NumPy-Discussion December 2003

numpy-discussion@python.org

42 participants
61 discussions

PyArray_ContiguousFromObject & PyDECREF
by Michiel Jan Laurens de Hoon 16 May '11

16 May '11

Dear pythoneers, In Chapter 12 ("Writing a C extension") in the Numerical Python manual, in the first example PyArray_ContiguousFromObject is used together with PyDECREF, while in the last example PyArray_ContiguousFromObject is used but the arrays are not PyDECREF'ed before the function returns. From Chapter 13 ("C API Reference") it seems that a PyDECREF is needed in both cases. Or am I missing something here? Is PyDECREF not needed for some reason in the last example? Thanks in advance, Michiel de Hoon U Tokyo First example: ============== static PyObject * trace(PyObject *self, PyObject *args) { PyObject *input; PyArrayObject *array; double sum; int i, n; if (!PyArg_ParseTuple(args, "O", &input)) return NULL; array = (PyArrayObject *) PyArray_ContiguousFromObject(input, PyArray_DOUBLE, 2, 2); if (array == NULL) return NULL; n = array->dimensions[0]; if (n > array->dimensions[1]) n = array->dimensions[1]; sum = 0.; for (i = 0; i < n; i++) sum += *(double *)(array->data + i*array->strides[0] + i*array->strides[1]); Py_DECREF(array); return PyFloat_FromDouble(sum); } Last example: ============= static PyObject * matrix_vector(PyObject *self, PyObject *args) { PyObject *input1, *input2; PyArrayObject *matrix, *vector, *result; int dimensions[1]; double factor[1]; double real_zero[1] = {0.}; long int_one[1] = {1}; long dim0[1], dim1[1]; extern dgemv_(char *trans, long *m, long *n, double *alpha, double *a, long *lda, double *x, long *incx, double *beta, double *Y, long *incy); if (!PyArg_ParseTuple(args, "dOO", factor, &input1, &input2)) return NULL; matrix = (PyArrayObject *) PyArray_ContiguousFromObject(input1, PyArray_DOUBLE, 2, 2); if (matrix == NULL) return NULL; vector = (PyArrayObject *) PyArray_ContiguousFromObject(input2, PyArray_DOUBLE, 1, 1); if (vector == NULL) return NULL; if (matrix->dimensions[1] != vector->dimensions[0]) { PyErr_SetString(PyExc_ValueError, "array dimensions are not compatible"); return NULL; } dimensions[0] = matrix->dimensions[0]; result = (PyArrayObject *)PyArray_FromDims(1, dimensions, PyArray_DOUBLE); if (result == NULL) return NULL; dim0[0] = (long)matrix->dimensions[0]; dim1[0] = (long)matrix->dimensions[1]; dgemv_("T", dim1, dim0, factor, (double *)matrix->data, dim1, (double *)vector->data, int_one, real_zero, (double *)result->data, int_one); return PyArray_Return(result); } -- Michiel de Hoon, Assistant Professor University of Tokyo, Institute of Medical Science Human Genome Center 4-6-1 Shirokane-dai, Minato-ku Tokyo 108-8639 Japan http://bonsai.ims.u-tokyo.ac.jp/~mdehoon

3 2

Fw: [Numpy-discussion] numarray.records - get/set item
by Sebastian Haase 16 Jan '04

16 Jan '04

My situation where I got onto this, is having one field named 'mmm' ("MinMaxMean") being an 3 element array. Now, to assign the values first I tried: self.hdrArray = makeHdrArray(self.h) #this makes the record array self.hdr = self.hdrArray[0].field #this is my shortcut to the bound member function # it essentially is a solution (hack) for the getitem part # but regarding setitem I had to learn that "assigning to a function" is illigal in Python - as opposed to C++ #so to do assignment I need to do: self.hdr('mmm')[0], self.hdr('mmm')[1], self.hdr('mmm')[2] = (mi,ma,av) now that I'm looking at it, self.hdrArray[0].setfield('mmm', (mi,ma,av)) would probably be better... How about adding an attribute 'f' which could serve as a "proxy" to allow: myRec.f.mmm = (mi,ma,av) and maybe even additionally: myRec.f['mmm'] = (mi,ma,av) Regards, Sebastian ----- Original Message ----- From: "Perry Greenfield" <perry(a)stsci.edu> To: "Sebastian Haase" <haase(a)msg.ucsf.edu>; <numpy-discussion(a)lists.sourceforge.net> Sent: Thursday, December 04, 2003 3:08 PM Subject: RE: [Numpy-discussion] numarray.records - get/set item > > Hi, > > Is it maybe a good idea to add this to the definition of 'class Record' > > > > class Record: > > """Class for one single row.""" > > <snip> > > def __getitem__(self, fieldName): > > return self.array.field(fieldName)[self.row] > > def __setitem__(self, fieldName, value): > > self.array.field(fieldName)[self.row] = value > > > > I don't know about the implications if __delitem __ and so on are not > > defined. > > I just think it would look quite nice to say > > myRecArr[0]['mmm'] = 'hallo' > > as opposed to > > myRecArr[0].setfield('mmm', 'hallo') > > > > Actually I would even like > > myRecArr[0].mmm = 'hallo' > > > > This should be possible by defining __setattr__. > > It would obviously only work for fieldnames that do not contain '.' or ' ' > > or ... > > > > Any comments ? > > > > > We've had many internal discussions about doing this. The latter was > considered a problem because of possible name collisions of field > names with other attributes or methods. The former is not bothered > by this problem, but we decided to be conservative on this and see > how strong the need was. We are interested in other opinions. > > Perry

