B ú‹d§«ã@sœdZddlZddlZddlZddlZddlZddlmZm Z ddl m Z ddl m Z mZddlmZmZmZmZmZmZmZmZmZdgZGd d„dƒZdS) zˆ The base `Sampler` class containing various helpful functions. All other samplers inherit this class either explicitly or implicitly. éNé)ÚResultsÚprint_fn)ÚUnitCube)Ú sample_unifÚSamplerArgument) Úget_seed_sequenceÚget_print_funcÚprogress_integrationÚIteratorResultÚ RunRecordÚget_neff_from_logwtÚcompute_integralsÚ DelayTimerÚ _LOWL_VALÚSamplerc @seZdZdZd2dd„Zdd„Zdd „Zd d „Zd d „Zd3dd„Z dd„Z dd„Z dd„Z e dd„ƒZe dd„ƒZe dd„ƒZd4dd„Zdd „Zd!d"„Zd#d$„Zd%d&„Zd'd(„Zddd)ejejd*d*d*df d+d,„Zdddejdd*d*dd*dd-df d.d/„Zd5d0d1„ZdS)6ra8 The basic sampler object that performs the actual nested sampling. Parameters ---------- loglikelihood : function Function returning ln(likelihood) given parameters as a 1-d `~numpy` array of length `ndim`. prior_transform : function Function transforming a sample from the a unit cube to the parameter space of interest according to the prior. npdim : int, optional Number of parameters accepted by `prior_transform`. live_points : list of 3 `~numpy.ndarray` each with shape (nlive, ndim) Initial set of "live" points. Contains `live_u`, the coordinates on the unit cube, `live_v`, the transformed variables, and `live_logl`, the associated loglikelihoods. update_interval : int Only update the bounding distribution every `update_interval`-th likelihood call. first_update : dict A dictionary containing parameters governing when the sampler should first update the bounding distribution from the unit cube to the one specified by the user. rstate : `~numpy.random.Generator` `~numpy.random.Generator` instance. queue_size: int Carry out likelihood evaluations in parallel by queueing up new live point proposals using (at most) this many threads/members. pool: pool Use this pool of workers to execute operations in parallel. use_pool : dict, optional A dictionary containing flags indicating where the provided `pool` should be used to execute operations in parallel. rFcCsÖ||_||_||_| |_| |_|dd…\|_|_|_| rF|d|_nd|_t |jƒ|_ t j |j t d|_t j |j t d|_||_d|_d|_i|_| |_|jdkr¬t|_n| j|_| |_|  dd¡|_|  dd¡|_|  dd¡|_|  dd¡|_|jr||_nd|_g|_d |_d |_ d |_!d|_"|j |_#t$ %|j d |j ¡|_&d |_'d |_(d |_)d|_*d|_+||_,| dd|j ¡|_-| dd¡|_.d|_/d|_0t1|jƒg|_2d|_3d |_4| |_5d |_6d|_7d|_8t9ƒ|_:dS)Né)ÚdtypeÚprior_transformTÚ loglikelihoodÚ propose_pointZ update_boundrrgð?FgÚZ min_ncalléZmin_effg$@);rrÚnpdimÚncdimÚblobÚlive_uÚlive_vÚ live_loglÚ live_blobsÚlenÚnliveÚnpÚzerosÚintÚ live_boundÚlive_itÚrstateÚscaleÚmethodÚkwargsÚpoolÚmapÚMÚuse_poolÚgetÚuse_pool_ptformZ use_pool_loglÚuse_pool_evolveZuse_pool_updateÚ queue_sizeÚqueueÚnqueueÚunusedÚusedÚitÚncallÚmathÚlogÚdlvÚ added_liveÚeffÚciteÚ save_samplesÚ save_boundsÚbound_update_intervalÚfirst_bound_update_ncallÚfirst_bound_update_effÚlogl_first_updateÚunit_cube_samplingrÚboundÚnboundÚncall_at_last_updateÚ logvol_initÚ plateau_modeÚplateau_counterÚplateau_logdvolr Ú saved_run)ÚselfrrrZ live_pointsZupdate_intervalZ first_updater'r2r+r.rrIr©rOú\/work/yifan.wang/ringdown/master-ringdown-env/lib/python3.7/site-packages/dynesty/sampler.pyÚ__init__Gsl   zSampler.