer on a card with ~~4992~~ GPU cores, 8.7 Teraflops of single precision and 5.6 Teraflops of double precision processing power, and ~~24 GB graphic DDR5~~ memory.		the [http://www.nvidia.com/object/tesla-servers.html Tesla P100] Pascal architecture for servers, a supercomputer on a PCIe card with 3584 GPU cores, 9.3 Teraflops of single precision and 4.7 Teraflops of double precision processing power, and 16384 MB of memory.

			An example of a very useful Scikit is the [http://scikit-image.org/docs/dev/api/skimage.restoration.html image restoration package] which includes Lucy-Richardson deconvolution. A program that makes use of it for astronomical fits images is in the [http://prancer.physics.louisville.edu/classes/650/python/examples examples directory] of this site. Look for fits_lucy_richardson.py under the convolution examples.

	We will return to use ~~the~~ SciPy image processing routines later.

			We will return to use other SciPy image processing routines later.

WikiSysop at 04:25, 28 January 2017

2017-01-28T04:25:28Z

← Older revision		Revision as of 04:25, 28 January 2017
Line 757:		Line 757:
	std = y.std()		std = y.std()
	var = y.var()		var = y.var()


	=== SciKits Packages ===		=== SciKits Packages ===
Line 765:		Line 766:

	We will return to use the SciPy image processing routines later.		We will return to use the SciPy image processing routines later.


	== Examples ==		== Examples ==
Line 781:		Line 783:

	that contain most of the code shown in this section.		that contain most of the code shown in this section.


	== Assignments ==		== Assignments ==

	For the assigned homework to use these ideas, see the [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_assignments assignments] section.		For the assigned homework to use these ideas, see the [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_assignments assignments] section.

WikiSysop: /* SciKits Packages */

2015-02-03T21:43:58Z

SciKits Packages

← Older revision		Revision as of 21:43, 3 February 2015
Line 760:		Line 760:
	=== SciKits Packages ===		=== SciKits Packages ===

	We mention SciKits in the context of SciPy because it is a collection of SciPy applications that extend into many different domains of physics, astronomy, and engineering. The best introduction is to browse the [http://scikits.appspot.com/scikits SciKits website] and explore the variety there. In addition to image processing, you'll find speech processing, light scattering from non-spherical particles, sparse matrices, machine learning, vector field plotting, and more. Some are fully developed, others are works in progress. One of the most promising is the [http://scikits.appspot.com/cuda scikits.cuda] package, which interfaces Python to CUDA, NVidia's parallel processing through graphical processing units (GPU). The most recent hardware at this date (February ~~2013~~), is		We mention SciKits in the context of SciPy because it is a collection of SciPy applications that extend into many different domains of physics, astronomy, and engineering. The best introduction is to browse the [http://scikits.appspot.com/scikits SciKits website] and explore the variety there. In addition to image processing, you'll find speech processing, light scattering from non-spherical particles, sparse matrices, machine learning, vector field plotting, and more. Some are fully developed, others are works in progress. One of the most promising is the [http://scikits.appspot.com/cuda scikits.cuda] package, which interfaces Python to CUDA, NVidia's parallel processing through graphical processing units (GPU). The most recent hardware at this date (February 2015), is the
	the [http://~~nvidianews~~.nvidia.com/~~Releases~~/~~NVIDIA-Introduces-GeForce-GTX-TITAN-DNA-of-the~~-~~World-s-Fastest-Supercomputer-Powered-by-World-s-Fa-925~~.~~aspx GeForce GTX Titan~~], a supercomputer on a card with ~~2668~~ GPU cores, 4.5 Teraflops of single precision and 1.3 Teraflops of double precision processing power, and 6 GB graphic DDR5 memory.		the [http://www.nvidia.com/object/tesla-servers.html Tesla K80] for servers, a supercomputer on a card with 4992 GPU cores, 8.7 Teraflops of single precision and 5.6 Teraflops of double precision processing power, and 24 GB graphic DDR5 memory.

