| c312f4e | Mohammad Akhlaghi | 28 November 2020, 22:18:36 UTC | README-hacking.md: updated paper to cite for using Maneage Until now, we were asking the users of Maneage to cite the first paper that used its primoridal version (arXiv:1505:01664). But there is now a paper that fully describes the concept (arXiv:2006.03018). With this commit, in the 'citation' section of 'README-hacking.md' we now ask to cite the new paper. | 28 November 2020, 22:18:36 UTC |
| b4a273b | Boud Roukema | 28 November 2020, 17:58:33 UTC | Another try to fix modern_revolver_hacks.patch | 28 November 2020, 17:58:33 UTC |
| 87d6d0f | Boud Roukema | 28 November 2020, 17:44:37 UTC | Try to fix modern_revolver_hacks.patch This commit tries to fix modern_revolver_hacks.patch. | 28 November 2020, 17:44:37 UTC |
| fbab1ed | Boud Roukema | 28 November 2020, 17:15:11 UTC | Try to fix bary/geom confusion Unless we want to completely revise revolver for the confusing terminology "barycenter" that refers to geometrical centres, we can at least write "geom" for the files used in this particular project. This commit aims to do that. This commit also uses a loop for run-plots instead of repeated code. This commit has not yet been tested. | 28 November 2020, 17:15:11 UTC |
| 789d849 | Boud Roukema | 28 November 2020, 03:20:06 UTC | Merge branch 'prereferee_changes' into postreferee_corrections This commit merges the elaphrocentre definition improvement and the extra Nadathur reference. | 28 November 2020, 03:20:06 UTC |
| 80c5aa9 | Boud Roukema | 28 November 2020, 03:01:11 UTC | Highlighting changes done so far This commit does some further modification of the small number of changes made to the text so far and adds highlighting with \postrefereechanges{ ... } to highlight the changes for the referee + editor, who are not expected to do git diffs. | 28 November 2020, 03:01:11 UTC |
| 43c51f2 | Marius Peper | 16 November 2020, 09:48:19 UTC | Start on trivial corrections and add comments where we have to modify the text This commit starts to solve a few of the trivial corrections. It also starts to mark which parts of the text need to be overworked from my point of view. | 16 November 2020, 09:48:19 UTC |
| fcc77a7 | Mohammad Akhlaghi | 15 November 2020, 13:28:57 UTC | Building final PDF: pdf-build-final has to be given an explicit yes Until now, when the 'pdf-build-final' configuration variable (defined in 'reproduce/analysis/config/pdf-build.conf') was given any string a PDF would be built. This was very confusing, because people could put a 'no' and the PDF would still be built! With this commit, only when this variable has a value of 'yes' will the PDF be built. If given any other string (or no string at all), it will not produce a PDF. This issue was reported by Zahra Sharbaf. | 15 November 2020, 13:32:57 UTC |
| 9cfb8d3 | Mohammad Akhlaghi | 13 November 2020, 19:11:18 UTC | README.md: added commands to delete all Docker images Until now we had described the basic commands on how to create and use Docker images, but we hadn't mentioned how you can delete them. With this commit the commands necessary for deleting Docker images have also been added at the bottom of the section on Docker. | 13 November 2020, 19:11:18 UTC |
| 07ca2b7 | Marius Peper | 12 November 2020, 02:18:51 UTC | Enable the pipeline to perform allresults_LSBGs.py for all three centres in one run This commit lets allresults_LSBGs.py run on all three output-files of revolver. It will do the analysis for all three centres. A new version of the revolver hack is needed hence before we have not created a file containing the IDs of the DM particles laying in a fixed sphere around the barycentre. This commit directly deals with point 7 of the referees report. | 12 November 2020, 02:18:51 UTC |
| 2b39a67 | Mohammad Akhlaghi | 18 October 2020, 13:37:16 UTC | Recipes for final initialize and verify targets not on stdout The LaTeX macro files for these two subMakefiles are created on every run of './project make'. So their commands are also printed every time and hardly ever will a normal user want to modify or change these. So to avoid populating the standard output of a Maneaged project with all these extra lines every time (possibly getting mixed with the important analysis or LaTeX outputs), an '@' has been placed at the start of the recipes. With an '@' at the start of the recipe, Make is instructed to not print the commands it wants to run in the standard output. | 18 October 2020, 13:37:16 UTC |
| 3be9c0b | Boud Roukema | 11 October 2020, 11:01:22 UTC | Add missing ref Nadathur19BOSSRevolver for Revolver | 11 October 2020, 11:01:22 UTC |
