[Numpy-discussion] Bradford Cross wants to stay in touch on LinkedIn

LinkedIn Bradford Cross requested to add you as a connection on LinkedIn: -- Mike, I'd like to add you to my professional network on LinkedIn. - Bradford Cross Accept invitation from Bradford Cross http://www.linkedin.com/e/QUQsB3LtSZ2GEKV2lSD

Re: [Numpy-discussion] Building Numpy: could not read symbols Bad value error

On Sat, Jun 19, 2010 at 4:01 PM, Michael Green wrote: > Well I used the instructions found here > < > http://www.scipy.org/Installing_SciPy/Linux#head-eecf834fad12bf7a625752528547588a93f8263c > > > > I started by building lapack and it built successfully. > > I then proceeded with ATLAS using the

Re: [Numpy-discussion] Building Numpy: could not read symbols Bad value error

Well I used the instructions found here I started by building lapack and it built successfully. I then proceeded with ATLAS using the following configure line (that does contain -fPIC): ../configure -Fa al

Re: [Numpy-discussion] reduce array by computing min/max every n samples

Benjamin Root wrote: > Brad, I think you are doing it the right way, but I think what is > happening is that the reshape() call on the sliced array is forcing a > copy to be made first. The fact that the copy has to be made twice > just worsens the issue. I would save a copy of the reshape res

Re: [Numpy-discussion] reduce array by computing min/max every n samples

On Sat, Jun 19, 2010 at 15:37, Benjamin Root wrote: > On that note, would it be a bad idea to have a function that returns a > min/max tuple?  Performing two iterations to gather the min and the max > information versus a single iteration to gather both at the same time would > be useful. I have

Re: [Numpy-discussion] reduce array by computing min/max every n samples

Brad, I think you are doing it the right way, but I think what is happening is that the reshape() call on the sliced array is forcing a copy to be made first. The fact that the copy has to be made twice just worsens the issue. I would save a copy of the reshape result (it is usually a view of the

Re: [Numpy-discussion] Ticket #1455, the bias keyword.

On Jun 19, 2010, at 1:50 PM, Charles R Harris wrote: > Hi All, > > I'm looking at ticket #1455. At the moment I've added the ddof keyword to cov > and corrcoef and set it up so that it overrides the bias keyword when ddof is > something other than its current default of None. The question is,

Re: [Numpy-discussion] Beginer question about Boost Python Numeric

Hi, I think I figured out by using numeric::array::set_module_and_type( "numpy", "ndarray"); thanks! yong On Fri, Jun 18, 2010 at 8:17 PM, Yongning Zhu wrote: > Hi, > > I am new to boost python, thanks for the list, and hopefully, I will > find some one to help me. > I am adding the boost pytho

[Numpy-discussion] Ticket #1455, the bias keyword.

Hi All, I'm looking at ticket #1455 . At the moment I've added the ddof keyword to cov and corrcoef and set it up so that it overrides the bias keyword when ddof is something other than its current default of None. The question is, should I deprecate th

Re: [Numpy-discussion] ImportError: /usr/lib/python2.6/dist-packages/numpy/core/multiarray.so: undefined symbol: NPY_COPY_PYOBJECT_PTR

On Sat, Jun 19, 2010 at 11:13 PM, Sandro Tosi wrote: > Hello, > I'm upgrading the Debian package for Numpy to 1.4.1. > > Compilation goes well, but when I install the package for test and > import numpy I got this traceback: Do you have numpy already installed ? Because debian splits the numpy pa

[Numpy-discussion] ImportError: /usr/lib/python2.6/dist-packages/numpy/core/multiarray.so: undefined symbol: NPY_COPY_PYOBJECT_PTR

Hello, I'm upgrading the Debian package for Numpy to 1.4.1. Compilation goes well, but when I install the package for test and import numpy I got this traceback: $ python -V Python 2.6.5+ $ python -c "import numpy" Traceback (most recent call last): File "", line 1, in File "/usr/lib/python2