The choice of luminosity function filter alpha parameter and possibly a bug with consolidate.

[Hoptune]

Hi Eli,

I am running redMaPPer on SDSS DR8 photometric catalog to reproduce your catalog, while I find the clusters I got are far less than your result. A possible reason could be the alpha parameter I use for the luminosity filter, i.e.:

In Rykoff et al. 2014 (and Rykoff et al. 2016) you mentioned that you set alpha=1 so I set calib_lumfunc_alpha=1 in the begining. While I noticed that in your cal_example.yml you set calib_lumfunc_alpha=-1. And I do find using calib_lumfunc_alpha=-1 will make my results more similar with yours. So I'm wondering which one is correct? Or am I missing something that these two alpha doesn't mean the same thing?

The other thing is that during the cluster finding procedure the name of the final catalog for each pixel is ended with "?????_final_catalog.fit". While in line 80-83 of redmapperconsolidatetask.py (as shown below) it seems to be still searching for files ended with "?????_final.fit", which will cause an error when calling redmapper_consolidate_run.py.

# find the files
catfiles = sorted(glob.glob(os.path.join(self.config.outpath, '%s_*_?????_final.fit' % (self.config.outbase))))

self.config.logger.info("Found %d catalog files in %s" % (len(catfiles), self.config.outpath))

By the way, I slightly modified your redmapper_batch.py to make it compatible with PBS system. If you like I can send a pull request though I'm not sure whether my codes will work for your original purpose. :)

Thanks for your work and public codes which are really helpful!

Best regards,
Zhiwei

[erykoff]

Apologies for the confusion on the sign convention of alpha between different papers. It should definitely be -1.0 in the code which is the default.

Thanks for catching that error, and I'll patch it. I would definitely welcome a PR on PBS.