bilby issueshttps://git.ligo.org/lscsoft/bilby/-/issues2019-04-23T06:24:09Zhttps://git.ligo.org/lscsoft/bilby/-/issues/243improving CPNest integration2019-04-23T06:24:09ZVivien Raymondimproving CPNest integrationAdd cpnest https://github.com/johnveitch/cpnest to bilby. Maybe using something similar to https://git.ligo.org/john-veitch/lalsuite/blob/liwrapper/lalinference/python/lalinference_cpnest.py ...Add cpnest https://github.com/johnveitch/cpnest to bilby. Maybe using something similar to https://git.ligo.org/john-veitch/lalsuite/blob/liwrapper/lalinference/python/lalinference_cpnest.py ...FutureJohn Douglas Veitchjohn.veitch@ligo.orgJohn Douglas Veitchjohn.veitch@ligo.orghttps://git.ligo.org/lscsoft/bilby/-/issues/245Proposal library2019-06-19T14:41:01ZVivien RaymondProposal libraryin the `gw` package, we want to have a set of GW-specific jump proposals which can then be used by the samplers able to. The first step of this work is to define a structure and API of such a proposal library.in the `gw` package, we want to have a set of GW-specific jump proposals which can then be used by the samplers able to. The first step of this work is to define a structure and API of such a proposal library.FutureChristopher BerryMatthew CarneyChristopher Berryhttps://git.ligo.org/lscsoft/bilby/-/issues/275Add the "dynamic" nested sampler from `dynesty`.2019-04-23T06:28:37ZAtul DivakarlaAdd the "dynamic" nested sampler from `dynesty`.Is it possible to include 'dynamic' input in the dynesty sampler? It helps to get better posterior distributions, but currently if I write it turned on as:
result = bilby.run_sampler(
likelihood=likelihood, priors=prior, sampler='dy...Is it possible to include 'dynamic' input in the dynesty sampler? It helps to get better posterior distributions, but currently if I write it turned on as:
result = bilby.run_sampler(
likelihood=likelihood, priors=prior, sampler='dynesty', nlive=500,
injection_parameters=injection_param, outdir=outdir, label=label,
bound='multi', sample='rwalk', walks=25, verbose=True, **dynamic=True**)
I get the error:
09:47 **bilby WARNING : Supplied argument 'dynamic' not an argument of 'Dynesty', removing.**
09:47 bilby INFO : Using sampler Dynesty with kwargs {'bound': 'multi', 'sample': 'rwalk', 'verbose': True, 'check_point_delta_t': 600, 'nlive': 500, 'first_update': None, 'npdim': None, 'rstate': None, 'queue_size': None, 'pool': None, 'use_pool': None, 'live_points': None, 'logl_args': None, 'logl_kwargs': None, 'ptform_args': None, 'ptform_kwargs': None, 'enlarge': None, 'bootstrap': None, 'vol_dec': 0.5, 'vol_check': 2.0, 'facc': 0.5, 'slices': 5, 'walks': 25, 'update_interval': 300, 'print_func': <bound method Dynesty._print_func of <bilby.core.sampler.dynesty.Dynesty object at 0x1a3a11acf8>>, 'dlogz': 0.1, 'maxiter': None, 'maxcall': None, 'logl_max': inf, 'add_live': True, 'print_progress': True, 'save_bounds': True}Futurehttps://git.ligo.org/lscsoft/bilby/-/issues/282Save emcee chain during the run2019-02-08T02:43:58ZColm Talbotcolm.talbot@ligo.orgSave emcee chain during the runCurrently, if an emcee job crashes or is killed the chain is lost. There's a suggested method for writing each iteration to disk as it's generated [here](http://dfm.io/emcee/current/user/advanced/#incrementally-saving-progress). We shoul...Currently, if an emcee job crashes or is killed the chain is lost. There's a suggested method for writing each iteration to disk as it's generated [here](http://dfm.io/emcee/current/user/advanced/#incrementally-saving-progress). We should definitely do something like this. Preferably with the ability to resume from an old chain.https://git.ligo.org/lscsoft/bilby/-/issues/292Viewing angle as parameter2020-07-26T12:31:36ZCarl-Johan HasterViewing angle as parameterIt'd be good to be able to specify viewing angle as a parameter in which to sample. Here I define `viewing angle = |iota|`, ie this is simply saying that I don't care about the face-on vs. face-off properties.
