[AstroPy] Summer scholar

Rene Breton superluminique@gmail....
Fri Jul 30 12:36:43 CDT 2010

Hi Joe,

That's an interesting suggestion and I just searched a bit and came 
across a scipy package called PyWavelets! Have you ever heard about it? 
I've never used it myself since last time I had to use wavelets stuff 
was a while ago, well before this came to existence.

It certainly seems to me that working on a PyWavelets sub-package as 
part of the AstroPy library would be a useless duplicate of effort unless...

I'd look into the next couple things:
1. Does PyWavelets have all the functionalities that one (you) is 
looking for? If so, problem solved. If not, maybe it would be a good 
idea to implement what's missing in that package in order to keeps 
things well organized, especially if the things you have in mind are 
general enough and not too specific to astronomers.
2. If needed, we could think about a specialized wavelets sub-package in 
AstroPy with that suits some very specific need of astronomers (not sure 
what it would be though). Again, to me the "requirement" would be to 
make it be some kind of front-end to PyWavelets. By doing things this 
way, we don't include further dependencies, which is certainly 
fundamental to AstroPy. If one wants to use the wavelets sub-package, 
then one can make sure that PyWavelets is working and that's all. But 
again, I think solution #1 would be better because it would avoid 
possible strong divergence in the future.


On 10-07-30 01:19 PM, Joe Harrington wrote:
> Unless it's been done, I'd love to see this IDL/Matlab/FORTRAN code
> http://paos.colorado.edu/research/wavelets/
> http://paos.colorado.edu/research/wavelets/bams_79_01_0061.pdf
> implemented in Python.  *Lots* of people use it, and the lack of it is
> a factor in some people's decision to stick with IDL.  There are a few
> Python wavelet codes and I haven't looked at any of them, so maybe all
> that needs to be done is to implement the high-level stuff there
> (tests and examples) in an existing Python wavelet code.  An
> assessment of those and their suitability for tasks as outlined in the
> accompanying paper would be a great contribution for 8 weeks of
> undergrad work!  I used this package, e.g., in my most recent ApJ
> paper: http://arxiv.org/abs/1005.3570
> See especially the nearly-noiseless global transform!
> --jh--
> Date: Fri, 30 Jul 2010 14:16:32 +1000
> From: Wolfgang Kerzendorf<wkerzendorf@googlemail.com>
> Subject: [AstroPy] Summer scholar
> To: astropy@scipy.org
> Message-ID:<4C525220.1080508@gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>    Dear all,
> Here at Mt. Stromlo Observatory we will have summer scholars in November
> - February. There was a recent call for ideas for ~8 week summer
> projects. I can definitley co-supervise as this will all happen at
> Stromlo, but would need help with fleshing out an idea. Is there
> something we can offer for a physics undergrad to code for 8 weeks.
> Maybe build a coord base class? Is that too complex? Other ideas? Maybe
> we should also think about getting people in the next GSoC round?
> What are people's feeling on this?
> Cheers
>     Wolfgang
> _______________________________________________
> AstroPy mailing list
> AstroPy@scipy.org
> http://mail.scipy.org/mailman/listinfo/astropy

More information about the AstroPy mailing list