[yt-users] Column density plot

Nathan Goldbaum nathan12343 at gmail.com
Tue Nov 4 16:46:26 PST 2014


I've updated the pull request I created earlier for John Regan's issue to
include a fix for this, since they're related:
https://bitbucket.org/yt_analysis/yt/pull-request/1292/ensure-profiled-data-are-in-the-correct/diff

Would you mind giving it a test drive?

On Tue, Nov 4, 2014 at 1:39 PM, Nathan Goldbaum <nathan12343 at gmail.com>
wrote:

>
>
> On Tue, Nov 4, 2014 at 1:37 PM, Stephanie Tonnesen <stonnes at gmail.com>
> wrote:
>
>> I want the cylindrical_r bins to be in linear space.  When I do just the
>> first line, I still get everything in log space.  When I include
>> "logs=logs" in the profile call, it crashes.
>>
>> alld = ds.all_data()
>> gal = alld.cut_region("obj['density'] > 1.01e-28")
>> logs = {('index', 'cylindrical_r'): False}
>> #logs = {('gas', 'cell_mass'): False}
>> profile = yt.create_profile(gal,
>>                             [('index', 'cylindrical_r')],          # the
>> bin field
>>                             [('gas', 'cell_mass')],  # profile field
>>                             weight_field=None, n_bins=17,logs=logs)
>> coldens = profile["gas","cell_mass"]/1.6733e-24
>> coldens[0] = coldens[0]/(3.14159*profile.x[0]*profile.x[0])
>> i = 1
>> while i < len(coldens):
>>     coldens[i] = coldens[i]/((3.14159*profile.x[i]*profile.x[i]) -
>> (3.14159*profile.x[i-1]*profile.x[i-1]))
>>     i = i+1
>> mp.plot(profile.x/3.086e21,coldens)
>> mp.show()
>> print profile.x
>>
>> ---------------------------------------------------------------------------KeyError                                  Traceback (most recent call last)<ipython-input-23-a4ffd8615abf> in <module>()      6                             [('index', 'cylindrical_r')],          # the bin field      7                             [('gas', 'cell_mass')],  # profile field----> 8                             weight_field=None, n_bins=17,logs=logs)      9 coldens = profile["gas","cell_mass"]/1.6733e-24     10 coldens[0] = coldens[0]/(3.14159*profile.x[0]*profile.x[0])
>> /Users/stephanietonnesen/yt-x86_64/src/yt-hg/yt/data_objects/profiles.pyc in create_profile(data_source, bin_fields, fields, n_bins, extrema, logs, units, weight_field, accumulation, fractional)   1326                 for f in bin_fields]   1327     else:-> 1328         logs = [logs[bin_field[-1]] for bin_field in bin_fields]   1329     if extrema is None:   1330         ex = [data_source.quantities["Extrema"](f, non_zero=l)
>> KeyError: 'cylindrical_r'
>>
>>
> This looks like a bug.  Can you file an issue?  It would be great if you
> included a script exhibiting the behavior that makes use of one of the
> public datasets on yt-project.org/data
>
>
>>
>> --
>> Dr. Stephanie Tonnesen
>> Alvin E. Nashman Postdoctoral Fellow
>> Carnegie Observatories, Pasadena, CA
>> stonnes at gmail.com
>>
>> On Tue, Nov 4, 2014 at 1:31 PM, Nathan Goldbaum <nathan12343 at gmail.com>
>> wrote:
>>
>>>
>>>
>>> On Tue, Nov 4, 2014 at 1:29 PM, Stephanie Tonnesen <stonnes at gmail.com>
>>> wrote:
>>>
>>>> Okay, I think I must be missing something because right now I am still
>>>> getting log-spaced bins:
>>>>
>>>> alld = ds.all_data()
>>>> gal = alld.cut_region("obj['density'] > 1.01e-28")
>>>> logs = {('index', 'cylindrical_r'): False}
>>>> logs = {('gas', 'cell_mass'): False}
>>>>
>>>
>>> Do you want both of these to be linearly spaced?  Right now you're
>>> overwriting the first one when you redeclare "logs".