| c473a5e | Boud Roukema | 11 October 2020, 10:45:06 UTC | Clarify definition of elaphrocentre This commit clarifies in the text that the definition of the elaphrocentre uses potentials calculated per particle by RAMSES. | 11 October 2020, 10:45:06 UTC |
| 04bda5a | Boud Roukema | 09 October 2020, 10:33:58 UTC | Update README-hacking.md with elaphrocentre ArXiv:2010.03742 This commit updates README-hacking.md with the URIs for the 'elaphrocentre' galaxy formation pipeline paper arXiv:2010.03742. This makes three papers currently in the peer review pipeline: arXiv:2006.03018, arXiv:2007.11779, and arXiv:2010.03742, each chronologically corresponding to various stages of the review process. | 09 October 2020, 13:28:21 UTC |
| 540f29b | Boud Roukema | 09 October 2020, 08:38:31 UTC | Restore parameters for N = 32^3 | 09 October 2020, 08:38:31 UTC |
| 19839a6 | Boud Roukema | 09 October 2020, 07:07:18 UTC | Update README.md with ArXiv and SWH IDs This commit updates README.md with the ArXiv and SWH IDs of the paper. The source URLs of the references are placed close to the citations, since some people will read README.md as a plain text file (and may not scroll all the way down), without making the paragraph itself too messy to read in plain text. The markdown rendered version should, in principle, not show the list of references; it should only show the URLs as live links. | 09 October 2020, 07:07:18 UTC |
| 6d86ff9 | Boud Roukema | 08 October 2020, 04:14:30 UTC | cleaner arXiv version | 08 October 2020, 04:14:30 UTC |
| cd3bd7d | Boud Roukema | 08 October 2020, 04:11:15 UTC | Data Availability Statement MNRAS new requirement | 08 October 2020, 04:11:15 UTC |
| 724a7c8 | Boud Roukema | 08 October 2020, 00:40:55 UTC | Add softwareheritage (swh) ID and URL | 08 October 2020, 00:40:55 UTC |
| 5621d99 | Boud Roukema | 07 October 2020, 22:22:48 UTC | Fix my error in terms of packages; minor fix in title | 07 October 2020, 22:22:48 UTC |
| 5665f48 | Boud Roukema | 07 October 2020, 22:12:09 UTC | Minor smoothing of log file instructions in README.md | 07 October 2020, 22:12:09 UTC |
| 2b097c5 | Boud Roukema | 07 October 2020, 22:05:17 UTC | Merge branch 'N128cubed' of codeberg:boud/elaphrocentre into N128cubed | 07 October 2020, 22:05:17 UTC |
| 6bbc299 | Marius Peper | 07 October 2020, 22:02:55 UTC | Minor edit of README.md This commit recommends the user to create logs when running/configuring the pipeline. | 07 October 2020, 22:02:55 UTC |
| b3f433d | Boud Roukema | 07 October 2020, 22:01:27 UTC | Fix make dist-arXiv for this paper With this commit, the command `./project make dist-arXiv` or `./project make dist-arxiv` should correctly create a tar.gz file for ArXiv. However, if you have old files in `.build/tex/build/` or `.build/tex/build/fig`, they may be included too. So double-check and clean those out first, or tidy up the tar.gz file, or improve the `make` rule. | 07 October 2020, 22:01:27 UTC |
| 7e5a841 | Boud Roukema | 07 October 2020, 21:27:25 UTC | Update README.md - paper name; elaphrocentre specific info This commit adds some updates to README.md - paper name, Zenodo ID; info specific to the elaphrocentre galaxy pipeline project; overall directory structure (which is not at all obvious to people new to Maneage). | 07 October 2020, 21:27:25 UTC |
| a8bccd5 | Boud Roukema | 07 October 2020, 20:44:19 UTC | Explain strong tail of insignificant values of GalsizeFracSlope | 07 October 2020, 20:44:19 UTC |
| 81e4878 | Boud Roukema | 07 October 2020, 20:05:53 UTC | Add cos variance to infall tau slope The cosmic variance for the infall tau slope is high enough that we need to state it. This commit calculates and states this value. | 07 October 2020, 20:05:53 UTC |
| f95ad34 | Boud Roukema | 06 October 2020, 19:06:36 UTC | Merge branch 'elaphro' into N128cubed | 06 October 2020, 19:06:36 UTC |
| c6b5713 | Boud Roukema | 06 October 2020, 14:00:32 UTC | Minor text smoothing This commit removes the "A closer examination ..." sentence and does some other minor text smoothing. | 06 October 2020, 19:05:41 UTC |
| 374c6b8 | Marius Peper | 06 October 2020, 11:28:21 UTC | Fix a small typo in the text | 06 October 2020, 19:05:34 UTC |
| 42ed3ef | Boud Roukema | 05 October 2020, 18:09:47 UTC | Text smoothing: 3. results, 4. discussion, 5. conclusion Minor text smoothing; some corrections of wrong summaries/inferences or somewhat misleading or confusing text. Trivia: two cases of LSGB corrected to LSBG; MNRAS wants 3-author papers to be cited with '*' the first time, to list all three authors (if I remember correctly). | 06 October 2020, 19:05:26 UTC |
