Implementations of nonlocal contributions to GR#86
Open
dirian wants to merge 20 commits intolesgourg:2.4from
Open
Implementations of nonlocal contributions to GR#86dirian wants to merge 20 commits intolesgourg:2.4from
dirian wants to merge 20 commits intolesgourg:2.4from
Conversation
* devel: updated version number to 2.0.3 late source cut approximation removed for t0 source for better Tphi, Ephi changed default value of 2 neglect_delta_k parameters for better Tphi, Ephi corrected trivial bug in neglect approx of vectors in parser: filename written in parser_init, instead of just before (new argument in parser_init) corrected little bug: in output transfer functions, delta)cdm and theta_cdm were weird untrack compiled python wrapper CPU corrected Quick bug fix for CPU
…s, matching the curvature paper * devel: (26 commits) version number updated to 2.0.4 removed lots of unused variables detected by compiler updated value of eisw_lisw_split_z split isw in lisw and eisw increased a bit def settings for hyperspherical and fixed small things robust cl_ref now, and better q_size_max estimate decreased slightly closed ref precision to avoid memory overflow fixed reference settings in closed case fixed cl_ref, but problem when using it in closed case corrected even even more errors in sw-isw decomposition corrected even more errors in sw-isw decomposition corrected further errors in sw-isw decomposition corrected bug in isw-sw decomposition ignore more files in python directory improved q-sampling and integration along q; added temperature decomposition (sw, isw...) index_q_closed defined, transfer_get_q_list() moved up revised k and q sampling. Works well but last step needed: trap+spli in closed in spectra add boundary value for communication improved and updated multipole cut approximation Tentative Wrap of CosmoHammer ...
* devel: (59 commits) updated external Pk files and codes for higher kmax commented the part calling bandpowers in spectra updated version number to 2.1.0 now fixed allocation issue in spectra fixed issue with allocations in spectra.c (for iso modes) changed sign of nc_lens transfer and set dNdz_analytic params like in classgal corrected bug when reading evolution file makefile updated for tar updated cpp wrapper added tests in test_classy.pyL isocurvature, large kmax corrected test codes, added test_nonlinear fixed memory leak in halofit Brought classy wrapper to new version of nonlinear put back function nonlinear_k_nl_at_z added a precision variable halofit_min_k_max include/transfer.h new nonlinear module producing pk_nl nut not yet cl_nl Added macros for writing output files consistently Perturbation variables can now be printed to files for user specified k-values. Added a protection to the pk function, and derived params to CosmoHammer ...
…tylistic details * devel: added version nunber 2.1.1 Improved testings wiht nose fixed memory leak that appeared only for isocurvature modes and no Pk Added feature: write thermodynamics file improved details of testing input values of w_fld improved style of test_loops fixed misplaced free in transfer_free Macro-ised the write background part. background_functions is now an internal function, never called eslewhere
…rror in baryon drag redshift introduced in 2.1.0 * devel: updated version number to 2.1.2 can ask for lensed polarisation without temperature removed obsolete warning for tensor polarisation fixed case r=0, and removed useless lines in background changed comment for halofit Update the model of Halofit used to match that of arXiv:1208.2701, which works better on small scales and incorporates a correction for dark energy. always write a # in headers fixed bug in clauclation of z_d when printing perturbation evolution, print also a corrected bug in print_variables: was printed only in last approx interval fixed bug appearing when tau_d is very small and z_d is above 10000 get warning if pass file which is neither .ini nor .pre q smapling robust in range -0.5+0.5, and omegak now limited to this range z values in recfast derived in more robust way, avoiding compiler-dependent rounding issues
* devel: Test class restored Update the test to test tensor modes replaced input parameter N_eff by N_ur (N_eff deprecated but still works) three options for tensor method now working almost done with neutrino tensors still in this process, tm_photons_only and tm_exact work, not tm_massless_approximation in precess of implementing massless neutrino approx in tensors, still buggy removed warning for neutrinos in tensors ncdm in tensors works ndm still wrong in course of impelmenting massive neutrinos in tensors; compiles, but not tested and not finsihed Added massless neutrino contribution to tensor modes
* devel: updated version number to 2.3.1 Corrected bugs related to dcdm Added support for omega_dcdm Moved dcdm parameters around inside background structure.
Warning: this is a cherry picked version of the devel branch (see the cherry picked commit). * 2.3.2: Cherry picked version of the devel branch (2.3.2) added the derived param r_0002 CPU.py now computes the ratios
* 2.4: (129 commits) fixed tests for density cl fixed default value of H_n for inflation_H fixed density_cl to use spectra and not lensing removed tabs Finetuning and bugfixing related to new output functionality transformed the README to rst syntax added tests and pushed version number inflation_V_end can now also deal with inflation ending up at imposed value phi_end spectrum.c all polished polished all primordial.c but find_phi_pivot() started to polish source/primordial.c polished inflation part in explanatory.ini polished inflation-related parameters in common.h and input.c polished primordial.h inflation_V_end works now. Just needs polishing inflation_V_end supposed to be coded, but bugs remain started to encode proper time in inflation find_pivot2() works fixed at least "evolve" with option _end_inflation_ making progress inflation_V_end ...