>>>
>>>
>>>> profile = yt.create_profile(gal,
>>>>                             [('index', 'cylindrical_r')],          #
>>>> the bin field
>>>>                             [('gas', 'cell_mass')],  # profile field
>>>>                             weight_field=None, n_bins=17)
>>>>
>>>
>>> Did you forget to pass logs in here?
>>>
>>>
>>>> coldens = profile["gas","cell_mass"]/1.6733e-24
>>>> coldens[0] = coldens[0]/(3.14159*profile.x[0]*profile.x[0])
>>>> i = 1
>>>> while i < len(coldens):
>>>>     coldens[i] = coldens[i]/((3.14159*profile.x[i]*profile.x[i]) -
>>>> (3.14159*profile.x[i-1]*profile.x[i-1]))
>>>>     i = i+1
>>>> mp.plot(profile.x/3.086e21,coldens)
>>>> mp.show()
>>>> print profile.x
>>>>
>>>> [  2.59949407e+20   3.49922639e+20   4.71037249e+20   6.34071835e+20
>>>>    8.53535665e+20   1.14895993e+21   1.54663592e+21   2.08195484e+21
>>>>    2.80255739e+21   3.77257363e+21   5.07833019e+21   6.83603292e+21
>>>>    9.20210863e+21   1.23871263e+22   1.66745367e+22   2.24458981e+22
>>>>    3.02148330e+22] cm
>>>>
>>>>
>>>>
>>>> --
>>>> Dr. Stephanie Tonnesen
>>>> Alvin E. Nashman Postdoctoral Fellow
>>>> Carnegie Observatories, Pasadena, CA
>>>> stonnes at gmail.com
>>>>
>>>> On Tue, Nov 4, 2014 at 12:32 PM, Nathan Goldbaum <nathan12343 at gmail.com
>>>> > wrote:
>>>>
>>>>>
>>>>>
>>>>> On Tue, Nov 4, 2014 at 12:26 PM, Stephanie Tonnesen <stonnes at gmail.com
>>>>> > wrote:
>>>>>
>>>>>> Okay, I've got that, but does that mean that I can't set something in
>>>>>> the parenthesis to force the bins to be in linear space?  and I need to do
>>>>>> the take_log setting earlier in the code?
>>>>>>
>>>>>
>>>>> You can, you would need to have something like:
>>>>>
>>>>> logs = {('gas', 'cell_mass'): False}
>>>>>
>>>>> profile = yt.create_profile(gal,
>>>>>                             [('index', 'cylindrical_r')],          #
>>>>> the bin field
>>>>>                             [('gas', 'cell_mass')],  # profile field
>>>>>                             weight_field=None, logs=logs)
>>>>>
>>>>>
>>>>>>
>>>>>> Thanks,
>>>>>> Stephanie
>>>>>>
>>>>>> --
>>>>>> Dr. Stephanie Tonnesen
>>>>>> Alvin E. Nashman Postdoctoral Fellow
>>>>>> Carnegie Observatories, Pasadena, CA
>>>>>> stonnes at gmail.com
>>>>>>
>>>>>> On Tue, Nov 4, 2014 at 11:10 AM, Nathan Goldbaum <
>>>>>> nathan12343 at gmail.com> wrote:
>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Nov 4, 2014 at 10:59 AM, Stephanie Tonnesen <
>>>>>>> stonnes at gmail.com> wrote:
>>>>>>>
>>>>>>>> Thank you both!  I ended up generally following Suoqing's method,
>>>>>>>> but tried to use create_profile.  I want my radial spacing to be linear,
>>>>>>>> and when I look at the help page:
>>>>>>>>
>>>>>>>> http://yt-project.org/docs/dev/reference/api/generated/yt.data_objects.profiles.create_profile.html#yt.data_objects.profiles.create_profile
>>>>>>>>
>>>>>>>> I read that I should set logs=False
>>>>>>>>
>>>>>>>> So I type in:
>>>>>>>>
>>>>>>>> profile = yt.create_profile(gal,
>>>>>>>>                             [('index', 'cylindrical_r')],
>>>>>>>> # the bin field
>>>>>>>>                             [('gas', 'cell_mass')],  # profile field
>>>>>>>>                             weight_field=None, logs=False)
>>>>>>>>
>>>>>>>> and yt does not like that.  Nor does it like take_log or
>>>>>>>> log_space.  Can someone tell me what I should be setting there?
