Numpy wraps the FORTRAN libs.  I'm certain this is what she should try:

http://numpy.scipy.org/

Bet your friend would find it groovy too.  Let Python rise to the
occasion, like a Loch Ness from the sea (lake, whatever).

Kirby

On Thu, Oct 20, 2011 at 1:55 PM, Chris McDonald <[email protected]> wrote:
> I have a friend that would very likely be interested in this as well.
> He does a heavy amount of scientific computing but most of it is in
> Fortran, he is looking to move to Python eventually though and I think
> this would be ideal for him. I on the other hand have lots of python
> experience but non in the big dataset/scientific world but I can help
> in the same way offered by Andrew, a web developer who knows how to
> distribute tasks using queues and databases.
>
> I'll ping him with regards to this and see what he thinks.
>
> -Wraithan
>
> On Thu, Oct 20, 2011 at 1:46 PM, Andrew Brookins <[email protected]> 
> wrote:
>> Heather (and everyone),
>>
>> On Oct 14, 2011, at 8:22 PM, Heather Lintz wrote:
>>
>>> Hi Pythonistas,
>>>
>>> I am a Corvallis ghost member of your group. My name is Heather. I am also
>>> an ecologist working on climate change topics. I program all the time in
>>> MATLAB, and less often in R and Python. So far, I have only used Python to
>>> do some ArcGIS tasks using the ArcGIS library in Python (and some other
>>> basic libraries too). However, I now have a couple somewhat hefty new
>>> projects I would like to accomplish in Python. I was wondering if there is a
>>> good time/place to catch some of you and talk about some potential
>>> Python tutoring with these tasks in mind. I already have some experience
>>> with the language (for example, I posted some of my code below that I wrote
>>> awhile and forgot about). Is the monthly meet-up a good place for this?
>>> You seem to have agendas for those meetings perhaps?
>>>
>>> Here are the projects I have in mind that I would like to work on:
>>>
>>> 1. Code a statistical algorithm and divide and delegate computation tasks to
>>> multiple processors on a Linux system. The processors would each generate
>>> results and the results would be pooled for an optimization.
>>>
>>> 2. Import RNA Seq data generated from the Illumina High Seq 2000 and learn
>>> how to manipulate INSANELY large bioinformatics/genomics data sets. I
>>> especially like
>>> to do statistics on such data (things that I normally do in MATLAB).
>>> But this time it would be treating the INSANELY LARGE AMOUNT of data as a
>>> matrix to manipulate it, etc. in Python.
>>>
>>> I'd like to come up twice a month for Python 'tutoring' to get these
>>> projects accomplished and learn Python better. There's nothing like wisdom
>>> from other programmers to help. Would this interest any of you? Can you
>>> recommend someone in your group that is great at
>>> scientific-python-programming-teaching challenges?
>>>
>>> Many thanks,
>>> Heather
>>>
>>>
>>> P.s. Here's my previous dinky Python code that I already forgot about. It's
>>> the max of my ability.
>>>
>>> ########################
>>> # Import system modules
>>> ########################
>>>
>>> import sys, string, os, arcgisscripting, copy, glob, linecache, csv from
>>> quantile import quantile
>>>
>>> # Create the Geoprocessor
>>> gp = arcgisscripting.create()
>>> gp.overwriteoutput = 1
>>>
>>> ####################################################################
>>> #READ DATA FROM EACH ASC FILE AND CALCULATE QUANTILES FROM EACH FILE
>>> ####################################################################
>>>
>>> q1=[]
>>> q2=[]
>>> q3=[]
>>>
>>> os.chdir(ascDIR)
>>> runlist=os.listdir(ascDIR)
>>> print repr(runlist)
>>> print len(runlist)
>>> for file in runlist:
>>>   print repr(file)
>>>   gq=[]
>>>   x=open(file,'r')
>>>   for i in xrange(6):
>>>       x.readline()
>>>   z= x.readline()
>>>   while z != '':
>>>       z=z.strip().split()
>>>       for num in z:
>>>           num=float(num)
>>>           if num > -1:
>>>               gq.append(num)
>>>       z= x.readline()
>>>   a=quantile(gq, .25,  qtype = 7, issorted = False)
>>>   #print a
>>>   b=quantile(gq, .5,  qtype = 7, issorted = False)
>>>   c=quantile(gq, .75,  qtype = 7, issorted = False)
>>>   q1.append(a)
>>>   q2.append(b)
>>>   q3.append(c)
>>> print len(q1), len(q2), len(q3)
>>>
>>> outfile = open("outfile.txt", "w")
>>> for i in xrange(len(q1)):
>>>   outfile.write("%12.3e%12.3e%12.3e\n" % (q1[i], q2[i], q3[i]))
>>> outfile.close()
>>>
>>> outfile = open("outfilezones.txt", "w")
>>> for i in xrange(len(q1)):
>>>   outfile.write(runlist)
>>> outfile.close()
>>>
>>>
>>>
>>> _______________________________________________
>>> Portland mailing list
>>> [email protected]
>>> http://mail.python.org/mailman/listinfo/portland
>>
>> This is a really awesome question.
>>
>> There are a few hack-nights around town that aren't oriented around a 
>> specific language that you could attend. Check: http://calagator.org/
>>
>> I'm not aware of a regular Python tutoring/workshop style meeting (maybe 
>> someone else is). However, I would be very interested in participating in 
>> and helping to organize such a meeting.
>>
>> As for your particular problem, I haven't done any scientific computing, but 
>> I can share my experience using Python for web development. Maybe there is 
>> some technology crossover (message queues, or even just your database layer? 
>> Oops, I dunno, I'm not a scientist!).
>>
>> Anyway, "large bioinformatics/genomics data"? Sounds awesome!
>>
>> Best,
>> Andrew
>> _______________________________________________
>> Portland mailing list
>> [email protected]
>> http://mail.python.org/mailman/listinfo/portland
>>
> _______________________________________________
> Portland mailing list
> [email protected]
> http://mail.python.org/mailman/listinfo/portland
>
_______________________________________________
Portland mailing list
[email protected]
http://mail.python.org/mailman/listinfo/portland

Reply via email to