o RŀgZ@sVdZddlZddlmZedezddlZWney*ddlmZeddwej Z ddZ ej d Z ee ZGd d d Zd d ZddZddZ    d5ddZdZ       d6ddZddZddZddZ   d7ddZdd Zd8d!d"Zd#d$Zd%d&Zd'd(Zd)d*Z d+d,Z!d-d.Z"d/d0Z#d1d2Z$d3d4Z%dS)9aA state-emitting MarkovModel. Note terminology similar to Manning and Schutze is used. Functions: train_bw Train a markov model using the Baum-Welch algorithm. train_visible Train a visible markov model using MLE. find_states Find the a state sequence that explains some observations. load Load a MarkovModel. save Save a MarkovModel. Classes: MarkovModel Holds the description of a markov model N)BiopythonDeprecationWarningzThe 'Bio.MarkovModel' module is deprecated and will be removed in a future release of Biopython. Consider using the hmmlearn package instead.)MissingPythonDependencyErrorzRPlease install NumPy if you want to use Bio.MarkovModel. See http://www.numpy.org/cCs@i}t|ddd}t|d}|D] \}}||||<q|S)zAReturn a dictionary of values with their sequence offset as keys.N) enumeratelen)valuesdentriesnindexkeyrC/var/www/html/myenv/lib/python3.10/site-packages/Bio/MarkovModel.py itemindex1s   rgYnc@s$eZdZdZ dddZddZdS) MarkovModelz+Create a state-emitting MarkovModel object.NcCs"||_||_||_||_||_dS)zInitialize the class.N)statesalphabet p_initial p_transition p_emission)selfrrrrrrrr__init__Ds  zMarkovModel.__init__cCs.ddlm}|}t|||d|S)z9Create a string representation of the MarkovModel object.r)StringIO)iorsaveseekread)rrhandlerrr__str__Ns   zMarkovModel.__str__NNN)__name__ __module__ __qualname____doc__rrrrrrrAs  rcCs*|}||std|d||S)zNRead the first line and evaluate that begisn with the correct start (PRIVATE).z I expected z but got )readline startswith ValueError)rstartlinerrr_readline_and_check_startXs r*cCs|t|d}|dd}t|d}|dd}t||}t|t|}}t||_t|d}tt|D]}t|d||d}t|d|j|<q9t||f|_ t|d }tt|D]!}t|d||d}d d |ddD|j |ddf<qet||f|_ t|d }tt|D]!}t|d||d}d d |ddD|j |ddf<q|S)z.Parse a file handle into a MarkovModel object.zSTATES:rNz ALPHABET:zINITIAL: :rz TRANSITION:cSg|]}t|qSrfloat.0vrrr yzload..z EMISSION:cSr-rr.r0rrrr3r4) r*splitrrnpzerosrranger/rr)rr)rrmmNMirrrload`s,      * *r=c Cs|j}|dd|jd|dd|jd|dtt|jD]}|d|j|d|j|ddq&|d tt|jD]}|d|j|ddd d |j|DdqF|d tt|jD]}|d|j|ddd d |j|DdqmdS)z$Save MarkovModel object into handle.zSTATES:   z ALPHABET: z INITIAL: r+z: gz TRANSITION: cs|]}t|VqdSNstrr1xrrr zsave..z EMISSION: csrArBrCrErrrrGrHN) writejoinrrr8rrrr)r9rwr<rrrrs&44rc st|t|}}|std|dur"t|}|j|fkr"td|dur6t|}|j||fkr6td|durJt|}|j||fkrJtdg} t||D]} | fdd| DqRdd| D} t| d krrtd t||| ||||d } | \} }}t ||| ||S) a Train a MarkovModel using the Baum-Welch algorithm. Train a MarkovModel using the Baum-Welch algorithm. states is a list of strings that describe the names of each state. alphabet is a list of objects that indicate the allowed outputs. training_data is