This would be nice to have ...It'd be good to be able to specify viewing angle as a parameter in which to sample. Here I define `viewing angle = |iota|`, ie this is simply saying that I don't care about the face-on vs. face-off properties.
This would be nice to have when doing followup of GRBs and so onFuturehttps://git.ligo.org/lscsoft/bilby/-/issues/294Stopping conditions to produce N independent samples2019-06-25T04:04:59ZChristopher BerryStopping conditions to produce N independent samplesInvestigate how different samplers produce independent samples, so we can aim to produce a desired number of samples per runInvestigate how different samplers produce independent samples, so we can aim to produce a desired number of samples per runFutureChristopher BerryChristopher Berryhttps://git.ligo.org/lscsoft/bilby/-/issues/303Add detecor frame sky-parametrization2019-06-25T03:51:09ZCarl-Johan HasterAdd detecor frame sky-parametrizationIt's convenient to sample in this parametrization of the sky-coordinates
This is implemented in `LALInference` in https://git.ligo.org/lscsoft/lalsuite/blob/master/lalinference/src/DetectorFixedSkyCoords.cIt's convenient to sample in this parametrization of the sky-coordinates
This is implemented in `LALInference` in https://git.ligo.org/lscsoft/lalsuite/blob/master/lalinference/src/DetectorFixedSkyCoords.cFutureJohn Douglas Veitchjohn.veitch@ligo.orgJohn Douglas Veitchjohn.veitch@ligo.orghttps://git.ligo.org/lscsoft/bilby/-/issues/310Nestle steps argument doesn't work2019-07-26T05:56:21ZRhys GreenNestle steps argument doesn't workIf using the method = 'classic' argument, there should be an option to set the number of mcmc steps between samples, this is currently not accepted by bilbyIf using the method = 'classic' argument, there should be an option to set the number of mcmc steps between samples, this is currently not accepted by bilbyFuturehttps://git.ligo.org/lscsoft/bilby/-/issues/311Determine default for sampler, sampler settings, jump proposals2019-06-25T04:09:18ZEric ThraneDetermine default for sampler, sampler settings, jump proposalsThis work is already underway, but I'm creating a git issue to link to the review readiness [wiki](https://git.ligo.org/lscsoft/bilby/wikis/review_script).
@gregory.ashton, @christopher\-berry, @paul\-lasky, @vivienThis work is already underway, but I'm creating a git issue to link to the review readiness [wiki](https://git.ligo.org/lscsoft/bilby/wikis/review_script).
@gregory.ashton, @christopher\-berry, @paul\-lasky, @vivien1.0.0Christopher BerryChristopher Berryhttps://git.ligo.org/lscsoft/bilby/-/issues/315Fake sampler for likelihood consistency check2019-03-21T10:24:55ZCarl-Johan HasterFake sampler for likelihood consistency checkImplement a "fake sampler" which reads in a set of parameters from a file, and then computes the likelihood for those paramters.
This would be a really useful consistency check between the ROQ and nonROQ likelihood functions, as well as ...Implement a "fake sampler" which reads in a set of parameters from a file, and then computes the likelihood for those paramters.
This would be a really useful consistency check between the ROQ and nonROQ likelihood functions, as well as an useful check for when the likelihood function is updated/appended (as, in order to accept a merge request to the likelihood function it needs to sufficiently reproduce the likelihood values from a fiducial set of parameter samples)FutureMichael PuerrerMichael Puerrerhttps://git.ligo.org/lscsoft/bilby/-/issues/316emcee pre-release version does not run2019-02-21T03:45:41ZRhys Greenemcee pre-release version does not runUsing the emcee pre-release version
and running
```
emcee_result = bilby.run_sampler(
likelihood=likelihood, priors=priors, sampler='emcee' )
```
raised the error:
```python
home/c1572221/src/bilby/bilby/core/sampler/emcee.pyc ...Using the emcee pre-release version
and running
```
emcee_result = bilby.run_sampler(
likelihood=likelihood, priors=priors, sampler='emcee' )
```
raised the error:
```python
home/c1572221/src/bilby/bilby/core/sampler/emcee.pyc in run_sampler(self)
194 for sample in tqdm(sampler.sample(**self.sampler_function_kwargs),
195 total=self.nsteps):
--> 196 points = np.hstack([sample[0], np.array(sample[3])])
197 # import IPython; IPython.embed()
198 with open(out_file, "a") as ff:
TypeError: 'State' object does not support indexing
```0.4.1https://git.ligo.org/lscsoft/bilby/-/issues/321Evidence calculation for all!2020-02-06T11:50:33ZCarl-Johan HasterEvidence calculation for all!Make sure to check which samplers can produce evidences (and Bayes Factors), and how good they are at computing them.Make sure to check which samplers can produce evidences (and Bayes Factors), and how good they are at computing them.https://git.ligo.org/lscsoft/bilby/-/issues/325Add support for ROQ rescaling2019-05-08T02:35:14ZMichael PuerrerAdd support for ROQ rescalingRescaling of bases as in table I of https://arxiv.org/pdf/1604.08253.pdf.Rescaling of bases as in table I of https://arxiv.org/pdf/1604.08253.pdf.Carl-Johan HasterCarl-Johan Hasterhttps://git.ligo.org/lscsoft/bilby/-/issues/343Combine multiple runs2019-05-23T07:55:34ZColm Talbotcolm.talbot@ligo.orgCombine multiple runsWe need a method for combining the output of multiple runs, this will need to be treated a little differently for different samplers.