WikiSysop: /* Matrix and vector math in NumPy */

2015-01-29T21:37:45Z

Matrix and vector math in NumPy

← Older revision		Revision as of 21:37, 29 January 2015
Line 337:		Line 337:
	2x + y = 1		2x + y = 1
	3y + z = 1		3y + z = 1
	x + 4z =		x + 4z = 1

	We would write		We would write

WikiSysop at 04:28, 8 December 2013

2013-12-08T04:28:58Z

← Older revision		Revision as of 04:28, 8 December 2013
Line 1:		Line 1:
	Python provides a framework on which numerical and scientific data processing can be built. As part of our [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_for_Physics_and_Astronomy short course on Python for Physics and Astronomy] we will look at the capabilities of the [http://www.numpy.org NumPy], [http://www.scipy.org/ SciPy] and [http://scikits.appspot.com/scikits SciKits] packages. This is a brief overview with a few examples drawn primarily from the excellent but short introductory book ''SciPy and NumPy'' by [http://www.oreillynet.com/pub/au/4782 Eli Bressert (O'Reilly 2012)].		Python provides a framework on which numerical and scientific data processing can be built. As part of our [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_for_Physics_and_Astronomy short course on Python for Physics and Astronomy] we will look at the capabilities of the [http://www.numpy.org NumPy], [http://www.scipy.org/ SciPy] and [http://scikits.appspot.com/scikits SciKits] packages. This is a brief overview with a few examples drawn primarily from the excellent but short introductory book ''SciPy and NumPy'' by [http://www.oreillynet.com/pub/au/4782 Eli Bressert (O'Reilly 2012)]. The National Radio Astronomy Observatory's Common Astronomy Software site has a
			[http://casaguides.nrao.edu/index.php?title=NumpyBasics helpful introduction to NumPy].

	== NumPy==		== NumPy==

WikiSysop: /* Fourier Transforms in NumPy */

2013-08-19T02:34:01Z

Fourier Transforms in NumPy

← Older revision		Revision as of 02:34, 19 August 2013
Line 459:		Line 459:
	[[File:Numpy_fft.png \| center \| 600 px]]		[[File:Numpy_fft.png \| center \| 600 px]]

	In practical applications it is desirable to have programs that will read files containing time or frequency data, perform the Fourier transform, and export a file with the transform from frequency to time, or from time to frequency. Two programs that are useful in this way are included in the examples section:		In practical applications it is desirable to have programs that will read files containing time or frequency data, perform the Fourier transform, and export a file with the transform from frequency to time, or from time to frequency. Two programs that are useful in this way are included in the [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_examples examples] section. They depend on SciPy, which is described in the following.

	*fft_f2t.py		*fft_f2t.py
	*fft_t2f.py		*fft_t2f.py

	== SciPy and SciKits ==		== SciPy and SciKits ==

WikiSysop: /* Fourier Transforms in NumPy */

2013-08-19T02:31:47Z

Fourier Transforms in NumPy

← Older revision		Revision as of 02:31, 19 August 2013
Line 459:		Line 459:
	[[File:Numpy_fft.png \| center \| 600 px]]		[[File:Numpy_fft.png \| center \| 600 px]]

			In practical applications it is desirable to have programs that will read files containing time or frequency data, perform the Fourier transform, and export a file with the transform from frequency to time, or from time to frequency. Two programs that are useful in this way are included in the examples section:

			*fft_f2t.py
			*fft_t2f.py

	== SciPy and SciKits ==		== SciPy and SciKits ==

WikiSysop: /* Examples */

2013-08-19T02:27:49Z

Examples

← Older revision		Revision as of 02:27, 19 August 2013
Line 772:		Line 772:
	*norm_probability_plot.example		*norm_probability_plot.example
	*numpy_fft.example		*numpy_fft.example
			*fft_f2t.py
			*fft_t2f.py

	that contain most of the code shown in this section.		that contain most of the code shown in this section.

WikiSysop: /* Examples */

2013-02-26T06:49:58Z

Examples

← Older revision		Revision as of 06:49, 26 February 2013
Line 765:		Line 765:
	For examples of NumPy and SciPy, see the [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_examples examples] section. Look for		For examples of NumPy and SciPy, see the [http://prancer.physics.louisville.edu/astrowiki/index.php/Python_examples examples] section. Look for

			*curve_fit_plot.example
	*integrate_function_plot.example		*integrate_function_plot.example
	*integrate_numerical_plot.example		*integrate_numerical_plot.example