* 2.4: updated version number to 2.4.1 deleted some old reference output files updated the README.rst file changed default value of T_ncdm, added printout of N_eff total, and related remarks in explanatory.ini when asking for lensing but not lCl, return error tentative fix to CPU fixed -x option Bug-fix on CPU default h value adapted to take into account m=0.06eV in planck paper and m=0 in default model default values and explanatory values set to Planck2013+WP bestfit
* 2.4: restore output directory
* 2.4: version number updated to 2.4.2 Moved inflation tests from primordial module to input module and updated test_class.py accordingly. minor stylistic modifs of previous commit k_output_values are now added to actual k-list Fixed hardcoding in fix of issue lesgourg#19 Fixed a wrong allocation which would lead to segfaults on some 32 bit machines (if sizeof(double*)<sizeof(double) ) added growth factors D, in write background Fixed Github issue lesgourg#19 which was a bug for perturbations output removed deprecated classy function cleanup Fixed bug where the C-compiler defined in the Makefile would not be used by Cython when compiling the wrapper. Also avoids using --user when prefix have been set in distutils.cfg Fixed CPU.py, created script now works fixed compatibility with CosmoHammer changed the naming of output files from CPU.py modified default behaviour: if no field is asked, plot them all
…tion, dens-lens correlations) * 2.4: lensing potential in curved space: corrected a bug in the previous bug fix, and checked that everything OK now corrected the reading of the flag on the spot in input module Previous fix for longstanding issue fixed one bug but introduced another. This is the real fix. corrected something in reading annihilation parameters added few comments for recfast parameter definitions added few comments and tests to the DM annihilation part updated version number to 2.4.3 Corrected long-standing rare bug in ndf15 output all dens[i]-lens[j] correlations, not just i<j solved a bug: last_index_back and last_index_thermo not initialised in perturb_solve() updated hydrogen fraction like in Galli et al 2013 updated heating fraction using Galli et al 2013 corrected same exponents in annihilation part of hyrec corrected some flag and exponent typos in annihilation part Two remaining lines had same bug which could lead to seg-fault on 32 bit machines. corrected error in evolution function s impact on lensing added a comment concerning delta_m that needs to be generalised to ddm corrected lensing potential expression in cruved space corrected several minor typos affecting nCl but by negligible amount fixed missing curvature factor relevant only for nCl
* 2.4: change version number to 2.4.4 rewritten small part related to annihilation to avoid rare bug slightly revised the expression and sampling of the jumps in the reio_bins_tanh scheme, to make it more robust and avoid rare interpolation errors fixed two bugs concerning reio_bins_tanh fixed bug in total velocity transfer function (rho_cdm contribution in sync was missing) fixed bug when using hyrec + annihilation
…ecaying DM * 2.4: version number changed to v2.4.5 included dcdm in definition of fnu in halofit series of fixes to remove numerical errors from halofit implementation relevant only for quintessence: corrected an abs() into fabs() renoved selection bias from shear spectrum, only relevant for number count (thanks Aurelien Benoit-Levy) corrected bug in interface to halofit, problematic only with decaying CDM (Omega_drdcdm>0) improved reco-reio matching, helps to get more precise z_reio(tau_reio) search Improved error handling for shooting method not to generate input errors for MontePython
Owner
|
Hi Yves, I still hesitate between merging this to class_public, or leaving your pull request open, so that other users can import the changes from there. We should discuss at some point. |
Author
|
Hi Julien,
I think that it would be better to leave the pull request open so that everyone can have access to it, which was our initial goal.
I could also put an independent version on my git page, much as how you did for hi_class, and remove the pull request if you think that it is cleaner that way.
In any ways, we are currently on another nonlocal model that we would add to our branch if it happens to have a nice phenomenology, so the nonlocal modifications would be even larger and I guess that this motivates further the non-merge to class_public.
Best,
Yves
…________________________________
De : lesgourg <[email protected]>
Envoyé : samedi 25 mars 2017 13:04
À : lesgourg/class_public
Cc : Yves Dirian; Author
Objet : Re: [lesgourg/class_public] Implementations of nonlocal contributions to GR (#86)
Hi Yves, I still hesitate between merging this to class_public, or leaving your pull request open, so that other users can import the changes from there. We should discuss at some point.
-
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#86 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ANUTs5kg1-aBsK-8aUhozRQPKeE4Hsjeks5rpQK4gaJpZM4HnZDX>.
{"api_version":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name":"GitHub"},"entity":{"external_key":"github/lesgourg/class_public","title":"lesgourg/class_public","subtitle":"GitHub repository","main_image_url":"https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open in GitHub","url":"https://github.com/lesgourg/class_public"}},"updates":{"snippets":[{"icon":"PERSON","message":"@lesgourg in #86: Hi Yves, I still hesitate between merging this to class_public, or leaving your pull request open, so that other users can import the changes from there. We should discuss at some point."}],"action":{"name":"View Pull Request","url":"#86 (comment)"}}}
|
![https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open](https://cloud.githubusercontent.com/assets/143418/17495839/a5054eac-5d88-11e6-95fc-7290892c7bb5.png","avatar_image_url":"https://cloud.githubusercontent.com/assets/143418/15842166/7c72db34-2c0b-11e6-9aed-b52498112777.png","action":{"name":"Open){kind=link}
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two nonlocal modifications of Einstein's General Relativity are implemented into CLASSv2.4.3. The models are exposed in arXiv:1403.6068 (see also app.A of arXiv:1602.03558 for details on the implementation, and corresponding equations used).
For running CLASS including these modifications, the .ini file requires a model specification (see 9) of the explanatory).