[LSST|dm-users 649] Re: List/Description of Available Algorithm Names

Jim Bosch jbosch at astro.princeton.edu
Tue May 12 17:41:25 PDT 2015


On Tue, May 12, 2015 at 8:26 PM, Will Dawson <will at dawsonresearch.com>
wrote:

> OK it looks like some of what I may want are in meas_modelfit,
> meas_extensions_shapeHSM, and meas_extesions_photometryKron repositories.
> It looks like meas_modelfit is installed with a common `eups disturb
> install last_apps`.
>
> — Are there eups distributions for the extensions or should I clone the
> github repositories and associate them with the loadLSST.bash some how?
>
>
Hmm, I'd have thought there were, but maybe they never got added to the
main build after we got them working again recently.

If you clone the GitHub repos, you should be able to set them up just by
doing "setup -r . -t <some-tag>" while in the cloned directory (where
some-tag is the tag you used to install the stack).  You can then run
"scons opt=3 install declare" to install it and declare it so you can set
it up by name and version later.

And if you have any trouble with those repos, Perry Gee (who I happen to
know is somewhat local for you) would probably be a good resource in
getting them working, as he was the one who did a lot of the work in
getting them working again recently.


> I also realized that I should be using lsst.meas.base instead of
> lsst.meas.algorithms and thus SingleFrameMeasurementConfig rather than
> SourceMeasurementConfig (a previous suggestion by Jim makes sense now). And
> on a related note I guess I should something like:
>
> measureConfig = SingleFrameMeasurementConfig()
> measureConfig.algorithms.names = [‘base_SdssShape’]
>
> where ‘base_SdssShape’ has been used rather than ‘shape.sdss’.
>
> — Is all of this correct?
>

Yup, all correct.


> — If so then what algorithms.names should I use if I want to specify
> meas_modelfit or one of the extensions? I don’t a plugins.py file in these
> repositories like I do in meas_base.
>
>
If you import those packages into Python, they'll register themselves in
the list of available algorithms, and then you can get the complete list
(and its configuration) by inspecting an instance of
SingleFrameMeasurementConfig (i.e. look in the .plugins dict).

Jim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.lsst.org/pipermail/dm-users/attachments/20150512/176873da/attachment.html>


More information about the dm-users mailing list