| cf699db | Boud Roukema | 05 October 2020, 12:16:46 UTC | Text smoothing: abstract, 1. intro, 2. method Several small changes in the abstract, intro and method. Please double-check. :) | 06 October 2020, 19:05:17 UTC |
| 8e54b02 | Boud Roukema | 02 October 2020, 15:43:26 UTC | Update TODO: non-pegase sage hacks reported upstream The TODO point of reporting sage hacks upstreams is removed in this commit after reporting the -fcommon issue: https://github.com/darrencroton/sage/issues/16 = https://archive.today/lCjWw and proposing a pull request for the sprintf risk of a buffer exploit: https://github.com/darrencroton/sage/pull/17 = https://archive.today/OnDnX . | 06 October 2020, 19:05:10 UTC |
| 57d300e | Boud Roukema | 06 October 2020, 14:00:32 UTC | Minor text smoothing This commit removes the "A closer examination ..." sentence and does some other minor text smoothing. | 06 October 2020, 14:00:32 UTC |
| c098dfa | Marius Peper | 06 October 2020, 11:28:21 UTC | Fix a small typo in the text | 06 October 2020, 11:28:21 UTC |
| 2e425af | Boud Roukema | 05 October 2020, 18:09:47 UTC | Text smoothing: 3. results, 4. discussion, 5. conclusion Minor text smoothing; some corrections of wrong summaries/inferences or somewhat misleading or confusing text. Trivia: two cases of LSGB corrected to LSBG; MNRAS wants 3-author papers to be cited with '*' the first time, to list all three authors (if I remember correctly). | 05 October 2020, 18:09:47 UTC |
| eb293e3 | Boud Roukema | 05 October 2020, 12:16:46 UTC | Text smoothing: abstract, 1. intro, 2. method Several small changes in the abstract, intro and method. Please double-check. :) | 05 October 2020, 12:16:46 UTC |
| bba2584 | Boud Roukema | 02 October 2020, 15:43:26 UTC | Update TODO: non-pegase sage hacks reported upstream The TODO point of reporting sage hacks upstreams is removed in this commit after reporting the -fcommon issue: https://github.com/darrencroton/sage/issues/16 = https://archive.today/lCjWw and proposing a pull request for the sprintf risk of a buffer exploit: https://github.com/darrencroton/sage/pull/17 = https://archive.today/OnDnX . | 02 October 2020, 15:43:26 UTC |
| 2e4f399 | Boud Roukema | 02 October 2020, 11:32:03 UTC | Text: minor smoothing; Zenodo table ID; Fig 15 table link In this commit: minor tidying of some of Marius' recent text fixes; sentence length reduction and some minor improvements in logical flow. The zenodo ID 4062461 has been added (not to be confused with the zenodo ID of my CC-BY-SA-4.0 Cosmology lecture: https://zenodo.org/record/4060261 ). | 02 October 2020, 12:02:44 UTC |
| d3a3a5d | Boud Roukema | 02 October 2020, 11:32:03 UTC | Text: minor smoothing; Zenodo table ID; Fig 15 table link In this commit: minor tidying of some of Marius' recent text fixes; sentence length reduction and some minor improvements in logical flow. The zenodo ID 4062461 has been added (not to be confused with the zenodo ID of my CC-BY-SA-4.0 Cosmology lecture: https://zenodo.org/record/4060261 ). | 02 October 2020, 11:32:03 UTC |
| b677efb | Mohammad Akhlaghi | 02 October 2020, 00:12:35 UTC | TexLive's xstring package is now necessary After a fresh build of Maneage with a newly downloaded TeXLive, I noticed that it is complaining about not finding 'xstring.sty', apparently some package that depeneded on it is no longer including it itself! It is thus now added to the packages that are built by Maneage's TeXLive. | 02 October 2020, 00:12:35 UTC |
| 0f1b015 | Boud Roukema | 29 September 2020, 11:22:47 UTC | Merge branch 'elaphro' into N128cubed | 29 September 2020, 11:22:47 UTC |
| 355cc1c | Boud Roukema | 29 September 2020, 11:18:31 UTC | Add missing import os to plot_centres_voidstat.py | 29 September 2020, 11:18:31 UTC |
| bfa5639 | Boud Roukema | 29 September 2020, 09:59:39 UTC | Try fix for macOS/Darwin in plot_centres_voidstat.py | 29 September 2020, 09:59:39 UTC |
| 6bb9d50 | Boud Roukema | 28 September 2020, 23:08:14 UTC | Tidier fix for IO_GADGET_WITH_PATH - path may not be absolute | 28 September 2020, 23:08:20 UTC |
| 5352896 | Marius Peper | 28 September 2020, 23:01:13 UTC | Fix typo when setting the environmental variable holding the location of io_gadget.so | 28 September 2020, 23:01:13 UTC |
| a9fb242 | Boud Roukema | 28 September 2020, 22:11:40 UTC | Hardwire io_gadget.so location if macOS/Darwin This commit hardwires the io_gadget.so location if the system is suspected to be macOS/Darwin. | 28 September 2020, 22:11:40 UTC |
