[Numpy-discussion] append_fields behavior with a (sort of) empty array

First consider this example of column_stack import numpy as np x = np.random.random((5,1)) y = np.random.random((5,3)) arr = np.column_stack((x[[]], y)) # this fails which is expected arr = np.column_stack((x[:,[]], y)) # this happily works I guess because x[:,[]] # array([], shape=(5, 0), dtyp

Re: [Numpy-discussion] Request for a bit more info on structured arrays in the "basics" page

On Sun, Mar 6, 2011 at 11:12 PM, Ralf Gommers wrote: > On Sun, Mar 6, 2011 at 1:10 AM, Skipper Seabold wrote: >> On Sat, Mar 5, 2011 at 9:28 AM, Ralf Gommers >> wrote: >>> On Sat, Mar 5, 2011 at 8:09 AM, Russell E. Owen wrote: The page

[Numpy-discussion] Inaccuracy in documentation of np.linalg.pinv

I think the documentation for np.linalg.pinv contains some inaccuracies. Most importantly, Moore-Penrose is not defined by the solution to the least-square problem. It was defined by the unique solution to 4 equations. Since SVD can be easily shown to satisfy the same 4 equations, it is the Moor

Re: [Numpy-discussion] Is there a memory-efficient alternative to choose?

On Mar 8, 2011, at 11:56 AM, Josh Hykes wrote: > At some point in my code, I need to do a cell-wise multiplication of the > properties with a state variable. The ideal method would (1) be fast (no > Python loops) and (2) not waste memory constructing an entire property map. > My best attempt usi

Re: [Numpy-discussion] Is this the optimal way to take index along a single axis?

On Tue, Mar 8, 2011 at 3:03 PM, Jonathan Taylor wrote: > I am wanting to use an array b to index into an array x with dimension > bigger by 1 where the element of b indicates what value to extract > along a certain direction.  For example, b = x.argmin(axis=1). > Perhaps I want to use b to create

[Numpy-discussion] Is this the optimal way to take index along a single axis?

I am wanting to use an array b to index into an array x with dimension bigger by 1 where the element of b indicates what value to extract along a certain direction. For example, b = x.argmin(axis=1). Perhaps I want to use b to create x.min(axis=1) but also to index perhaps another array of the sam

[Numpy-discussion] Is there a memory-efficient alternative to choose?

Hello, I am writing a small PDE code in Python using a Cartesian mesh. Each mesh cell is one of a few materials, with associated properties. I store these properties in a dictionary and have a "mesh map" that tells me which material is in each cell. At some point in my code, I need to do a cell-w

Re: [Numpy-discussion] How to sum weighted matrices

Or just with a dot: === In [17]: np.tensordot(weights, matrices, (0,0)) Out[17]: array([[ 5., 5., 5.], [ 5., 5., 5.]]) In [18]: np.dot(matrices.T,weights).T Out[18]: array([[ 5., 5., 5.], [ 5., 5., 5.]]) == make matrices.T C_CONTIGUOUS for maximum speed. -n On Mon, Mar

Re: [Numpy-discussion] Boolean indexing

On Tue, Mar 8, 2011 at 07:59, Sam Tygier wrote: > Hi > > I am having an  issue with boolean slicing. it seems to work fine for > reading a value, but I can use it to set a value: > > import numpy > b =  numpy.array([[1,2],[3,4],[5,6],[7,8],[9,10]]) > m = numpy.array([0,1,0,0,0], dtype=bool) > prin

Re: [Numpy-discussion] Boolean indexing

2011-03-08 14:29:07 GMT, Sturla Molden: >A "boolean slice" cannot be indexed with the dot product of dimensions and strides, hence the copy. >You probably want to use masked arrays instead. Masked array does not seem to help. when i do: am = numpy.ma.array(a, mask=a[n]['name']=="foo") am['x'] +=

Re: [Numpy-discussion] Boolean indexing

Den 08.03.2011 14:59, skrev Sam Tygier: > I think the boolean slicing is making a copy instead of a view. Yes. > is there > a way around this? A "boolean slice" cannot be indexed with the dot product of dimensions and strides, hence the copy. You probably want to use masked arrays instead. St

Re: [Numpy-discussion] rewriting NumPy code in C or C++ or similar

Den 08.03.2011 05:05, skrev Dan Halbert: > Thanks, that's a good suggestion. I have not written Fortran since 1971, > but it's come a long way. I was a little worried about the row-major vs > column-major issue, but perhaps that can be handled just by remembering > to reverse the subscript order be

[Numpy-discussion] [ANN] Bento 0.0.5, a packaging solution for python software

Hi, I am pleased to announce a new release of bento, a packaging solution for python which aims at reproducibility, extensibility and simplicity. You can take a look at its main features on bento's main documentation page (http://cournape.github.com/Bento). The main features of this 0.0.5 rele

[Numpy-discussion] Boolean indexing

Hi I am having an issue with boolean slicing. it seems to work fine for reading a value, but I can use it to set a value: import numpy b = numpy.array([[1,2],[3,4],[5,6],[7,8],[9,10]]) m = numpy.array([0,1,0,0,0], dtype=bool) print b[m] print b[m][0,0] b[m][0,0] = -1 print b[m][0,0] I think th

Re: [Numpy-discussion] rewriting NumPy code in C or C++ or similar

On Mon, 7 Mar 2011 23:07:32 -0500, Dan Halbert wrote: > On 3/7/2011 9:25 PM, Nathaniel Smith wrote: > > On Mon, Mar 7, 2011 at 3:36 PM, Dan Halbert wrote: > >> Or is there some higher-level compiled array language that looks something > >> like NumPy code? > > > > You might want to try Eigen: >

Re: [Numpy-discussion] [cython-users] errors with numpy 1.5.1

Tue, 08 Mar 2011 00:12:35 -0800, Robert Bradshaw wrote: [clip] > Unfortunately, I don't think anyone's actively working on it right now. > It's probably a superficial fix for someone who knows NumPy and Cython > decently well, but I have no idea myself (not having looked that deeply > into it.) Doe

Re: [Numpy-discussion] [cython-users] errors with numpy 1.5.1

On Fri, Mar 4, 2011 at 4:19 PM, Choy wrote: > Hello -- > > I recently upgraded from python 2.6 -> 2.7, cython 0.13 -> 14.1, and > numpy 1.4.1 -> 1.5.1.  Unfortunately, one of my cython modules no > longer works. > > I tracked my bug and used the web to see that there is already a > ticket for exac