__init__cCs tdƒ‚dS)NzShould be overriden)Ú RuntimeError)rNÚfnamerOrOrPÚsaveŸsz Sampler.savecGs tdƒ‚dS)NzShould be overriden)rR)rNÚargsrOrOrPr¢szSampler.propose_pointcGs tdƒ‚dS)NzShould be overriden)rR)rNrUrOrOrPÚ evolve_point¥szSampler.evolve_pointcOs tdƒ‚dS)NzShould be overriden)rR)rNrUr*rOrOrPÚupdate_proposal¨szSampler.update_proposalNcCs tdƒ‚dS)NzShould be overriden)rR)rNÚsubsetrOrOrPÚupdate«szSampler.updatecCs||_d|_t|_dS)N)Ú__dict__r+r,r-)rNÚstaterOrOrPÚ __setstate__®szSampler.__setstate__cCs*|j ¡}xdD]}||kr||=qW|S)z#Get state information for pickling.)r-r+)rZÚcopy)rNr[ÚkrOrOrPÚ __getstate__³s    zSampler.__getstate__c Cs|jj|j|jfd|_|jrBt t|  |j t  |j¡¡ƒ¡|_ n t tt |j t  |j¡ƒƒ¡|_ t dd„|j  t  |j ¡¡Dƒ¡|_tj|jtd|_tj|jtd|_g|_d|_d|_d|_d|_|j|_t|jƒg|_d|_d|_d|_d|_ d |_!d |_"t#ƒ|_$d S) zRe-initialize the sampler.)ÚsizecSsg|] }|j‘qSrO)Úval)Ú.0Ú_rOrOrPú Êsz!Sampler.reset..)rrrTFN)%r'Úrandomr!rrr0r"ÚarrayÚlistr-rÚasarrayrr,rrr#r$r%r&r3r4r5r6r7r8rrrFrGrEr<rJrKrLr rM)rNrOrOrPÚreset¼s0 "z Sampler.resetc Cs|i}x dD]}t |j|¡||<q W|jrøt ¡¼t d¡d|jfd|jdfd|dfd|j fd |d fd |d fg}x"d D]}|  d |||f¡q„WxdD]}|  |||f¡q¨W|  dt  |d¡f¡|  d|df¡WdQRXnt dƒ‚|j rt|  dt |j¡f¡|  dtj|jdtdf¡|  dtj|jdtdf¡|  dt |jd¡f¡t|ƒS)zrSaved results from the nested sampling run. If bounding distributions were saved, those are also returned.) ÚncÚvÚidr7ÚuÚlogwtÚloglÚlogvolÚlogzÚlogzvarÚhrÚignorer!Zniterrr8rjr=Zsamplesrkr)rlr7rmZsamples_)rnrorprqZlogzerrrrÚ informationrsNzYou didn't save any samples!rFZ bound_iterÚ bounditer)rZ samples_boundÚboundidxr()r"rfrMr?ÚwarningsÚcatch_warningsÚ simplefilterr!r7r=ÚappendÚsqrtÚ ValueErrorr@r]ÚdeepcopyrFr$r)rNÚdr^ÚresultsrOrOrPr€äs2      zSampler.resultscCs<|jd}t|ƒdks&t t |¡¡r*dStt |¡ƒSdS)a Estimate the effective number of posterior samples using the Kish Effective Sample Size (ESS) where `ESS = sum(wts)^2 / sum(wts^2)`. Note that this is `len(wts)` when `wts` are uniform and `1` if there is only one non-zero element in `wts`. rnrN)rMr r"ZisneginfÚmaxr rh)rNrnrOrOrPÚ n_effective s zSampler.n_effectivecCs|jS)zw Return list of papers that should be cited given the specified configuration of the sampler. )r>)rNrOrOrPÚ citationsszSampler.citationsc CsÎ|dkr|j}||jk}||j|jk}|j|jk}|jrB|rB|sj|jsL|sj|jrf|jdk rf||jksj|rÊ|tkr~|j |k}nt dƒ}|j |d}|j r¤|j  |¡|jd7_||_|jrÊd|_||_dS)z¼ Here we update the bound depending on the situation The arguments are the loglstar and number of calls if force is true we update the bound no matter what N)rXrF)r8rBrArHr=rCrErDrrÚslicerYr@rFr{rG) rNÚloglstarr8ÚforceZcall_check_firstZ call_checkZefficiency_checkrXrFrOrOrPÚupdate_bound_if_needed&s*       zSampler.update_bound_if_neededc Cs”|jdkr:t |j|k¡df}t|dƒdkr>tdƒ‚nd}|js–g}g}|j}x¨|j|j kr’|j |Ž\}}|  |¡|  |¡|jd7_qTWnf|j j |j |j|jfd}t |j¡ddd…dd…ft |j |j¡dd…ddf}t}|j |_|j dkrt|j |j ƒ}n|j g}|jr0|j} nt} g}xFt|j ƒD]8} |  t|| ||| |j|j|j|| |jd¡qDWt| ||ƒƒ|_dS) z7Sequentially add new live point proposals to the queue.Zunifrz©No live points are above loglstar. Do you have a likelihood plateau ? It is also possible that you are trying to sample excessively around the very peak of the posteriorrOr)r`N)rmr…Úaxesr(rrZrseedr*)r)r"Znonzerorr rRrErVr4r2rr{r'rerÚidentityrr#rrr1r-r,Úrangerr(rrr*rgr3) rNr…rUZ point_queueZ axes_queuerVÚpointrˆZseedsZmapperÚirOrOrPÚ _fill_queuePsP      zSampler._fill_queuecCsT|jdkr| |¡|j d¡\}}}}}|jd7_|jd8_|||||fS)z0Grab the first live point proposal in the