| 9f41b27 | Boud Roukema | 28 September 2020, 18:55:30 UTC | Attempted fix for macOS/Darwind LD_LIBRARY_PATH It is an error in python to call the value-by-key of a set (proper name?) type object if the key does not exist. However, using os.environ.get allows checking if the key exists without a fatal error. This commit should hopefully work for macOS/Darwin if python-3.8 in macOS/Darwin uses LD_LIBRARY_PATH for its linking library path. | 28 September 2020, 18:55:30 UTC |
| 96c215b | Marius Peper | 27 September 2020, 22:19:31 UTC | Minor edits to paper.tex | 27 September 2020, 22:19:31 UTC |
| 56068fe | Boud Roukema | 26 September 2020, 21:06:46 UTC | Fix for macOS/Darwin anti-LD_LIBRARY_PATH policy A recent commit used an alternative name, LD_LIBRARY_PATH_PYTHON, to pass the value of LD_LIBRARY_PATH from `analyse-plot.mk` to `reproduce/analysis/bash/run-plots`. This commit continues (maybe solves :)) this bypass procedure by passing the environment variable LD_LIBRARY_PATH_PYTHON from `reproduce/analysis/bash/run-plots` to reproduce/analysis/python/allresults_LSBGs.py and again using its value to set LD_LIBRARY_PATH. | 26 September 2020, 21:06:46 UTC |
| 21fe310 | Boud Roukema | 26 September 2020, 19:09:46 UTC | Minor: briefly show why cal H cap cal V makes sense | 26 September 2020, 19:09:46 UTC |
| 43ca74b | Marius Peper | 26 September 2020, 17:01:03 UTC | Rename f_{void} to give it a more self-explanatory name As suggested by Ahsan we rename f_{void} to f_{\,\cal{H}\cap \cal{V}}. f_{\,\cal{H}\cap \cal{V}} is the fraction of particles that are detected to be in a halo and are found to belong to a void at the same time. We think that this name is more self-explanatory. | 26 September 2020, 17:01:03 UTC |
| 90eb6f2 | Boud Roukema | 26 September 2020, 01:18:17 UTC | Record verification failures even if allowed If `verify-outputs` is not 'yes' in `verify-outputs.conf`, then we still want to record the failures, along with the successes, in .build/tex/macros/verify.tex . This commit implements this. For a run on branch 'elaphro', i.e. for a 32^3-particle simulation, you should expect about 3/4 of the parameters to be considered failures, as recorded in `.build/tex/macros/verify.tex` . The branch 'N128cubed', for 128^3 particles, should give successes for all parameters if your build and run have successfully matched what we expect. | 26 September 2020, 01:25:21 UTC |
| 7becbb3 | Boud Roukema | 26 September 2020, 01:18:17 UTC | Record verification failures even if allowed If `verify-outputs` is not 'yes' in `verify-outputs.conf`, then we still want to record the failures, along with the successes, in .build/tex/macros/verify.tex . This commit implements this. For a run on branch 'elaphro', i.e. for a 32^3-particle simulation, you should expect about 3/4 of the parameters to be considered failures, as recorded in `.build/tex/macros/verify.tex` . The branch 'N128cubed', for 128^3 particles, should give successes for all parameters if your build and run have successfully matched what we expect. | 26 September 2020, 01:22:22 UTC |
| 719e061 | Boud Roukema | 26 September 2020, 01:10:39 UTC | Quantify verification procedure in paper.tex This commit gives a more quantified and commit-identified definition of the statistical verification used in the paper. | 26 September 2020, 01:10:39 UTC |
| f4cdc3e | Boud Roukema | 26 September 2020, 01:05:53 UTC | Quantify verification procedure in paper.tex This commit gives a more quantified and commit-identified definition of the statistical verification used in the paper. | 26 September 2020, 01:05:53 UTC |
| b3b1d44 | Boud Roukema | 25 September 2020, 23:40:21 UTC | Merge branch 'elaphro' into N128cubed | 25 September 2020, 23:40:21 UTC |
| 4ed0db0 | Boud Roukema | 24 September 2020, 08:58:11 UTC | Cosmic variance: verify.mk checks and values in text This commit adds the estimates of our key parameters from 10 runs at 128^3 of commit 3725d92 and a mechanism of using these (or an equivalent set of 'verify' outputs) to verify the results of a single full run in a more statistically reasonable way than was done up to now. This method is briefly described by updating the reproducibility/cosmic variance pair of paragraphs in the text, and by adding \pm_{\mathrm{cv}} to the parameters where the cosmic variance is noticeable. Resolved conflicts: - paper.tex (particle mass - fact first, comment second) - reproduce/analysis/config/verify-outputs.conf (update for verify method) | 25 September 2020, 23:38:16 UTC |
| ab15662 | Boud Roukema | 25 September 2020, 23:04:32 UTC | Try to bypass overzealous macOS system integrity protection This commit implements a hack that seems to be needed for macOS/Darwin/xnu. Apparently, "system integrity protection" prevents LD_LIBRARY_PATH from being passed to a command even though that variable is explicitly prepended as an environment variable to the command. We solve the issue by giving the variable a different name in `analyse-plot.mk`, and in `run-plot` we restore the usual name. | 25 September 2020, 23:32:06 UTC |
