Re: [Matplotlib-users] Analog of processing map() or protovis scale?

Uri Laserson, on 2011-01-16 17:41,  wrote:
> Hi all,
> 
> Does there already exist some python implementation (in MPL or other) of an
> easy-to-use 1D scale transformation?  This is something analogous to
> processing's map function or protovis's scale functionality.  It would work
> something like:
> 
> s = linear().domain(5,100).range(13000,15000)
> 
> or
> 
> s = root(p=5).domain(0.1,0.6).range(0,1)
> 
> There could be multiple versions, including linear, log, symlog, root
> (power), discrete, etc.

Hi Uri,

I think that the closest we have matplotlib is
matplotlib.colors.Normalize[1] and matplotlib.colors.LogNorm[2], but
both of these have a fixed range of the 0-1 (which is the reason
they are in colors). Both of these do end up with an inverse
method that you could leverage to get an arbitrary range, though.

1. http://matplotlib.sourceforge.net/api/colors_api.html#matplotlib.colors.Normalize
2. http://matplotlib.sourceforge.net/api/colors_api.html#matplotlib.colors.LogNorm

-- 
Paul Ivanov
314 address only used for lists,  off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7