CosmoCoffee Forum Index CosmoCoffee

 
 FAQFAQ   SearchSearch  MemberlistSmartFeed   MemberlistMemberlist    RegisterRegister 
   ProfileProfile   Log inLog in 
Arxiv New Filter | Bookmarks & clubs | Arxiv ref/author:

[CosmoMC] Error with params.ini file - Planck likelihood not found
 
Post new topic   Reply to topic    CosmoCoffee Forum Index -> Computers and software
View previous topic :: View next topic  
Author Message
Mason Ng



Joined: 17 May 2017
Posts: 22
Affiliation: The University of Auckland

PostPosted: June 15 2017  Reply with quote

I have managed to successfully build CosmoMC and the Planck likelihood on the cluster environment. I have also run (separately)

Code:
mpirun -np 2 ./cosmomc test.ini
mpirun -np 2 ./cosmomc test_planck.ini


which both produce outputs that are as expected. I am now trying to obtain test chains through params_generic.ini and I am getting the following error in the job log:

Code:
Number of MPI processes:           2
 Error :params_generic.ini , include file not found: ./data/clik/hi_l/plik/plik_
 dx11dr2_HM_v18_TT.clik
 MpiStop:            0
 Error :params_generic.ini , include file not found: ./data/clik/hi_l/plik/plik_
 dx11dr2_HM_v18_TT.clik
application called MPI_Abort(MPI_COMM_WORLD, 1818192232) - process 0
In: PMI_Abort(1818192232, application called MPI_Abort(MPI_COMM_WORLD, 1818192232) - process 0)
 MpiStop:            1
application called MPI_Abort(MPI_COMM_WORLD, 1818192232) - process 1
In: PMI_Abort(1818192232, application called MPI_Abort(MPI_COMM_WORLD, 1818192232) - process 1)
slurmstepd: error: *** STEP 64118033.0 ON compute-a1-052 CANCELLED AT 2017-06-15T14:16:51 ***
srun: Job step aborted: Waiting up to 122 seconds for job step to finish.
srun: error: compute-a1-052: tasks 0-1: Killed


It is probably something to do with the params_generic.ini file - it is here:

Code:

#Sample parameters for cosmomc used as a generic sampler
#Write your likelihood function in calclike.f90

generic_mcmc=T
parameterization=generic

#Folder where files (chains, checkpoints, etc.) are stored
root_dir = chains/

#Root name for files produced
file_root = planck_best_fit

#action = 0, to MCMC, action=1, postprocess .data file
action = 0

#general settings
#Planck 2015 likelihoods
INCLUDE(./data/clik/hi_l/plik/plik_dx11dr2_HM_v18_TT.clik)
INCLUDE(./data/clik/low_l/commander/commander_rc2_v1.1_l2_29_B.clik)
#NOTE: I have tried both "INCLUDE" and "DEFAULT".
#You can either use numbered parameters, or, usually better, define a parameter name file as below
#num_theory_params = 2
#e.g. for 2D Gaussian
#param[param1] = 0 -20 20 1 1
#param[param2] = 0 -20 20 1 1

#alternative using parameter names from your generic_example.paramnames file
ParamNamesFile = ./batch2/params_CMB_defaults.ini
#param[mypar1] = 0 -20 20 1 1
#param[mypar2] = 0 -20 20 1 1

#Max samples to get
samples = 100000

#Use vanilla MCMC here since no speed hierarchy defined
sampling_method = 1
use_fast_slow = F

estimate_propose_matrix = F
delta_loglike = 2

propose_scale = 2.4

#Temperature at which to Monte-Carlo
temperature = 1

#Feedback level ( 2=lots,1=chatty,0=less,-1=minimal)
feedback = 1

#Can re-start from the last line of previous run (.txt file)
continue_from =

#Increase to oversample fast parameters,e.g. if space is odd shape
oversample_fast = 1

#Can use covariance matrix for proposal density, otherwise use settings below
#Covariance matrix can be produced using "getdist" prorgram.
propose_matrix =

#If action = 1
redo_likelihoods = T
redo_theory = F
redo_cls = F
redo_pk = F
redo_skip = 0
redo_outroot =
redo_thin = 1
#If large difference in log likelihoods may need to offset to give sensible weights
#for exp(difference in likelihoods)
redo_likeoffset =  0

#Number of distinct points to sample
#Every accepted point is included

#number of steps between independent samples
#if non-zero all info is dumped to file file_root.data
#if you change this probably have to change output routines in code too
indep_sample = 0

#number of samples to disgard at start
#May prefer to set to zero and remove later
burn_in = 0

#If zero set automatically
num_threads = 0

#MPI mode multi-chain options (recommended)
#MPI_Converge_Stop is a (variance of chain means)/(mean of variances) parameter that can be used to stop the chains
#Set to a negative number not to use this feature. Does not guarantee good accuracy of confidence limits.
MPI_Converge_Stop = 0.01
#if MPI_LearnPropose = T, the proposal density is continally updated from the covariance of samples so far (since burn in)
MPI_LearnPropose = T

#Can optionally also check for convergence of confidence limits (after MPI_Converge_Stop  reached)
MPI_Check_Limit_Converge = T

#if MPI_Check_Limit_Converge = T, give tail fraction to check (checks both tails):
MPI_Limit_Converge = 0.025
#And the permitted percentil chain variance in units of the standard deviation (small values v slow):
MPI_Limit_Converge_Err = 0.1
#which parameter's tails to check. If zero, check all parameters:
MPI_Limit_Param = 0

#If have covmat, R to reach before updating proposal density (increase if covmat likely to be poor)
#Only used if not varying new parameters that are fixed in covmat
MPI_Max_R_ProposeUpdate = 2
#As above, but used if varying new parameters that were fixed in covmat
MPI_Max_R_ProposeUpdateNew = 30


#if blank this is set from system clock
rand_seed =
Back to top
View user's profile  
Antony Lewis



Joined: 23 Sep 2004
Posts: 1274
Affiliation: University of Sussex

PostPosted: June 15 2017  Reply with quote

The provided params_generic (intended for doing on-cosmology runs) does not have any of those INCLUDE lines. See test_planck.ini for how to include Planck files and the readme for how to install them.
Back to top
View user's profile [ Hidden ] Visit poster's website
Mason Ng



Joined: 17 May 2017
Posts: 22
Affiliation: The University of Auckland

PostPosted: June 15 2017  Reply with quote

Antony Lewis wrote:
The provided params_generic (intended for doing on-cosmology runs) does not have any of those INCLUDE lines. See test_planck.ini for how to include Planck files and the readme for how to install them.


Thanks for the reply. I did try both "DEFAULT" and "INCLUDE" lines and they both result in the same error message.

I have seen test_planck.ini before as well, and I see that it uses the likelihood from batch2 (so with .ini). I was under the impression that I could do the same thing if I did

Code:
DEFAULT(./data/clik/hi_l/plik/plik_dx11dr2_HM_v18_TT.clik)


It seems like I had the wrong idea?
Back to top
View user's profile  
Antony Lewis



Joined: 23 Sep 2004
Posts: 1274
Affiliation: University of Sussex

PostPosted: June 15 2017  Reply with quote

[quote="Mason Ng"]
Antony Lewis wrote:

It seems like I had the wrong idea?

Yes
Back to top
View user's profile [ Hidden ] Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    CosmoCoffee Forum Index -> Computers and software All times are GMT + 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group. Sponsored by WordWeb online dictionary and dictionary software.