| a01e3a2 | Boud Roukema | 25 September 2020, 23:31:32 UTC | Fix Mpc/h in L_box size | 25 September 2020, 23:31:32 UTC |
| e7d5ba5 | Boud Roukema | 25 September 2020, 22:38:04 UTC | Several small changes This commit makes several small improvements to the text thanks to suggestions by several people in our cosmology group. | 25 September 2020, 23:30:34 UTC |
| b7b26be | Boud Roukema | 25 September 2020, 23:04:32 UTC | Try to bypass overzealous macOS system integrity protection This commit implements a hack that seems to be needed for macOS/Darwin/xnu. Apparently, "system integrity protection" prevents LD_LIBRARY_PATH from being passed to a command even though that variable is explicitly prepended as an environment variable to the command. We solve the issue by giving the variable a different name in `analyse-plot.mk`, and in `run-plot` we restore the usual name. | 25 September 2020, 23:04:32 UTC |
| 1a5a3e3 | Boud Roukema | 25 September 2020, 22:38:04 UTC | Several small changes This commit makes several small improvements to the text thanks to suggestions by several people in our cosmology group. | 25 September 2020, 22:38:04 UTC |
| e46cf4d | Boud Roukema | 25 September 2020, 18:33:39 UTC | Improve consistency of round and curly brackets The convention in Maneage for bracketting in makefiles is (it seems to me) to use $(round brackets), not ${curly brackets}, at the 'make' level, and $$nobracketsnorunderscores for variables intended to pass through to the bash shell; while our convention in elaphrocentre modifies this to use $${curly_brackets_and_allow_underscores} for variables to be passed through to bash scripts. However, sometimes we haven't followed this convention fully. According to the GNU make manual [1], ${curly brackets} are just as valid as $(round brackets) at the 'make' level, but mixing the two types is discouraged. This commit hypothesises that there is a bug in the GNU make ore the GNU bash implemention in macOS/Darwin/xnu that occurs when a colon ':' is included indirectly (through nesting) in a ${curly_bracket}, by improving the consistency of using round brackets in `reproduce/analysis/make/analyse-plot.mk`. (Some locale variables are also explicitly set to English in `reproduce/analysis/bash/run-plots`.) In principle, this commit should not cause any difference in behaviour of the pipeline. However, it might fix a bug in macOS/Darwin/xnu. [1]: https://www.gnu.org/software/make/manual/html_node/Syntax-of-Functions.html#Syntax-of-Functions | 25 September 2020, 20:34:49 UTC |
| 02c55a4 | Marius Peper | 24 September 2020, 22:37:33 UTC | Ensure that the analyse-plot step is started when the parameters for revolver are changed Until this commit the analyse-plot step was not always started when we change the input parameters of our void finder, i.e. detect-voids-params.conf . To me it is still not clear why the analyse-plot step hasn't started after the void finder ran. This commit ensures that the analyse-plot step is started by making the parameter-file of the void finder to a dependency of the analyse-plot step. | 25 September 2020, 20:34:34 UTC |
| 68a0546 | Marius Peper | 24 September 2020, 14:38:43 UTC | Increase the ZOBOV_BUFFER for a 32^3 run to avoid a crash in revolver This commit increases the variable ZOBOV_BUFFER for the 32^3 particle run because for some seeds the pipeline can crash when revolver tries to identify voids. ZOBOV splits the box into several subboxes when searching for voids. ZOBOV_BUFFER specifies the overlap between these subboxes. Especially for low particle runs it is possible that ZOBOV crashes when ZOBOV_BUFFER isn't chosen properly. This commit increases ZOBOV_BUFFER a lot so that these crashes should not occur anymore. | 25 September 2020, 20:34:21 UTC |
| 9375723 | Boud Roukema | 24 September 2020, 12:52:23 UTC | Force the use of C and English locales Many European languages use the decimal comma, e.g. pi \approx 3,14159 instead of pi \approx 3.14159 . Many GNU shell tools handle these automatically in the user's locale. We'll assume that only English and the decimal point are used. | 25 September 2020, 20:34:12 UTC |