1 1

RE: [Numpy-discussion] Slow conversion from list to arrays
by Nadav Horesh 30 Dec '03

30 Dec '03

It works excellent. I see also that there is a corresponding update in the CVS. Happy new (and peaceful) year, Nadav. -----Original Message----- From: Todd Miller [mailto:jmiller@stsci.edu] Sent: Tue 30-Dec-03 16:34 To: Nadav Horesh Cc: numpy-discussion Subject: Re: [Numpy-discussion] Slow conversion from list to arrays On Thu, 2003-12-25 at 03:27, Nadav Horesh wrote: > The conversion for a list of large arrays into a matrix is very slow in > numarray while reasonably fast: > > >>> from numarray import * > >>> a = [arange(200000), arange(200000), arange(200000), arange(200000)] > >>> aa = array(a) # Very slow > > -------------------------------- > > >>> from Numeric import * > >>> a = [arange(200000), arange(200000), arange(200000), arange(200000)] > >>> aa = array(a) # Fast > > Python version: 2.3.3 > numarray version: Official 0.8 > > Tested under WinXP and RH9 linux. > > Any workaround? > Try concatenate. For the array sizes you're looking at, numarray does OK. You may have to adjust the result shape. Todd > Nadav > > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: IBM Linux Tutorials. > Become an expert in LINUX or just sharpen your skills. Sign up for IBM's > Free Linux Tutorials. Learn everything from the bash shell to sys admin. > Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click > _______________________________________________ > Numpy-discussion mailing list > Numpy-discussion(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/numpy-discussion -- Todd Miller Space Telescope Science Institute 3700 San Martin Drive Baltimore MD, 21030 (410) 338 - 4576

1 0

US STOCK MARKET - MBTT Technology---MERRILL LYNCH On Sector.....gaille
by Harriet Meredith 30 Dec '03

