a Kbe+@sjddlZddlmZGdddZGdddeZddZd d Zd d Zed ddgZ GdddZ dS)N) namedtuplec@steZdZdZdddZeddZeddZd d Zd d Z eee Z d dZ ddZ ddZ ddZddZdS) AlignedSenta# Return an aligned sentence object, which encapsulates two sentences along with an ``Alignment`` between them. Typically used in machine translation to represent a sentence and its translation. >>> from nltk.translate import AlignedSent, Alignment >>> algnsent = AlignedSent(['klein', 'ist', 'das', 'Haus'], ... ['the', 'house', 'is', 'small'], Alignment.fromstring('0-3 1-2 2-0 3-1')) >>> algnsent.words ['klein', 'ist', 'das', 'Haus'] >>> algnsent.mots ['the', 'house', 'is', 'small'] >>> algnsent.alignment Alignment([(0, 3), (1, 2), (2, 0), (3, 1)]) >>> from nltk.corpus import comtrans >>> print(comtrans.aligned_sents()[54]) 'So why should EU arm...'> >>> print(comtrans.aligned_sents()[54].alignment) 0-0 0-1 1-0 2-2 3-4 3-5 4-7 5-8 6-3 7-9 8-9 9-10 9-11 10-12 11-6 12-6 13-13 :param words: Words in the target language sentence :type words: list(str) :param mots: Words in the source language sentence :type mots: list(str) :param alignment: Word-level alignments between ``words`` and ``mots``. Each alignment is represented as a 2-tuple (words_index, mots_index). :type alignment: Alignment NcCs:||_||_|dur tg|_nt|tus0J||_dSN)_words_mots Alignment alignmenttype)selfwordsmotsrr 1lib/python3.9/site-packages/nltk/translate/api.py__init__/s  zAlignedSent.__init__cCs|jSr)rr r r rr 8szAlignedSent.wordscCs|jSr)rrr r rr <szAlignedSent.motscCs|jSr) _alignmentrr r r_get_alignment@szAlignedSent._get_alignmentcCs"tt|jt|j|||_dSr)_check_alignmentlenr r r)r rr r r_set_alignmentCszAlignedSent._set_alignmentcCsNdddd|jD}dddd|jD}d|d|d|jdS)z_ Return a string representation for this ``AlignedSent``. :rtype: str z[%s]z, css|]}d|VqdSz'%s'Nr .0wr r r Oz'AlignedSent.__repr__..css|]}d|VqdSrr rr r rrPrz AlignedSent())joinrrr)r r r r r r__repr__IszAlignedSent.__repr__cCs:d}|d7}|jD]}|d|d|d7}q|jD]}|d|d|d7}q4|jD]*\}}|d|j|d|j|d7}qVtt|jd D]$}|d |j||j|d 7}qtt|jd D]$}|d |j||j|d 7}q|d d dd|jD7}|d d dd|jD7}|d7}|S)z< Dot representation of the aligned sentence zgraph align { znode[shape=plaintext] "z_source" [label="z"] z_target" [label="z _source" -- "z _target" z)"{}_source" -- "{}_source" [style=invis] z)"{}_target" -- "{}_target" [style=invis] z{rank = same; %s}  css|]}d|VqdS)z "%s_source"Nr rr r rrurz&AlignedSent._to_dot..css|]}d|VqdS)z "%s_target"Nr rr r rrvr})rrrrangerformatr)r sruvir r r_to_dotTs,  $  zAlignedSent._to_dotc Cs~|d}d}z$tjdd|gtjtjtjd}Wn.tyd}ztd|WYd}~n d}~00||\}}|dS)zR Ipython magic : show SVG representation of this ``AlignedSent``. utf8Zsvgdotz-T%s)stdinstdoutstderrz0Cannot find the dot binary from Graphviz packageN) r)encode subprocessPopenPIPEOSError ExceptionZ communicatedecode)r Z dot_stringZ output_formatZprocesseouterrr r r _repr_svg_|s   zAlignedSent._repr_svg_cCsBd|jddd}d|jddd}d|d|dS)zn Return a human-readable string representation for this ``AlignedSent``. :rtype: str r!Nz...z 'z'>)rrr)r sourcetargetr r r__str__szAlignedSent.__str__cCst|j|j|jS)zm Return the aligned sentence pair, reversing the directionality :rtype: AlignedSent )rrrrinvertrr r rr>szAlignedSent.invert)N)__name__ __module__ __qualname____doc__rpropertyr r rrrrr)r9r=r>r r r rrs     ( rc@sVeZdZdZddZeddZddZdd Zdd d Z d dZ ddZ ddZ d S)rac A storage class for representing alignment between two sequences, s1, s2. In general, an alignment is a set of tuples of the form (i, j, ...) representing an alignment between the i-th element of s1 and the j-th element of s2. Tuples are extensible (they might contain additional data, such as a boolean to indicate sure vs possible alignments). >>> from nltk.translate import Alignment >>> a = Alignment([(0, 0), (0, 1), (1, 2), (2, 2)]) >>> a.invert() Alignment([(0, 0), (1, 0), (2, 1), (2, 2)]) >>> print(a.invert()) 0-0 1-0 2-1 2-2 >>> a[0] [(0, 1), (0, 0)] >>> a.invert()[2] [(2, 1), (2, 2)] >>> b = Alignment([(0, 0), (0, 1)]) >>> b.issubset(a) True >>> c = Alignment.fromstring('0-0 0-1') >>> b == c True cCs:t||}|tgkr*tdd|Dnd|_d|_|S)Ncss|]}|dVqdSrNr rpr r rrrz$Alignment.