| 792b8e9 | Boud Roukema | 25 September 2020, 18:33:39 UTC | Improve consistency of round and curly brackets The convention in Maneage for bracketting in makefiles is (it seems to me) to use $(round brackets), not ${curly brackets}, at the 'make' level, and $$nobracketsnorunderscores for variables intended to pass through to the bash shell; while our convention in elaphrocentre modifies this to use $${curly_brackets_and_allow_underscores} for variables to be passed through to bash scripts. However, sometimes we haven't followed this convention fully. According to the GNU make manual [1], ${curly brackets} are just as valid as $(round brackets) at the 'make' level, but mixing the two types is discouraged. This commit hypothesises that there is a bug in the GNU make ore the GNU bash implemention in macOS/Darwin/xnu that occurs when a colon ':' is included indirectly (through nesting) in a ${curly_bracket}, by improving the consistency of using round brackets in `reproduce/analysis/make/analyse-plot.mk`. (Some locale variables are also explicitly set to English in `reproduce/analysis/bash/run-plots`.) In principle, this commit should not cause any difference in behaviour of the pipeline. However, it might fix a bug in macOS/Darwin/xnu. [1]: https://www.gnu.org/software/make/manual/html_node/Syntax-of-Functions.html#Syntax-of-Functions | 25 September 2020, 18:33:39 UTC |
| 32578e1 | Marius Peper | 24 September 2020, 22:37:33 UTC | Ensure that the analyse-plot step is started when the parameters for revolver are changed Until this commit the analyse-plot step was not always started when we change the input parameters of our void finder, i.e. detect-voids-params.conf . To me it is still not clear why the analyse-plot step hasn't started after the void finder ran. This commit ensures that the analyse-plot step is started by making the parameter-file of the void finder to a dependency of the analyse-plot step. | 24 September 2020, 22:37:33 UTC |
| 2b8f35d | Marius Peper | 24 September 2020, 14:38:43 UTC | Increase the ZOBOV_BUFFER for a 32^3 run to avoid a crash in revolver This commit increases the variable ZOBOV_BUFFER for the 32^3 particle run because for some seeds the pipeline can crash when revolver tries to identify voids. ZOBOV splits the box into several subboxes when searching for voids. ZOBOV_BUFFER specifies the overlap between these subboxes. Especially for low particle runs it is possible that ZOBOV crashes when ZOBOV_BUFFER isn't chosen properly. This commit increases ZOBOV_BUFFER a lot so that these crashes should not occur anymore. | 24 September 2020, 14:38:43 UTC |
| c79c7b4 | Boud Roukema | 24 September 2020, 12:52:23 UTC | Force the use of C and English locales Many European languages use the decimal comma, e.g. pi \approx 3,14159 instead of pi \approx 3.14159 . Many GNU shell tools handle these automatically in the user's locale. We'll assume that only English and the decimal point are used. | 24 September 2020, 12:52:23 UTC |
| 8ed52cc | Boud Roukema | 24 September 2020, 08:58:11 UTC | Cosmic variance: verify.mk checks and values in text This commit adds the estimates of our key parameters from 10 runs at 128^3 of commit 3725d92 and a mechanism of using these (or an equivalent set of 'verify' outputs) to verify the results of a single full run in a more statistically reasonable way than was done up to now. This method is briefly described by updating the reproducibility/cosmic variance pair of paragraphs in the text, and by adding \pm_{\mathrm{cv}} to the parameters where the cosmic variance is noticeable. | 24 September 2020, 08:58:11 UTC |
| ee8c6be | Mohammad Akhlaghi | 24 September 2020, 01:15:40 UTC | Gnuastro's analysis configuration files removed Until now, the core Maneage branch included some configuration files for Gnuastro's programs. This was actually a remnant of the distant past when Maneage didn't actually build its own software and we had to rely on the host's software versions. This file contained the configuration files specific to Gnuastro for this project and also had a feature to avoid checking the host's own configuration files. However, we now build all our software ourselves with fixed configuration files (for the version that is being installed and its version is stored). So those extra configuration files were just extra and caused confusion and problems in some scenarios. With this commit, those extra files are now removed. Also, two small issues are also addressed in parallel with this commit: - When running './project make clean', the 'hardware-parameters.tex' macro file (which is created by './project configure' is not deleted. - The project title is now written into the default output's PDF's properties (through 'hypersetup' in 'tex/src/preamble-header.tex') through the LaTeX macro. All these issues were found and fixed with the help of Samane Raji. | 24 September 2020, 01:15:40 UTC |
