Mailman 3 January 2016 - yt-dev

PR Triage Hangout (Wednesday)
by Hilary Egan 08 Mar '16

08 Mar '16

Hi all, Here's the link to this week's PR triage hangout, at 12pm EST on Wednesday. https://plus.google.com/u/0/events/cpg5tqck8tuufar5nqe5oo790lg -- Hilary Egan Astrophysical and Planetary Science Department University of Colorado Boulder

1 5

POLL: A New Default Colormap
by Matthew Turk 01 Mar '16

01 Mar '16

Hi folks, The current default colormap that yt uses is called "Algae," and it's been with us for a while. Recently some of us on the yt-dev mailing list (which is open to join!) have begun to look at alternate colormaps that have better characteristics, particularly from an accessibility perspective. If you have the time, it would be really, *really* helpful if you could take this anonymous, two question (only one of which is mandatory!) poll about which of the four candidates we've come up with you like the best. http://goo.gl/forms/D4qQqkPIMq Thanks very much! The poll will be open at least a week, and I'll report back the results here sometime after that. -Matt

4 7

The future of yt
by Matthew Turk 28 Feb '16

28 Feb '16

Hi everyone, The last couple weeks I’ve been thinking a lot about the future of yt. What I’d like to propose is that we shift investment in yt as a single, monolithic codebase into yt as a project, or an ecosystem of projects. This came out of the discussion of the extension/affiliated packages, analysis modules, and so on. Britton has for a while been pitching the idea [which I will poorly paraphrase here] that yt can be the framework on top of which killer apps can be built. I think this is great. What’s holding us back in some ways from this is that yt is currently structured as a monolithic code base, with little to no discovery of other packages and apps and whatnot. We tried for a while to change this with The Barn, but it ended up not quite taking off. I think the time is right to try to change the way we think about yt to be more about yt the Project, rather than yt the Codebase; the core codebase is an important component of this, but not the whole of it. Encouraging an ecosystem of packages can have a few very important benefits: * External packages will confer greater individual credit to the folks who develop them. * External packages can be versioned and developed independently; the review process can be different. * yt’s core can be emphasized as a generic package, on top of which astronomy analysis can be built. * Packages can be maintained wherever, including alternate locations such as github. On the other hand, having packages inside the main distribution makes discoverability much, much easier. It also enables everything to be In The Box. And, the continuous integration and testing system is already set up for yt. But, these are all possible to overcome -- we can devise a strategy for adding packages to the CI system (and if they are externally managed, they can also rely on yt as a dependency and use whatever CI system they like!) and we can improve discoverability by refocusing the website to enable this. I've asked Kacper about adding new packages, and it's not as easy as it might seem, so we may need to be careful about how that process occurs; one possibility would be to provide servers and ready-made setups, but have individuals do the heavy lifting. We could even have something in the codebase that describes some packages that are available. External packages could have much looser dependency rules, which means they can be free to take advantage of things like OpenCL, numba, etc, without having to add them to the primary codebase. Synchronizing APIs and versions across extension packages may be difficult in some particular cases, but I suspect in practice will not be an issue, as long as we continue to have a reasonably stable *public* API, and graduate a few things (such as .blocks) into a public API from semi-private. To this end, of really encouraging an ecosystem of packages, I’d like to propose two things, in increasing order of disruptiveness. First: Encourage extension packages. This would mean: * Reorganize website to allow for extension packages to be displayed prominently * Add support for name-space packages in yt * (possible) split out some packages from analysis_modules, including halo finding * Codify process of extension package creation, including how to have CI set up for them and build system. The second, more disruptive proposal: * Split yt into subprojects. This would include spinning out the volume rendering and some or all of the frontends, and probably the testing infrastructure as well. * Split further astro-specific routines into an astro extension, and begin the process of doing this with other domains as well. (As in the long-simmering domain context YTEP.) I’ll invite comments from everyone, but particularly from folks who have either not contributed to an analysis module or extension package because of concerns that would be addressed by this, as well as from core developers this would impact. If the thread gets too unweildy we may also want to table this for the next yt team meeting. Thanks, Matt _______________________________________________ yt-dev mailing list yt-dev(a)lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

