Reproducing Fig 36 of Planck Collaboration VI 2018

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Felipe Maldonado
Posts: 26
Joined: November 10 2016
Affiliation: FSU

Reproducing Fig 36 of Planck Collaboration VI 2018

Post by Felipe Maldonado »

Dear Dr. Lewis,

I'm attempting to reproduce and later, modify, Figure 36 of Planck Collaboration VI (2018). I understand that this figure was produced with this code which is publicly available on github. I'm trying to do this because I've produced some Fisher forecasts for $N_{\mathrm{eff}}$ and I'd like to visualize and compare.

I cloned and installed CosmoMC (although I don't think it's necessary to do so), and put the directory in my path on my .bashrc file. However, when I attempt to run nnu_limits_decoupling.py, I get an error that I am guessing has to do with something in the pipeline not being found. The error message I get is the following:

Code: Select all

myname@mycomputer ~/Desktop/CosmoMC $ python batch3/outputs/nnu_limits_decoupling.py 
Match at T=0.794328234724, g=10.52749793104588 for match to g=10.685636748214073
max GeV 281.8382931264455
WARNING:root:setWithSubplotSize is deprecated, use set_with_subplot_size
WARNING:root:rcSizes is deprecated, use rc_sizes
WARNING:root:sampleAnalyser is deprecated, use sample_analyser
WARNING:root:samplesForRoot is deprecated, use samples_for_root
Traceback (most recent call last):
  File "batch3/outputs/nnu_limits_decoupling.py", line 178, in <module>
    samps = g.getSamples('nnu', dataroot)
  File "/home/felipe/Desktop/CosmoMC/python/planckStyle.py", line 195, in getSamples
    return self.sampleAnalyser.samplesForRoot(self.getRoot(paramtag, datatag))
  File "/home/felipe/Desktop/CosmoMC/python/planckStyle.py", line 191, in getRoot
    return self.getBatch().resolveName(paramtag, datatag, returnJobItem=returnJobItem)
AttributeError: 'NoneType' object has no attribute 'resolveName'
What am I doing wrong?
Thank you,
Felipe Maldonado
Antony Lewis
Posts: 1976
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: Reproducing Fig 36 of Planck Collaboration VI 2018

Post by Antony Lewis »

Sounds like it can't find the chain files. Make sure you downloaded the relevant chain grid from the PLA and set up the config as described at

https://cosmologist.info/cosmomc/readme_planck.html
Antony Lewis
Posts: 1976
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: Reproducing Fig 36 of Planck Collaboration VI 2018

Post by Antony Lewis »

Btw, I updated the script for py3 compatibility and made a more helpful error message on github master.
Felipe Maldonado
Posts: 26
Joined: November 10 2016
Affiliation: FSU

Re: Reproducing Fig 36 of Planck Collaboration VI 2018

Post by Felipe Maldonado »

Thank you very much.

-Felipe
Post Reply