| a41f181 | Boud Roukema | 24 September 2020, 00:59:10 UTC | Minor fix to rockstar diagnostic info | 24 September 2020, 01:13:07 UTC |
| 2c68613 | Boud Roukema | 24 September 2020, 00:59:10 UTC | Minor fix to rockstar diagnostic info | 24 September 2020, 00:59:10 UTC |
| 4ceebea | Boud Roukema | 24 September 2020, 00:42:05 UTC | Merge branch 'elaphro' into N128cubed | 24 September 2020, 00:42:05 UTC |
| 829abd2 | Boud Roukema | 24 September 2020, 00:36:02 UTC | Minor: add PYTHONPATH in analyse-plot.mk | 24 September 2020, 00:36:02 UTC |
| caf02fe | Boud Roukema | 23 September 2020, 23:51:16 UTC | Prepare for merge with N128cubed; tidy remnants of c75fc48 This commit tidies up some of the remaining effects in `paper.tex` related to the experimental commit c75fc48; and removes a deleted paragraph that accidentally got restored. | 24 September 2020, 00:34:13 UTC |
| 54a4baf | Boud Roukema | 23 September 2020, 23:24:00 UTC | Revert "Use the existing structures to make the analysis ..." This reverts commit c75fc48a13cec8abdeffd3030d275aa9076db5f9 and its followup commit 40c40a2 and tidies up some minor paper.tex conflicts, mainly to do with S vs \lambda, and one case of v_{\perp} accidentally being written as v_{\parallel}. | 24 September 2020, 00:30:33 UTC |
| 0059f74 | Boud Roukema | 24 September 2020, 00:17:17 UTC | Merge branch 'rockstar_dns' into N128cubed | 24 September 2020, 00:17:17 UTC |
| ef5f2c1 | Boud Roukema | 23 September 2020, 12:04:41 UTC | Rockstar diagnoses; localhost for server | 24 September 2020, 00:14:46 UTC |
| d742062 | Boud Roukema | 23 September 2020, 23:51:16 UTC | Prepare for merge with N128cubed; tidy remnants of c75fc48 This commit tidies up some of the remaining effects in `paper.tex` related to the experimental commit c75fc48; and removes a deleted paragraph that accidentally got restored. | 23 September 2020, 23:51:16 UTC |
| 6787f6a | Boud Roukema | 22 September 2020, 15:40:10 UTC | Add xstring to texlive-packages.conf; comment out h5py import On some machines, the texlive package 'xstring' is found to be missing, so this is added to `reproduce/software/config/texlive-packages.conf` in this commit. In `reproduce/analysis/python/allresults_LSBGs.py`, the import of h5py is commented out, because it appears that we don't use it at all. | 23 September 2020, 23:27:52 UTC |
| bf15f53 | Boud Roukema | 22 September 2020, 03:09:37 UTC | Maneage python scripts need the python path | 23 September 2020, 23:27:43 UTC |
| 52a17c0 | Boud Roukema | 22 September 2020, 03:01:30 UTC | Diagnosis: print out sys.path in main python script This commit adds one line to `reproduce/analysis/python/allresults_LSBGs.py` for diagnosis on a machine that claims `ModuleNotFoundError: No module named 'h5py'` on what is now line 43, even though h5py was installed correctly. | 23 September 2020, 23:27:31 UTC |
| 1d92d4b | Boud Roukema | 23 September 2020, 23:24:00 UTC | Revert "Use the existing structures to make the analysis ..." This reverts commit c75fc48a13cec8abdeffd3030d275aa9076db5f9 and its followup commit 40c40a2 and tidies up some minor paper.tex conflicts, mainly to do with S vs \lambda, and one case of v_{\perp} accidentally being written as v_{\parallel}. | 23 September 2020, 23:24:00 UTC |
| 40c40a2 | Boud Roukema | 23 September 2020, 23:18:06 UTC | Minor fixes double run over halo_is_in_void_min_fraction This does some minor fixes to get c75fc48 to work in 'elaphro' (32^3). | 23 September 2020, 23:18:06 UTC |
| b9218fd | Boud Roukema | 23 September 2020, 12:04:41 UTC | Rockstar diagnoses; localhost for server | 23 September 2020, 12:04:41 UTC |
| 098f719 | Boud Roukema | 23 September 2020, 04:32:25 UTC | Merge branch 'elaphro' of codeberg:boud/elaphrocentre into elaphro Conflicts, mainly involved S versus \lambda, were resolved by hand; some material may wrongly have been lost or re-inserted after deletion. | 23 September 2020, 04:32:25 UTC |
| 5d4e15c | Boud Roukema | 23 September 2020, 04:21:23 UTC | Rework conclusion; s/S/\lambda/g; bibliography This commit does some general reorganisation of the recent changes to the Conclusion, aiming to make the references to the spin parameter consistent throughout, and better reflecting the history of the parameter and how we use it, and how we happen to get a spin parameter result fully consistent with the known literature (to within our calculated statistical accuracy). This commit also replaces S by \lambda, since the latter is more common in recent literature. This commit also tidies the alphabetical order of the bibliography for author names that start with a lower case letter, adding two scripts to hack this. The rules `paper-full.tex` and `paper-full.pdf` are now used instead of `paper.pdf` in `reproduce/analysis/make/paper.mk`. | 23 September 2020, 04:21:23 UTC |