30 Dec '03

US Stock Market - Stock Profile of the Week Symbol: MBTT Market: OTC.BB Sector: ELECTRONICS Before we begin our profile we have very exciting, breaking news... Renowned Satellite Industry Expert Dr. Arthur R. Tilford Joins MB Tech BREAKING NEWS - (PRIMEZONE) - MB Tech, Inc. (OTC.BB: MBTT) announces that Dr. Arthur R. Tilford, formerly with Hughes Electronics (NYSE: GMH), has been retained by the Company to provide technical consulting and advisory services. Tilford is an aerospace industry veteran responsible for a total of 11 patents and patent applications within the field, having received Hughes' Chairman's Award for DirecDuo, solo engineering efforts in 1997. He is an expert system designer and Federal Licensing and Certifying Engineer. He has authored several books, including Satellite Fieldbook Quarterly(c) -- Hughes S&C, CA, the contents of which describe all parameters of every C/Ku band geo-satellite in the world and the over 400 page Satellite Workbook(c) for use in satellite engineering classes, taught by him, at the California State University (CSUF at CSULB). Tilford worked with Netune Communications, where he was responsible for providing transponder allocations, domestic and international link-budget calculations for narrow to very-wide-band satellite system and writing, calculating and applying for FCC and RA transmit licenses. He also assigned parameters for RF sub-system design and is an expert in radiation hazard compliance. Hanwook Bae, President of MB Tech, commented, ``The importance of Dr. Tilford joining the Company cannot be underestimated. He is an industry leader with numerous patents relating to antenna design and an immense knowledge of the satellite Internet market. His incredible technical knowledge will allow us to dramatically step up our new initiatives, catapulting us to the forefront of our industry.'' He added, ``This is yet another positive step forward in the overall paradigm shift for the Company. We have moved from being a manufacturer of satellite components, to a Company that will lead the consumer satellite electronics industry.'' STOCK PROFILE OF THE WEEK MB Tech is a global manufacturer and distributor of satellite components. The primary product MB Tech produces is "LNB" (Low-Noise Blockdown) converters, which are the essential element enabling DBS satellites to receive and convert satellite transmission signals. MB Tech currently serves the satellite television market, and is expanding to serve the satellite radio and military hardware sectors. MB Tech manufactures several proprietary LNB solutions that differentiate it from competitors. MB Tech's product line includes a dual-horn LNB which allows multiple set-top boxes to be connected to a single satellite dish, enabling viewing of multiple channels simultaneously on different television monitors, and a tri-horn LNB, which provides the ability to download signals from multiple satellites over a single dish. TECHNOLOGY MB Tech has improved on conventional designs and can manufacture units more efficiently and less expensively than the competition. MB Tech specializes in LNB electronics, it is the ONLY thing we do. An LNB converter transforms the satellite transmission into a useable signal. MB Tech has developed a special new breed of LNB converter which allows the signal to feed separate televisions or other 'boxes' with different channels. The signal coming in from one dish can be distributed through MB Tech's 'dual and triple horned' LNB converter such that multiple channels can be viewed, or different channels and Internet access can be delivered.. ALL AT THE SAME TIME! Other technological objectives have been to develop LNB for G.P.S. equipment, and small type two horn technology as well as UBC DSS type LNB. These objectives have been achieved and the Company is currently speaking with larger American based technology companies about the technology. PRODUCTS MB Tech products are innovative! We rely on superior technology to be an integral part of our business. We have made a serious committment to ongoing R&D to stay ahead of the curve. MB Tech, Inc traces its roots back to 2000 with the development of 3 types of LNB (DSS type). Continuous reinvention and rapid transformation are themes throughout our history. Led by our top-notch technology, our company offered a new breed of satellite broadcasting solutions to clients?solutions that customized LNB with 3 or 4 horns. Click here to see our product milestones. Within the next two years, MB Tech will spend over 80 million dollars on continuing Research and development of our products. We will continue a process of product renewal that will assure products with features and functionality that will be attractive to the market. MDI: MERRILL LYNCH COMMENTS ON AD TRENDS IN THE MEDIA SECTOR Merrill Lynch surveyed newspapers, TV stations and radio stations to try to understand the divergent trends between local and national advertising in the current recovery. They noted that U.S. ad expenditures tend to keep pace with or slightly outperform the economy over time. As such, the rise of media outlets should not necessarily lead to higher ad spending but most likely to market share shifts among the mediums. The surveys showed that local cable is becoming a more competitive force, but overall, they found no major structural issues impacting the dynamics between local and national ad spending. Merrill noted that national advertisers typically base their budgets on projected economic trends and therefore can afford to spend a bit ahead of a turn in the cycle. However, local advertisers, with more limited resources, are more likely to lag in spending coming out of a recovery as they need to see an immediate return on investment. They pointed out that a better economy should eliminate the differences. The firm believes there are many positive themes occurring with newspapers today (e.g. collaborative selling efforts, product innovation, renewed focus on readership) that are being masked by the downturn in department store spending, which appears to be secular, and the cyclical issue of help wanted, which is finally turning positive. >From a consumer perspective, they believe radio is a relatively protected medium. However, the industry remains beset with excess inventory, following unsustainable demand in the late 1990's. A reduction in spots and/or lower prices could limit near term growth, yet support long term growth, giving investors something to chase. TV is about to enter the growth phase of its two year cycle, but its low growth, single revenue stream is likely the main target for local cable TV and should also be impacted by increased DVR penetration, which gives the firm little to get excited about. IMPORTANT CORPORATE MILESTONES Oct. 2003 Receives 6.14 Million Dollar Purchase Order For 45,000 Flat Antennas Sept. 2003 Purchases Patent for Key Operating Mechanism of DBS Flat Antenna Sept. 2003 Completes Development of Flat Antenna May. 2002 Canada DBS (Star Choice) two horn LNB P/O for (USD 450,000) Feb. 2002 Canada DBS (Star Choice) two horn LNB P/O for (USD 675,000) Dec. 2001 Canada DBS (Star Choice) two horn LNB P/O for (USD 75,000) Sept. 2001 Canada DBS (Star Choice) two horn LNB P/O for (USD 1,200,000) July 2001 Patent application for LNB design and pattern May 2001 Purchase order of Canadian BS (two horn LNB) April 2001 Establishment of MB Tech, Inc. Jan. 2001 Canada DBS two horn LNB successful field test Sept. 2000 SBCA (Satellite Broadcasting & Communications Association) Membership Certificate Aug. 2000 Development of 3 types of LNB (CAN TYPE); application of patent of new device and patent of new design (app# 00-22512,22513, 22514) July 2000 SBCA (USA); DBS field test in S. Africa; DBS field test in Australia May 2000 DBS field test in Thailand; Media Cast (London); DBS field test in Russia April 2000 Application for patent of new device and patent of new design, patent app# 10-2000-0017958 March 2000 Development of 3 types of LNB (DSS type) Please note that MBTT had absolutley nothing to do with this report and is not a participant in any way. No more advertisements: www.precscription-service-now.com/away.html Stock Market Today is an independent research firm. This report is based on Stock Market Today's independent analysis but also relies on information supplied by sources believed to be reliable. This report may not be the opinion of MBTT management. Stock Market Today has also been retained to research and issue reports on MBTT. Stock Market Today may from time to time purchase or sell MBTT common shares in the open market without notice. The information contained in this report shall not constitute, an offer to sell or solicitation of any offer to purchase any security. It is intended for information only. Some statements may contain so-called "forward-looking statements". Many factors could cause actual results to differ. Investors should consult with their Investment Advisor concerning MBTT. Copyright 2003 © Stock Market Today Ltd. All Rights Reserved. This newsletter was distributed by SCI. SCI was paid two hundred twenty thousand shares MBTT stock to distribute this report. SCI is not affiiated with Stock Market Today and is not responsible for newsletter content. h rec xfl