Hopefully, outputs from nested samplers can all be combined in this way, if we wish. Similar for MCMC....We need a method for combining the output of multiple runs, this will need to be treated a little differently for different samplers.
Hopefully, outputs from nested samplers can all be combined in this way, if we wish. Similar for MCMC. I'm not sure how best, if at all to combine MCMC runs with nest runs.
Related is thinning for auto-correlation length for MCMC samplers, which may or may not be done currently.
Ideally, this will be possible from the results files.https://git.ligo.org/lscsoft/bilby/-/issues/349CPNest failed to checkpoint in HTCondor2019-06-24T10:55:02ZTsun-Ho PangCPNest failed to checkpoint in HTCondorI have been running a parameter estimation run with CPNest as the sampler on the cluster (CIT) as a condor job. After the job got idle and rerun, all the progress are lost. And there are no .pkl output at the outdir/cpnest/.I have been running a parameter estimation run with CPNest as the sampler on the cluster (CIT) as a condor job. After the job got idle and rerun, all the progress are lost. And there are no .pkl output at the outdir/cpnest/.https://git.ligo.org/lscsoft/bilby/-/issues/367Generate initial live points in dynesty2019-05-27T04:09:58ZColm Talbotcolm.talbot@ligo.orgGenerate initial live points in dynestyDynesty provides the option to provide the initial live points to the sampler.
For constrained priors doing this will increase the initial efficiency of the run as you won't place any samples out of the allowed range.
Note that the ini...Dynesty provides the option to provide the initial live points to the sampler.
For constrained priors doing this will increase the initial efficiency of the run as you won't place any samples out of the allowed range.
Note that the initial samples **must be drawn from the prior**.0.5.1https://git.ligo.org/lscsoft/bilby/-/issues/369parameterized eos sampling2020-05-11T01:48:07ZMonica Rizzoparameterized eos samplingwant to include functionality to sample in eos parameters for a few different parameterized modelswant to include functionality to sample in eos parameters for a few different parameterized modelsMatthew CarneyMonica RizzoMatthew Carneyhttps://git.ligo.org/lscsoft/bilby/-/issues/425Recent release of emcee (3.0.0) does not work within BILBY2019-11-07T21:01:34ZEthan PayneRecent release of emcee (3.0.0) does not work within BILBYThe recent update to emcee (to version 3.0.0) has broken its implementation within bilby. The traceback that appears when running a standard `gaussian_example.py` with emcee is:
```
Traceback (most recent call last):
File "gaussian_ex...The recent update to emcee (to version 3.0.0) has broken its implementation within bilby. The traceback that appears when running a standard `gaussian_example.py` with emcee is:
```
Traceback (most recent call last):
File "gaussian_example.py", line 53, in <module>
nwalkers=10, outdir=outdir, label=label, store=True, thin=100, live_dangerously=True)
File "/home/ethan/Research/code_libraries/bilby/bilby/core/sampler/__init__.py", line 175, in run_sampler
result = sampler.run_sampler()
File "/home/ethan/Research/code_libraries/bilby/bilby/core/sampler/emcee.py", line 345, in run_sampler
iterations -= self._previous_iterations
File "/home/ethan/Research/code_libraries/bilby/bilby/core/sampler/emcee.py", line 310, in _previous_iterations
return len(self.sampler.blobs)
File "/home/ethan/Research/dev3/lib/python3.6/site-packages/emcee/utils.py", line 26, in f
return func(*args, **kwargs)
File "/home/ethan/Research/dev3/lib/python3.6/site-packages/emcee/ensemble.py", line 491, in blobs
return self.get_blobs()
File "/home/ethan/Research/dev3/lib/python3.6/site-packages/emcee/ensemble.py", line 509, in get_blobs
return self.get_value("blobs", **kwargs)