>>>>>>>>
>>>>>>>>
>>>>>>> In this case, logs is a dictionary that maps field names to logging
>>>>>>> selection.  From the docstrings for create_profile:
>>>>>>>
>>>>>>> logs : dict of boolean values
>>>>>>>
>>>>>>>         Whether or not to log the bin_fields for the profiles.
>>>>>>>
>>>>>>>         The keys correspond to the field names. Defaults to the
>>>>>>> take_log
>>>>>>>         attribute of the field.
>>>>>>>
>>>>>>>
>>>>>>>> Thanks again!
>>>>>>>>
>>>>>>>> Stephanie
>>>>>>>>
>>>>>>>> --
>>>>>>>> Dr. Stephanie Tonnesen
>>>>>>>> Alvin E. Nashman Postdoctoral Fellow
>>>>>>>> Carnegie Observatories, Pasadena, CA
>>>>>>>> stonnes at gmail.com
>>>>>>>>
>>>>>>>> On Mon, Nov 3, 2014 at 9:45 PM, Suoqing JI <
>>>>>>>> suoqing at physics.ucsb.edu> wrote:
>>>>>>>>
>>>>>>>>> Hi Stephanie,
>>>>>>>>>
>>>>>>>>> profile.add_fields(‘cell_mass’, weight=None)
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Sorry I’ve missed the later part of my old script — after getting
>>>>>>>>> the profile[‘cell_mass’] it should be divided by the unit surface area. The
>>>>>>>>> following is the full code and it has been tested with AMR data:
>>>>>>>>>
>>>>>>>>> profile = BinnedProfile1D(mydisk, Nbin, 'cylindrical_r', rmin,
>>>>>>>>> rmax, log_space=True, lazy_reader=True, end_collect=False)
>>>>>>>>> profile.add_fields('cell_mass', weight=None)
>>>>>>>>>
>>>>>>>>> R = profile['cylindrical_r’]
>>>>>>>>> Sigma = profile[‘cell_mass']
>>>>>>>>>
>>>>>>>>> R_edge = np.logspace(np.log10(rmin), np.log10(rmax), num=Nbin+1)
>>>>>>>>> for i in range(0, Nbin): Sigma[i] = Sigma[i] /
>>>>>>>>> (np.pi*(R_edge[i+1]**2 - R_edge[i]**2.))
>>>>>>>>>
>>>>>>>>> And Nathan’s approach which takes the advantage of image buffer
>>>>>>>>> should also work.
>>>>>>>>>
>>>>>>>>> Best wishes,
>>>>>>>>> --
>>>>>>>>> Suoqing JI
>>>>>>>>> Ph.D Student
>>>>>>>>> Department of Physics
>>>>>>>>> University of California, Santa Barbara
>>>>>>>>> CA 93106, USA
>>>>>>>>>
>>>>>>>>> On Nov 3, 2014, at 8:40 PM, Nathan Goldbaum <nathan12343 at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Mon, Nov 3, 2014 at 5:29 PM, Suoqing JI <
>>>>>>>>> suoqing at physics.ucsb.edu> wrote:
>>>>>>>>>
>>>>>>>>>> Hi Stephanie,
>>>>>>>>>>
>>>>>>>>>> Maybe you could specify a disk, use BinnedProfile1D to create
>>>>>>>>>> bins along cylindrical radial ('cylindrical_r'), and do
>>>>>>>>>>
>>>>>>>>>> profile.add_fields(‘cell_mass’, weight=None)
>>>>>>>>>>
>>>>>>>>>> then you could plot profile[‘cell_mass’] vs.