a list of observations. Each observation is a list of objects from the alphabet. pseudo_initial, pseudo_transition, and pseudo_emission are optional parameters that you can use to assign pseudo-counts to different matrices. They should be matrices of the appropriate size that contain numbers to add to each parameter matrix, before normalization. update_fn is an optional callback that takes parameters (iteration, log_likelihood). It is called once per iteration. zNo training data given.N$pseudo_initial not shape len(states)5pseudo_transition not shape len(states) X len(states)5pseudo_emission not shape len(states) X len(alphabet)cg|]}|qSrrrEindexesrrr3r4ztrain_bw..cSr-r)rrErrrr3r4rz,I got training data with outputs of length 0)pseudo_initialpseudo_transitionpseudo_emission update_fn) rr'r6asarrayshaperappendmin _baum_welchr)rr training_datarRrSrTrUr:r;training_outputsoutputslengthsrFrrrrrPrtrain_bwsB      r_ic  Cs\|dur t|}nt||f}|durt||f}nt|||f}|dur,t||f}nt|||f}t|} t|} t|} |durLt|} nd} |durXt|}nd}|durdt|}nd}d}ttD]1}t}|D]}|t|||| | | | || 7}qr| dur| |||durt||dkrn |}qlt dtdd| | | fDS)zfImplement the Baum-Welch algorithm to evaluate unknown parameters in the MarkovModel object (PRIVATE).Ng?z%HMM did not converge in %d iterationscSsg|]}t|qSr)r6exp)r1_rrrr3&z_baum_welch..) _random_norm_copy_and_checkr6logr8MAX_ITERATIONSLOG0_baum_welch_onefabs RuntimeError)r:r;r\rrrrRrSrTrU lp_initial lp_transition lp_emissionlpseudo_initiallpseudo_transitionlpseudo_emission prev_llikr<llikr]rrrrZsV           rZc  Cst|} t|| ||||} t|| |||} t||| f} t| D]J} || }t||f}t|D])}t|D]"}| || ||||||| || d}||||<q8q2|t|| dddd| f<q!t|| f}t| D]} t|D]}t| |dd| f||| <q}qwt|}t|D]}t||ddf||<q|dddf}|durt||}|t|}t|D]5}t|D]}t| ||ddf|||||<q|durt|||||<||t||||<qt|D]K}t|t}t| D]} || }t|D]}t ||| ||| f||<qq |t|}|dur@t|||}|t|}|||ddf<qt| dd| fS)zExecute one step for Baum-Welch algorithm (PRIVATE). Do one iteration of Baum-Welch based on a sequence of output. Changes the value for lp_initial, lp_transition and lp_emission in place. rNr) r_forward _backwardr6r7r8_logsum _logvecaddrg logaddexp)r:r;r]rkrlrmrnrorpTfmatbmatlp_arctk lp_traverser<jlp lp_arcout_t lp_arcoutksumrrrrh)sh                (        rhc Cst||df}||dddf<td|dD]7}||d}t|D]*} t} t|D]} || |d|| | || |} t| | } q,| || |<q$q|S)zImplement forward algorithm (PRIVATE). Calculate a Nx(T+1) matrix, where the last column is the total probability of the output. rNrr6r7r8rgrw) r:rxrkrlrmr]matrixr|r}rlprobr<rrrrrss   ( rsc Cst||df}t|dddD]5}||}t|D]*}t} t|D]} || |d||| |||} t| | } q#| |||<qq|S)z'Implement backward algorithm (PRIVATE).rrr) r:rxrlrmr]rr|r}r<rrrrrrrts  ( rtcs&t|t|}}|durt|}|j|fkrtd|dur0t|}|j||fkr0td|durDt|}|j||fkrDtdgg}} t|t||D](\} } t| t| krctd|fdd| D| fdd| DqSt||| ||||} | \} }}t||| ||S) a Train a