File "/home/ethan/Research/dev3/lib/python3.6/site-packages/emcee/ensemble.py", line 524, in get_value
return self.backend.get_value(name, **kwargs)
File "/home/ethan/Research/dev3/lib/python3.6/site-packages/emcee/backends/backend.py", line 43, in get_value
raise AttributeError("you must run the sampler with "
AttributeError: you must run the sampler with 'store == True' before accessing the results
```
@moritz.huebner @colm.talbot
Cheers Gavin Wallace for spotting this issue :smile: https://git.ligo.org/lscsoft/bilby/-/issues/438CPNest resuming on HTCondor2020-05-14T12:53:24ZGregory Ashtongregory.ashton@ligo.orgCPNest resuming on HTCondorFutureColm Talbotcolm.talbot@ligo.orgColm Talbotcolm.talbot@ligo.orghttps://git.ligo.org/lscsoft/bilby/-/issues/447Dynesty sampling breaks if a checkpoint is being written before all prior sam...2020-01-16T05:55:37ZMoritz HuebnerDynesty sampling breaks if a checkpoint is being written before all prior samples are drawnNoticed because `bilby_pipe` jobs were getting put on hold. If drawing from the prior is not finished by the time the `bilby_pipe` job reaches its `periodic_restart_time` and tries to checkpoint, the below error occurs because the sample...Noticed because `bilby_pipe` jobs were getting put on hold. If drawing from the prior is not finished by the time the `bilby_pipe` job reaches its `periodic_restart_time` and tries to checkpoint, the below error occurs because the sampler is not yet instantiated. This puts `bilby_pipe` jobs on hold.
Drawing from the prior can take a long time if the priors are non-sensical, e.g., if both `prior.minimum = 0` and `prior.maximum = 0`. Therefore, a good solution to this issue would be for `bilby` to flag if priors are weird.
```
Writing checkpoint file bilby_pipe_test/result/S190425_initial_analysis_public_data_bilby_pipe_test_data0_1240215503-0171_analysis_L1V1_dynesty_resume.pickle
Traceback (most recent call last):
File "/local/condor/execute/dir_126817/condor_exec.exe", line 11, in <module>
load_entry_point('bilby-pipe==0.3.8', 'console_scripts', 'bilby_pipe_analysis')()
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby_pipe-0.3.8-py3.6.egg/bilby_pipe/data_analysis.py", line 322, in main
analysis.run_sampler()
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby_pipe-0.3.8-py3.6.egg/bilby_pipe/data_analysis.py", line 299, in run_sampler
**self.sampler_kwargs,
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/__init__.py", line 176, in run_sampler
result = sampler.run_sampler()
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/dynesty.py", line 234, in run_sampler
self.kwargs['nlive']))
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/base_sampler.py", line 449, in get_initial_points_from_prior
if self.check_draw(theta, warning=False):
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/base_sampler.py", line 470, in check_draw
if np.isinf(self.log_prior(theta)):
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/base_sampler.py", line 378, in log_prior
return self.priors.ln_prob(params)
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/prior.py", line 418, in ln_prob
for key in sample], axis=axis)
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/prior.py", line 418, in <listcomp>
for key in sample], axis=axis)
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/prior.py", line 1663, in ln_prob
return -0.5 * ((self.mu - val) ** 2 / self.sigma ** 2 + np.log(2 * np.pi * self.sigma ** 2))
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/dynesty.py", line 415, in write_current_state_and_exit
self.write_current_state(plot=False)
File "/home/isobel.romero-shaw/anaconda3/lib/python3.6/site-packages/bilby-0.6.3-py3.6.egg/bilby/core/sampler/dynesty.py", line 444, in write_current_state
unit_cube_samples=self.sampler.saved_u,
AttributeError: 'Dynesty' object has no attribute 'sampler'
```0.6.4