| Re: [eigen] Vector generators |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/eigen Archives
]
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] Vector generators
- From: Hauke Heibel <hauke.heibel@xxxxxxxxxxxxxx>
- Date: Sun, 25 Apr 2010 09:19:51 +0200
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=PQcSjxZR1wEQzJyVmE8mKriGFBPqJsM7ltcUASUOmRM=; b=DQYiXw8wkw0OG/A263nEytbHiN7hyGLvcbFb5ovXVouD7r1g35wbW6lRDrf0WpB6+1 ucEg/muVPMyJjzk+J3/12sOrwzN93uB//+Xq82JuwYnRo7VlNq0R9Va74jecHHNqVL7H K/+EPXLHtPsLkiDeI/ldbxYD09y/T/CF8F4ug=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; b=ALxGK2va1Wr0eCRw8NHJbGDz8D6KEdj4MpdAo4LyRcAmlbX03wX64mCUpX8XPpRQvV 2zTLF2X3/fu7clYgmX6oMCQHrw6vwytEFFHmoui3DrAZl3XjH5UCVh+bvA+Ixq8OZjGF ms0lYBRCAkATg7aFgSslel//1DHDepTDUWux8=
We have some functions which might be offering what you are looking
for in the development branch:
1) setLinSpaced(const Scalar &low, const Scalar &high, int size), see
here http://tinyurl.com/32lo4fr
2) LinSpaced (const Scalar &low, const Scalar &high, int size),
http://tinyurl.com/2uqfjhj
3) LinSpaced (Sequential_t, const Scalar &low, const Scalar &high,
int size), http://tinyurl.com/36zjs6g
The third method is the fastes since it is setting up to 4 values in
one step though there is a slight approximation error involved due to
successive summation.
All methods have in common, that you may (actually, will most likely)
end up with your last element being
high + small error
which is caused by typical rounding errors.
Regards,
Hauke
On Sat, Apr 24, 2010 at 8:48 PM, Manoj Rajagopalan <rmanoj@xxxxxxxxx> wrote:
> Hi,
>
> I frequently generate mesh-data of the form f((1:n)*factor), where n is the
> number of elements in the mesh, factor is some multiplicative factor
> (mesh-spacing), and f() is some function that operates on each of the
> generated values. Is something like this possible with concise syntax in
> eigen?
>
> The quantity (1:n) can be thought of as an extension of the constant-vector
> concept. A further extension is the vector generated by (start,finish,delta).
> These vectors, like eigen's constant-vectors, can be concisely encoded and
> expanded at assignment-time.
>
> Thanks,
> Manoj
>
>
>