1 0

Slow conversion from list to arrays
by Nadav Horesh 30 Dec '03

30 Dec '03

The conversion for a list of large arrays into a matrix is very slow in numarray while reasonably fast: >>> from numarray import * >>> a = [arange(200000), arange(200000), arange(200000), arange(200000)] >>> aa = array(a) # Very slow -------------------------------- >>> from Numeric import * >>> a = [arange(200000), arange(200000), arange(200000), arange(200000)] >>> aa = array(a) # Fast Python version: 2.3.3 numarray version: Official 0.8 Tested under WinXP and RH9 linux. Any workaround? Nadav

2 1

"error getting read buffer data ptr"
by Edward C. Jones 29 Dec '03

29 Dec '03

What are the usual causes for this traceback: Traceback (most recent call last): File "/home/edcjones/cvlib/phase03/IM/slide/test03.py", line 166, in ? test4() File "/home/edcjones/cvlib/phase03/IM/slide/test03.py", line 125, in test4 print arrout File "/usr/local/lib/python2.3/site-packages/numarray/numarraycore.py", line 631, in __str__ return array_str(self) File "/usr/local/lib/python2.3/site-packages/numarray/numarraycore.py", line 1159, in array_str return arrayprint.array2string( File "/usr/local/lib/python2.3/site-packages/numarray/arrayprint.py", line 187, in array2string separator, prefix) File "/usr/local/lib/python2.3/site-packages/numarray/arrayprint.py", line 140, in _array2string data = _gen.ravel(a) File "/usr/local/lib/python2.3/site-packages/numarray/generic.py", line 916, in ravel return reshape(array, (array.nelements(),)) File "/usr/local/lib/python2.3/site-packages/numarray/generic.py", line 907, in reshape v = v.view() libnumarray.error: NA_updateDataPtr: error getting read buffer data ptr

2 1

[Fwd: Re: [Numpy-discussion] "error getting read buffer data ptr"]
by Todd Miller 29 Dec '03

29 Dec '03

This is a response I forgot to reply-all. Edward tracked this problem down to a memory management bug in his C-code. -- Todd Miller Space Telescope Science Institute 3700 San Martin Drive Baltimore MD, 21030 (410) 338 - 4576

1 0

US STOCK MARKET - MBTT Technology---MERRILL LYNCH On Sector.....antonie
by Zachariah Kincaid 28 Dec '03

28 Dec '03

1 0

Linking numarray with ATLAS
by Nadav Horesh 28 Dec '03

28 Dec '03

The following recipe worked on RH9 and gentoo 1.4: 1. Set environment variable USE_LAPACK: export USE_LAPACK=1 2. Edit addon.py and make the following changes: 1. Change the first item of lapack_libs from 'blas' to 'atlas' 2. Change lapack_dirs from [] to ['directory of libatlas.a', 'directory of libg2c.a'] 3. Proceed with the usual installation: python setup.py ... I had some problem on the RH9 system that were solved by upgrading to a recent ATLAS package. Enjoy ;) Nadav.

1 0

Linking numarray with atlas
by Nadav Horesh 28 Dec '03

28 Dec '03

The following recipe worked with RH9 and gentoo linux: 1. Set environment variable USE_LAPACK:

1 0

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

NumPy-Discussion December 2003