__new__..r) frozenset__new__max_len_index)clspairsr r r rrHs $zAlignment.__new__cCstdd|DS)a Read a giza-formatted string and return an Alignment object. >>> Alignment.fromstring('0-0 2-1 9-2 21-3 10-4 7-5') Alignment([(0, 0), (2, 1), (7, 5), (9, 2), (10, 4), (21, 3)]) :type s: str :param s: the positional alignments in giza format :rtype: Alignment :return: An Alignment object corresponding to the string representation ``s``. cSsg|] }t|qSr ) _giza2pair)rar r r rz(Alignment.fromstring..)rsplit)rLr%r r r fromstringszAlignment.fromstringcCs|js||j|S)zN Look up the alignments that map from a given index or slice. )rK _build_index __getitem__)r keyr r rrTszAlignment.__getitem__cCstdd|DS)zI Return an Alignment object, being the inverted mapping. css*|]"}|d|df|ddVqdS)r rNr rEr r rrrz#Alignment.invert..)rrr r rr>szAlignment.invertNcCsVt}|js||s*ttt|j}|D]}|dd|j|Dq.t|S)z Work out the range of the mapping from the given positions. If no positions are specified, compute the range of the entire mapping. css|]\}}|VqdSrr )r_fr r rrrz"Alignment.range..)setrKrSlistr#rupdatesorted)r Z positionsZimagerFr r rr#szAlignment.rangecCs dt|S)M Produce a Giza-formatted string representing the alignment. z Alignment(%r))r\rr r rrszAlignment.__repr__cCsdddt|DS)r]r!css|]}d|ddVqdS)z%d-%dNrVr rEr r rrrz$Alignment.__str__..)rr\rr r rr=szAlignment.__str__cCs<ddt|jdD|_|D]}|j|d|qdS)z Build a list self._index such that self._index[i] is a list of the alignments originating from word i. cSsg|]}gqSr r )rrWr r rrPrz*Alignment._build_index..r rN)r#rJrKappend)r rFr r rrSszAlignment._build_index)N) r?r@rArBrH classmethodrRrTr>r#rr=rSr r r rrs  rcCs|d\}}t|t|fSN-rQint) pair_stringr(jr r rrNsrNcCs |d\}}}t|t|fSr`rb)rdr(rerFr r r _naacl2pair srfcsPt|tusJtfdd|Ds.tdtfdd|DsLtddS)ab Check whether the alignments are legal. :param num_words: the number of source language words :type num_words: int :param num_mots: the number of target language words :type num_mots: int :param alignment: alignment to be checked :type alignment: Alignment :raise IndexError: if alignment falls outside the sentence c3s*|]"}d|dkoknVqdSrDr rZpair) num_wordsr rrrz#_check_alignment..z&Alignment is outside boundary of wordsc3s6|].}|ddup,d|dko(knVqdS)r Nrr rg)num_motsr rrrz%Alignment is outside boundary of motsN)r rall IndexError)rhrirr )rirhrrs rPhraseTableEntry trg_phraselog_probc@s0eZdZdZddZddZddZdd Zd S) PhraseTablezs In-memory store of translations for a given phrase, and the log probability of the those translations cCs t|_dSr)dict src_phrasesrr r rr,szPhraseTable.__init__cCs |j|S)a Get the translations for a source language phrase :param src_phrase: Source language phrase of interest :type src_phrase: tuple(str) :return: A list of target language phrases that are translations of ``src_phrase``, ordered in decreasing order of likelihood. Each list element is a tuple of the target phrase and its log probability. :rtype: list(PhraseTableEntry) rqr src_phraser r rtranslations_for/s zPhraseTable.translations_forcCsLt||d}||jvr g|j|<|j|||j|jdddddS)z :type src_phrase: tuple(str) :type trg_phrase: tuple(str) :param log_prob: Log probability that given ``src_phrase``, ``trg_phrase`` is its translation :type log_prob: float )rmrncSs|jSr)rn)r6r r rKrz!PhraseTable.add..T)rUreverseN)rlrqr^sort)r rtrmrnentryr r radd>s   zPhraseTable.addcCs ||jvSrrrrsr r r __contains__MszPhraseTable.__contains__N)r?r@rArBrrurzr{r r r rro&s ro) r0 collectionsrrrGrrNrfrrlror r r r s b