>>>>>>>>>> profile['cylindrical_r’] and get the 1-D plot the surface density.
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> This will be a profile of the gas mass as a function of radius,
>>>>>>>>> but it's not quite a surface density profile.  That said, for an unweighted
>>>>>>>>> projection, I think it's the same up to a constant scaling factor.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Best wishes,
>>>>>>>>>> --
>>>>>>>>>> Suoqing JI
>>>>>>>>>> Ph.D Student
>>>>>>>>>> Department of Physics
>>>>>>>>>> University of California, Santa Barbara
>>>>>>>>>> CA 93106, USA
>>>>>>>>>>
>>>>>>>>>> On Nov 3, 2014, at 5:17 PM, Stephanie Tonnesen <stonnes at gmail.com>
>>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>> Hi yt-users,
>>>>>>>>>>
>>>>>>>>>> I would like to make a 1D plot of column density vs radius for a
>>>>>>>>>> disk (to compare with observations).  I can make a projectionplot, but want
>>>>>>>>>> something a bit more simple to look at.  I am using yt3.0.1--is there a
>>>>>>>>>> nice way to to this?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> The key is to use the numpy.digitize and numpy.bincount functions
>>>>>>>>> to find the histogram of the surface density as a function of radius.
>>>>>>>>> Here's an example:
>>>>>>>>>
>>>>>>>>> http://nbviewer.ipython.org/gist/ngoldbaum/af8e7f317efe8f115e8b
>>>>>>>>>
>>>>>>>>> This is a simplified version of what I've done for a project I'm
>>>>>>>>> working on right now, which involves making a ton of radial plots of
>>>>>>>>> projected quantities:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> https://bitbucket.org/ngoldbaum/galaxy_analysis/src/910f5a7e278247a36f25d62bdc478a7b5a7fe8ce/galanyl/galaxy_analyzer.py?at=default#cl-338
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Thanks!
>>>>>>>>>>
>>>>>>>>>> Stephanie
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> Dr. Stephanie Tonnesen
>>>>>>>>>> Alvin E. Nashman Postdoctoral Fellow
>>>>>>>>>> Carnegie Observatories, Pasadena, CA
>>>>>>>>>> stonnes at gmail.com
>>>>>>>>>>  _______________________________________________
>>>>>>>>>> yt-users mailing list
>>>>>>>>>> yt-users at lists.spacepope.org
>>>>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> yt-users mailing list
>>>>>>>>>> yt-users at lists.spacepope.org
>>>>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> yt-users mailing list
>>>>>>>>> yt-users at lists.spacepope.org
>>>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> yt-users mailing list
>>>>>>>>> yt-users at lists.spacepope.org
>>>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> yt-users mailing list
>>>>>>>> yt-users at lists.spacepope.org
>>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> yt-users mailing list
>>>>>>> yt-users at lists.spacepope.org
>>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> yt-users mailing list
>>>>>> yt-users at lists.spacepope.org
>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> yt-users mailing list
>>>>> yt-users at lists.spacepope.org
>>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>>
>>>>>
>>>>
>>>> _______________________________________________
>>>> yt-users mailing list
>>>> yt-users at lists.spacepope.org
>>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>>
>>>>
>>>
>>> _______________________________________________
>>> yt-users mailing list
>>> yt-users at lists.spacepope.org
>>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>>
>>>
>>
>> _______________________________________________
>> yt-users mailing list
>> yt-users at lists.spacepope.org
>> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.spacepope.org/pipermail/yt-users-spacepope.org/attachments/20141104/72b9e6bc/attachment.html>


More information about the yt-users mailing list