6 13

GSoC
by Matthew Turk 12 Feb '16

12 Feb '16

Hi folks, In the past we've explored doing Google Summer of Code, but without much success. I think this year the time is right, and we should get set up to do so. The timeline is here: https://developers.google.com/open-source/gsoc/timeline?hl=en We have until the 19th of February to prep our application; I don't think we should go in as a mentoring organization, but instead attempt to apply through a larger mentoring organization such as NumFOCUS. Is anyone willing to be a mentor? This would involve (virtual) checkins with students and spending a fair amount of time each week working with the student. It can be a gamble, but the program can be quite successful for both students and projects, and I personally think it's worth it. If you're interested, ping me off list in the next few days and let's start organizing. -Matt

2 1

Unit systems PR--needs some review
by John ZuHone 29 Jan '16

29 Jan '16

Hi devs, The unit system PR is still open for review: https://bitbucket.org/yt_analysis/yt/pull-requests/1904/switching-between-d… <https://bitbucket.org/yt_analysis/yt/pull-requests/1904/switching-between-d…> This is a big PR that touches in some way nearly every frontend and many other parts of the code, so I’d like to get about 5 approvals on it before we merge it in, especially from people that work on specific frontends. It has docs and tests, and there is a YTEP explaining the rationale behind it and what it does: https://bitbucket.org/yt_analysis/ytep/pull-requests/58/adding-ytep-0028-fo… <https://bitbucket.org/yt_analysis/ytep/pull-requests/58/adding-ytep-0028-fo…> This should go in before 3.3 is out. Best, John

2 2

Proposal: NumFOCUS Fiscal Sponsorship Agreement
by Matthew Turk 29 Jan '16

29 Jan '16

Hi everyone, The NumFOCUS organization is a 501(c)3 dedicated to both open source scientific software and computational science education ( numfocus.org ). I have recently been named to their board of directors. NumFOCUS supports many projects familiar to us here, including NumPy, Matplotlib, Jupyter, Astropy, and Sympy, as well as rOpenSci, Data Carpentry, Software Carpentry and Julia. After talking this over with Britton, I would like to propose that we join NumFOCUS through a comprehensive Fiscal Sponsorship Agreement. This process is outlined here: http://www.numfocus.org/apply-for-fiscal-sponsorship.html Organizations such as NumPy and rOpenSci have detailed their reasons for participating in this program here: https://mail.scipy.org/pipermail/numpy-discussion/2015-October/073926.html https://ropensci.org/blog/2014/10/01/numfocus-partnership/ Primarily, I think that this would help with our ability to exist independently of a single investigator; grants for programs such as workshops, project infrastructure, and so on can be managed by NumFOCUS (which has low overhead) and can be affiliated with the project. I think this is something that warrants discussion, and perhaps should be talked over in person during a team meeting, but I believe that this would be a strong step forward for us as a project and a community. -Matt

12 12

Issue #1167: The installation of rockstar in anaconda environment (yt_analysis/yt)
by Edward Leong 29 Jan '16

29 Jan '16

New issue 1167: The installation of rockstar in anaconda environment https://bitbucket.org/yt_analysis/yt/issues/1167/the-installation-of-rockst… Edward Leong: How to install the rockstar on yt in anaconda environment?

1 0

Issue #1166: YTFieldNotFound for angular momentum (yt_analysis/yt)
by Joyce Lee 28 Jan '16