queue.rr)r4rr3Úpopr6)rNr…rmrkrorjrrOrOrPÚ_get_point_values   zSampler._get_point_valuec Cs„|j}d}xl| |¡\}}}}}||7}||7}|dk rR|jsR|j||jdkd|jdkrj|j||d||kr Pq W||||fS)zjPropose points until a new point that satisfies the log-likelihood constraint `loglstar` is found.rN)rY)r8)r8rrErWr4r‡) rNr…r8Z ncall_accumrmrkrorjrrOrOrPÚ _new_points zSampler._new_pointccs˜|jrtdƒ‚nd|_t|jdƒdkrp|jdd}|jdd}|jdd}|jdd}|jd d}nd }d }d }|j}d }|js²t d t |j ¡d |j d ¡}ndt  d t |j ¡t  |j |¡ ¡}|j |j }t ||dt  d t |¡ |d ¡g¡}tj|dd }||7}t |j¡} t|jƒ} |jsX|jd } nd} x4t|j ƒD]$} | | } || || }}|j|  ¡}|j|  ¡}|jr¾|j|  ¡}nd}|j| }|j| }|j| }t|||||||ƒ\}}}}|}t d| ||¡}|jrL|j  t!| ||||||||d ||| |j"|d¡d|j#| |j$|_%t&| ||||||||d |||| |j%|dVqjWdS)z½Add the remaining set of live points to the current set of dead points. Instantiates a generator that will be called by the user. Returns the same outputs as :meth:`sample`.zIThe remaining live points have already been added to the list of samples!Trqréÿÿÿÿrrrsrorpggœuˆ<ä7þgð?r)ÚprependN)rlrmrkrorprnrqrrrsrjrwr7rvr(rgY@)ÚworstÚustarÚvstarr…rprnrqrrrsrjrÚworst_itrwrvr=Ú delta_logz)'r<r}r rMrIrJr"r:Zaranger!Úlog1prKÚexprLZ concatenateÚdiffZargsortrrrErGrŠrr]rrrr%r&r Ú logaddexpr?r{Údictr(r7r8r=r )rNrqrrrsr…rpZlogvolsZnrestZdlvsZ lsort_idxZloglmaxrvrŒÚidxr;r”r•Úold_blobÚ loglstar_newrwZpoint_itrnr—rOrOrPÚadd_live_points¿s    *        zSampler.add_live_pointscCsB|jr6d|_|jr>x*dD]}|j||j d…=qWntdƒ‚dS)zlRemove the final set of live points if they were previously added to the current set of dead points.F)rlrmrkrorprnrqrrrsrjrwr7rvr(Nz1No live points were added to the list of samples!)r<r?rMr!r})rNr^rOrOrPÚ_remove_live_pointsFs  zSampler._remove_live_pointsg{®Gáz„?Tc %#s$|dkrtj}|dkrtj}|ˆ_|ˆ_d} ˆjdksHtˆjdƒdkrdd} d} d} ˆj}d}d}nJˆjrv| svˆ  ¡‡fdd „d Dƒ\} } } }}t   dt   ˆj ¡|| ¡}d}d }xfttjƒD]V}||krÖd }| |krâd }t   dt   ˆj ¡|| ¡}|dk r||krd }||kr$d }|dk r~t  |¡s~ˆj}||kr~|rpˆjd d ˆj}ˆ  ¡d ˆ_||kr~d }ˆj  ¡dkrœt d¡d }|rʈjsÈt| | | ||d}ˆj |¡Pt  ˆj ¡}ˆj|}ˆj|}ˆjs6ˆj ˆj |k ¡}|dkr6d ˆ_|ˆ_t  dˆjd¡|ˆ_ˆjsFˆj }nt  !t  "ˆj|¡ ¡ }|dksnt#‚||8}ˆj$| %¡}ˆj&| %¡}ˆj |}ˆj'r´ˆj(| %¡}nd}ˆ )|¡\}}} }!| |!7} ˆj*|!7_*ˆj'rð| j'}"nd}"t+||| | ||| ƒ\}#} } } |}ˆj,s(ˆj-d}$nd}$ˆjrbˆj t||||||#| | | |!||$ˆj.|d¡|ˆj$|<|ˆj&|<| ˆj |<|$ˆj|<ˆjˆj|<ˆj'r¨|"ˆj(|<dˆjˆj*ˆ_/ˆjd7_ˆjrðˆjd8_ˆjdkrðd ˆ_t0||||||#| | | |!||||$ˆj/|dVqÄWdS)a, **The main nested sampling loop.** Iteratively replace the worst live point with a sample drawn uniformly from the prior until the provided stopping criteria are reached. Instantiates a generator that will be called by the user. Parameters ---------- maxiter : int, optional Maximum number of iterations. Iteration may stop earlier if the termination condition is reached. Default is `sys.maxsize` (no limit). maxcall : int, optional Maximum number of likelihood evaluations. Iteration may stop earlier if termination condition is reached. Default is `sys.maxsize` (no limit). dlogz : float, optional Iteration will stop when the estimated contribution of the remaining prior volume to the total evidence falls below this threshold. Explicitly, the stopping criterion is `ln(z + z_est) - ln(z) < dlogz`, where `z` is the current evidence from all saved samples and `z_est` is the estimated contribution from the remaining volume. Default is `0.01`. logl_max : float, optional Iteration will