| 210d5c3 | Marius Peper | 23 September 2020, 01:28:52 UTC | Merge branch 'elaphro' of codeberg:boud/elaphrocentre into elaphro | 23 September 2020, 01:28:52 UTC |
| c75fc48 | Marius Peper | 23 September 2020, 01:13:56 UTC | Use the existing structures to make the analysis for different values of halo_is_in_void_min_fraction This commit uses the structure that was developed to run chosen steps of the pipeline multiple times. If we want to compare include a comparison of the outcome of the pipeline with different input parameters we have to run chosen steps of the pipeline multiple times and save the outcome under a different name. This commit makes small changes in order to pass multiple values of halo_is_in_void_min_fraction to the pipeline. Currently the pipeline will start the analyse-plot step twice, for halo_is_in_void_min_fraction=0.5 and 0.99. This commit also modifies table 1 & 2 in paper.tex, adding the values of the second run to the table. | 23 September 2020, 01:13:56 UTC |
| 91f11d0 | Boud Roukema | 23 September 2020, 00:09:31 UTC | Merge branch 'elaphro' of codeberg:boud/elaphrocentre into elaphro | 23 September 2020, 00:09:31 UTC |
| 2fa1723 | Boud Roukema | 23 September 2020, 00:08:52 UTC | Fixes to paper.tex thanks to Ahsan This commit does minor improvements to `paper.tex` to make the text clearer, mostly based on Ahsan Nazer's feedback, plus a few other fixes. | 23 September 2020, 00:08:52 UTC |
| 3725d92 | Boud Roukema | 22 September 2020, 15:40:10 UTC | Add xstring to texlive-packages.conf; comment out h5py import On some machines, the texlive package 'xstring' is found to be missing, so this is added to `reproduce/software/config/texlive-packages.conf` in this commit. In `reproduce/analysis/python/allresults_LSBGs.py`, the import of h5py is commented out, because it appears that we don't use it at all. | 22 September 2020, 15:40:10 UTC |
| 5c52f60 | Boud Roukema | 22 September 2020, 03:09:37 UTC | Maneage python scripts need the python path | 22 September 2020, 03:09:37 UTC |
| 04a3215 | Boud Roukema | 22 September 2020, 03:01:30 UTC | Diagnosis: print out sys.path in main python script This commit adds one line to `reproduce/analysis/python/allresults_LSBGs.py` for diagnosis on a machine that claims `ModuleNotFoundError: No module named 'h5py'` on what is now line 43, even though h5py was installed correctly. | 22 September 2020, 03:01:30 UTC |
| 14d7c91 | Marius Peper | 22 September 2020, 01:39:25 UTC | Merge branch 'elaphro' of codeberg:boud/elaphrocentre into elaphro | 22 September 2020, 01:39:25 UTC |
| dc4c345 | Marius Peper | 22 September 2020, 01:30:27 UTC | Minor edits to the conclusion comparing our results to other works This commit adds a few sentences comparing our results to the work of other authors that analysed galaxies (or LSBGs/UDGs in detail). | 22 September 2020, 01:30:27 UTC |
| 3bdff4e | Boud Roukema | 21 September 2020, 20:53:45 UTC | Tighten reproducibility - freeze pipeline software versions Prior to this commit, we were not aware that github and bitbucket provided URLs for tar.gz downloads using individual commits without git tags. This lack of knowledge had the effect that any upstream update caused the install to fail, due to checksum mismatching. This situation may be acceptable in an interactive session, since the user can choose to accept to use the updated version, or provide his/her own fork in the case of disagreement, but this is not acceptable in an automated install. Moreover, this situation reduces the ability of the reader to reproduce the exact software contributing to a particular published paper. To solve this problem, this commit uses specific commit hashes for the actively developed cosmological software packages that do not currently change fast, but are very likely to change on a 5-10 year time scale. This affects: `reproduce/software/config/checksums.conf` `reproduce/software/config/urls.conf` `reproduce/software/config/versions.conf` `reproduce/software/make/high-level.mk` This commit also tidies up the use of directory names within the packages for the install sequence, since these names have well-defined patterns, separately for github and bitbucket (provided that a 12-character initial part of the commit hash is used in the case of bitbucket). This is a better option than using `ls` and a wildcard to find out the directory name. Non-cosmo packages did not need this change, because they have generally been installable from releases tagged with version numbers. | 21 September 2020, 21:09:59 UTC |