visible MarkovModel using maximum likelihoood estimates for each of the parameters. Train a visible MarkovModel using maximum likelihoood estimates for each of the parameters. states is a list of strings that describe the names of each state. alphabet is a list of objects that indicate the allowed outputs. training_data is a list of (outputs, observed states) where outputs is a list of the emission from the alphabet, and observed states is a list of states from states. pseudo_initial, pseudo_transition, and pseudo_emission are optional parameters that you can use to assign pseudo-counts to different matrices. They should be matrices of the appropriate size that contain numbers to add to each parameter matrix. NrLrMrNzstates and outputs not alignedcrOrrrE)states_indexesrrr3r4z!train_visible..crOrrrE)outputs_indexesrrr3r4) rr6rVrWr'rrX_mler)rrr[rRrSrTr:r;training_statesr\toutputststatesrFrrrr)rrr train_visibles@       rcCst|}|r ||}|D] }||dd7<q t|}t||f} |r+| |} |D]"}tt|dD]} || || d} } | | | fd7<q7q-tt| D]} | | ddft| | ddf| | ddf<qVt||f} |r~| |} t||f} t||D]\}}t||D]\}}| ||fd7<qqtt| D]} | | ddft| | ddf| | ddf<q|| | fS)z.csg|]}j|qSr)rrE)r9rrr36rb) rrr6rerVERY_SMALL_NUMBERrrrr_viterbir8r`) markov_modeloutputr:rkrlrmresultsr<rscorer)rQr9r find_states s   "rcCst|}g}t|D] }|dg|q t||f}||dd|df|dddf<td|D]9} || } t|D].} |dd| df|dd| f|| | f} t| } | | d|| | f<| || | <q=q3g}g}t|dd|df} | D]}||d|g|||dfq|r|\} }}| dkr|||fn||d| } | D]}|| d|g||fq|s|S)zSImplement Viterbi algorithm to find most likely states for a given input (PRIVATE).Nrr)rr8rXr6r7rpop)r:rkrlrmrrx backtracer<scoresr|r}ri_scoresrQ in_processrrrrrrr:s8 $ 0 $rcCszt|jdkr|t|}|St|jdkr9tt|D]}||ddft||ddf||ddf<q|Std)z"Normalize matrix object (PRIVATE).rNz&I cannot handle matrixes of that shape)rrWrr8r')rr<rrrrcs 2rcCst|}t|S)z%Normalize a uniform matrix (PRIVATE).)r6rrrWrrrr _uniform_normps rcCstj|}t|S)z$Normalize a random matrix (PRIVATE).)r6randomrrrrrrcvs rccCstj|dd}|j|krtdt|jdkr(tt|ddkr&td|St|jdkrMtt|D]}tt||ddkrJtd|q5|Std ) zFCopy a matrix and check its dimension. Normalize at the end (PRIVATE).r)copyzIncorrect dimensiong?g{Gz?zmatrix not normalized to 1.0rzmatrix %d not normalized to 1.0z&I don't handle matrices > 2 dimensions)r6arrayrWr'rrirr8)r desired_shaper<rrrrd|s  rdcCsFt|jdkrt|t|jf}n|}t}|D]}t||}q|S)z/Implement logsum for a matrix object (PRIVATE).r)rrWr6reshapeprodrgrw)rvecrnumrrrrus rucCsRt|t|ks Jdtt|}tt|D] }t||||||<q|S)z5Implement a log sum for two vector objects (PRIVATE).zvectors aren't the same length)rr6r7r8rw)logvec1logvec2sumvecr<rrrrvs rvcCst|}t|S)z-Return the exponential of a logsum (PRIVATE).)rur6r`)numbersrrrr _exp_logsums r)NNNN)NNNNNNNr rB)&r$warningsBiorwarnnumpyr6 ImportErrorrrwrrseedrrergrr*r=rr_rfrZrhrsrtrrrrrrrrcrdrurvrrrrrst      % E J^ ? .)