stop when the sampled ln(likelihood) exceeds the threshold set by `logl_max`. Default is no bound (`np.inf`). n_effective: int, optional Minimum number of effective posterior samples. If the estimated "effective sample size" (ESS) exceeds this number, sampling will terminate. Default is no ESS (`np.inf`). add_live : bool, optional Whether or not to add the remaining set of live points to the list of samples when calculating `n_effective`. Default is `True`. save_bounds : bool, optional Whether or not to save past distributions used to bound the live points internally. Default is `True`. save_samples : bool, optional Whether or not to save past samples from the nested sampling run (along with other ancillary quantities) internally. Default is `True`. Returns ------- worst : int Index of the live point with the worst likelihood. This is our new dead point sample. ustar : `~numpy.ndarray` with shape (npdim,) Position of the sample. vstar : `~numpy.ndarray` with shape (ndim,) Transformed position of the sample. loglstar : float Ln(likelihood) of the sample. logvol : float Ln(prior volume) within the sample. logwt : float Ln(weight) of the sample. logz : float Cumulative ln(evidence) up to the sample (inclusive). logzvar : float Estimated cumulative variance on `logz` (inclusive). h : float Cumulative information up to the sample (inclusive). nc : int Number of likelihood calls performed before the new live point was accepted. worst_it : int Iteration when the live (now dead) point was originally proposed. boundidx : int Index of the bound the dead point was originally drawn from. bounditer : int Index of the bound being used at the current iteration. eff : float The cumulative sampling efficiency (in percent). delta_logz : float The estimated remaining evidence expressed as the ln(ratio) of the current evidence. Nrrroggœuˆ<ä7þgœuˆ<ä7~csg|]}ˆj|d‘qS)r‘)rM)rbrc)rNrOrPrdàsz"Sampler.sample..)rsrqrrrproFT)Úprint_progresszFWe have reached the plateau in the likelihood we are stopping sampling)rqrrrsrprogð?)rlrmrkrorprnrqrrrsrjr7rvr(rgY@)r“r”r•r…rprnrqrrrsrjrr–rwrvr=r—)1ÚsysÚmaxsizer?r@r7r rMrIr<r¡r"r›rrrŠZisposinfr‚Úadd_final_liveZptprxÚwarnrœr{Zargminr&r%rJÚsumrKr:r!rLr;r˜r™ÚAssertionErrorrr]rrrrr8r rErGr(r=r )%rNÚmaxiterÚmaxcallÚdlogzÚlogl_maxr‚Úadd_liver@r?Úresumer8rsrqrrrpr…r—ZnplateauZstop_iterationsr7Zcurrent_n_effectiveZadd_infor“r–rwZcur_dlvr”r•rŸržrmrkrorjZnew_blobrnrvrO)rNrPÚsampleWs p                      zSampler.sampleé<c  Csî|dk r2t ¡t d¡t dt¡WdQRX|dkrV|rRd|jdd}nd}| rn|jrnt d¡dSt||ƒ\} }| dk rŒt| ƒ}z<|j }xpt |j ||||| d|| |d ƒD]L\}}||j 7}|rê|j d }||||||d | dk r¸| ¡r¸| | ¡q¸W|r\|j d }xBt | ¡ƒD]2\}}||j 7}|r&|||||d ||d q&Wt|jd |jdd\}}}}| ¡|jd<| ¡|jd<| ¡|jd<| ¡|jd<| dk rÈ| | ¡Wd| dk rÞ|  ¡|j ¡XdS)aU **A wrapper that executes the main nested sampling loop.