28 Jan '16

New issue 1166: YTFieldNotFound for angular momentum https://bitbucket.org/yt_analysis/yt/issues/1166/ytfieldnotfound-for-angula… Joyce Lee: I'm working with 3D simulation data (ORION) to find angular momentum vector. http://use.yt/upload/f1926f08 http://use.yt/upload/272030ab http://use.yt/upload/5e496e7c I run the following: ``` #!python import yt file = "data.0200.3d.hdf5" ds = yt.load(file) sp = ds.sphere("center", (0.1, 'pc')) L = sp.quantities.angular_momentum_vector() ``` and get the following traceback ``` #!python /Users/joycelee/yt-x86_64/src/yt-hg/yt/data_objects/derived_quantities.pyc in __call__(self, *args, **kwargs) 66 storage = {} 67 for sto, ds in parallel_objects(chunks, -1, storage = storage): ---> 68 sto.result = self.process_chunk(ds, *args, **kwargs) 69 # Now storage will have everything, and will be done via pickling, so 70 # the units will be preserved. (Credit to Nathan for this /Users/joycelee/yt-x86_64/src/yt-hg/yt/data_objects/derived_quantities.pyc in process_chunk(self, data, use_gas, use_particles) 458 rvals.extend([(data["all", "particle_specific_angular_momentum_%s" % axis] * 459 data["all", "particle_mass"]).sum(dtype=np.float64) \ --> 460 for axis in "xyz"]) 461 rvals.append(data["all", "particle_mass"].sum(dtype=np.float64)) 462 return rvals /Users/joycelee/yt-x86_64/src/yt-hg/yt/data_objects/data_containers.pyc in __getitem__(self, key) 244 Returns a single field. Will add if necessary. 245 """ --> 246 f = self._determine_fields([key])[0] 247 if f not in self.field_data and key not in self.field_data: 248 if f in self._container_fields: /Users/joycelee/yt-x86_64/src/yt-hg/yt/data_objects/data_containers.pyc in _determine_fields(self, fields) 513 raise YTFieldNotParseable(field) 514 ftype, fname = field --> 515 finfo = self.ds._get_field_info(ftype, fname) 516 else: 517 fname = field /Users/joycelee/yt-x86_64/src/yt-hg/yt/data_objects/static_output.pyc in _get_field_info(self, ftype, fname) 541 self._last_finfo = self.field_info[(ftype, fname)] 542 return self._last_finfo --> 543 raise YTFieldNotFound((ftype, fname), self) 544 545 def _setup_classes(self): YTFieldNotFound: Could not find field '('all', 'particle_specific_angular_momentum_x')' in data.0200.3d.hdf5. ```

1 0

Issue #1165: angular_momentum_vector derived quantity fails for an ORION dataset (yt_analysis/yt)
by Nathan Goldbaum 28 Jan '16

28 Jan '16

New issue 1165: angular_momentum_vector derived quantity fails for an ORION dataset https://bitbucket.org/yt_analysis/yt/issues/1165/angular_momentum_vector-de… Nathan Goldbaum: This is a weird one. With the following dataset (spread over three files): http://use.yt/upload/5e496e7c http://use.yt/upload/99731ae6 http://use.yt/upload/ac7409e1 and the following test script: ``` #!python import yt file = "data.0100.3d.hdf5" ds = yt.load(file) sp = ds.sphere("center", (0.1, 'pc')) L = sp.quantities.angular_momentum_vector() ``` you will get the following error: ``` #!python Traceback (most recent call last): File "test.py", line 7, in <module> L = sp.quantities.angular_momentum_vector() File "/Users/goldbaum/Documents/yt-hg/yt/data_objects/derived_quantities.py", line 72, in __call__ values[i].append(storage[key][i]) IndexError: list index out of range ``` When I step through in a debugger, it seems that the `parallel_objects` inside of the `DerivedQuantity` base class is not properly combining data from the two chunks that get processed. More specifically, the `parallel_passthrough` decorator on the `par_combine_objects` method is only returning data from the first chunk, not combining data from both chunks. This is totally weird because it's very well tested, often-exercised code. I'm not terribly familiar with the internals of the parallel analysis interface and would appreciate another set of eyes here. @brittonsmith @MatthewTurk maybe?

1 0

Default colormap
by Matthew Turk 28 Jan '16

28 Jan '16

Hi folks, For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default. In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt: Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ I love algae, but it's not the best from an accessibility perspective. I'd like to propose that we use a new default colormap. If we do this, I see two options: * Retain a "branding" by developing a new one either by using the techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.) * Use viridis (which we may then have to ship if we have older versions of matplotlib to support) -Matt _______________________________________________ yt-dev mailing list yt-dev(a)lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

10 22