** Iteratively replace the worst live point with a sample drawn uniformly from the prior until the provided stopping criteria are reached. Parameters ---------- maxiter : int, optional Maximum number of iterations. Iteration may stop earlier if the termination condition is reached. Default is `sys.maxsize` (no limit). maxcall : int, optional Maximum number of likelihood evaluations. Iteration may stop earlier if termination condition is reached. Default is `sys.maxsize` (no limit). dlogz : float, optional Iteration will stop when the estimated contribution of the remaining prior volume to the total evidence falls below this threshold. Explicitly, the stopping criterion is `ln(z + z_est) - ln(z) < dlogz`, where `z` is the current evidence from all saved samples and `z_est` is the estimated contribution from the remaining volume. If `add_live` is `True`, the default is `1e-3 * (nlive - 1) + 0.01`. Otherwise, the default is `0.01`. logl_max : float, optional Iteration will stop when the sampled ln(likelihood) exceeds the threshold set by `logl_max`. Default is no bound (`np.inf`). n_effective: int, optional Minimum number of effective posterior samples. If the estimated "effective sample size" (ESS) exceeds this number, sampling will terminate. Default is no ESS (`np.inf`). This option is deprecated and will be removed in a future release. add_live : bool, optional Whether or not to add the remaining set of live points to the list of samples at the end of each run. Default is `True`. print_progress : bool, optional Whether or not to output a simple summary of the current run that updates with each iteration. Default is `True`. print_func : function, optional A function that prints out the current state of the sampler. If not provided, the default :meth:`results.print_fn` is used. save_bounds : bool, optional Whether or not to save past bounding distributions used to bound the live points internally. Default is *True*. checkpoint_file: string, optional if not None The state of the sampler will be saved into this file every checkpoint_every seconds checkpoint_every: float, optional The number of seconds between checkpoints that will save the internal state of the sampler. The sampler will also be saved in the end of the run irrespective of checkpoint_every. NÚoncezaThe n_effective option to Sampler.run_nested is deprecated and will be removed in future releasesgü©ñÒMbP?gð?g{®Gáz„?zAYou are resuming a finished static run. This will not do anythingT) r©rªr«r¬r@r?r‚r®r­r)r«r¬)Ú add_live_itr«r¬rorp)rorprnrqrrrs)rxryÚfilterwarningsr¦ÚDeprecationWarningr!r<r rr8Ú enumerater¯rjr7Zis_timerTr rrMÚtolistÚcloserZ history_save)rNr©rªr«r¬r‚r­r¢Ú print_funcr@Zcheckpoint_fileZcheckpoint_everyr®ÚpbarÚtimerr8r7r€rŒZ new_logwtZnew_logzZ new_logzvarZnew_hrOrOrPÚ run_nestedsxN          zSampler.run_nestedcCsz|dkr t}t||ƒ\}}zH|j}|jd}x2t| ¡ƒD]"\}}|r:|||||dddq:WWd|dk rt| ¡XdS)ao **A wrapper that executes the loop adding the final live points.** Adds the final set of live points to the pre-existing sequence of dead points from the current nested sampling run. Parameters ---------- print_progress : bool, optional Whether or not to output a simple summary of the current run that updates with each iteration. Default is `True`. print_func : function, optional A function that prints out the current state of the sampler. If not provided, the default :meth:`results.print_fn` is used. Nrg{®Gáz„?)r²r«)rr r8r7rµr r·)rNr¢r¸r¹r8r7rŒr€rOrOrPr¥0s zSampler.add_final_live)rF)N)NF)TN)Ú__name__Ú __module__Ú __qualname__Ú__doc__rQrTrrVrWrYr\r_riÚpropertyr€r‚rƒr‡rrrr r¡r"Úinfr¯r»r¥rOrOrOrPrs\- K  ( (  *?" 3 )r¿r£rxr9r]Únumpyr"r€rrZboundingrZsamplingrrÚutilsrr r r r r rrrÚ__all__rrOrOrOrPÚs ,