B >?[*@sddlmZmZddlZddlmZddlmZeGdddeZ eGddde Z d d Z d d Z d dZedddgZGdddeZdS))print_functionunicode_literalsN) namedtuple)python_2_unicode_compatiblec@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:||_||_|dkr tg|_nt|tks0t||_dS)N)_words_mots Alignment alignmenttypeAssertionError)selfwordsmotsr r1lib/python3.7/site-packages/nltk/translate/api.py__init__3s  zAlignedSent.__init__cCs|jS)N)r)r rrrr<szAlignedSent.wordscCs|jS)N)r)r rrrr@szAlignedSent.motscCs|jS)N) _alignment)r rrr_get_alignmentDszAlignedSent._get_alignmentcCs"tt|jt|j|||_dS)N)_check_alignmentlenrrr)r r rrr_set_alignmentGszAlignedSent._set_alignmentcCsDdddd|jD}dddd|jD}d|||jfS)z_ Return a string representation for this ``AlignedSent``. :rtype: str z[%s]z, css|]}d|VqdS)z'%s'Nr).0wrrr Ssz'AlignedSent.__repr__..css|]}d|VqdS)z'%s'Nr)rrrrrrTszAlignedSent(%s, %s, %r))joinrrr)r rrrrr__repr__MszAlignedSent.__repr__cCs<d}|d7}x|jD]}|d||f7}qWx|jD]}|d||f7}q4Wx.|jD]$\}}|d|j||j|f7}qTWx:tt|jdD]$}|d|j||j|df7}qWx:tt|jdD]$}|d|j||j|df7}qW|d d d d |jD7}|d d d d |jD7}|d7}|S)z< Dot representation of the aligned sentence zgraph align { znode[shape=plaintext] z"%s_source" [label="%s"] z"%s_target" [label="%s"] z"%s_source" -- "%s_target" z)"%s_source" -- "%s_source" [style=invis] z)"%s_target" -- "%s_target" [style=invis] z{rank = same; %s}  css|]}d|VqdS)z "%s_source"Nr)rrrrrrysz&AlignedSent._to_dot..css|]}d|VqdS)z "%s_target"Nr)rrrrrrzs})rrrrangerr)r sruvirrr_to_dotXs(   zAlignedSent._to_dotcCsl|d}d}y$tjdd|gtjtjtjd}Wntk rRtdYnX||\}}|dS)zR Ipython magic : show SVG representation of this ``AlignedSent``. utf8Zsvgdotz-T%s)stdinstdoutstderrz0Cannot find the dot binary from Graphviz package) r%encode subprocessPopenPIPEOSError ExceptionZ communicatedecode)r Z dot_stringZ output_formatZprocessouterrrrr _repr_svg_s zAlignedSent._repr_svg_cCs<d|jddd}d|jddd}d||fS)zn Return a human-readable string representation for this ``AlignedSent``. :rtype: str rNz...z '%s'>)rrr)r sourcetargetrrr__str__szAlignedSent.__str__cCst|j|j|jS)zm Return the aligned sentence pair, reversing the directionality :rtype: AlignedSent )rrrrinvert)r rrrr9szAlignedSent.invert)N)__name__ __module__ __qualname____doc__rpropertyrrrrr rr%r4r8r9rrrrrs    ( rc@sVeZdZdZddZeddZddZdd Zdd d Z d dZ ddZ ddZ d S)r ac 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|]}|dVqdS)rNr)rprrrrsz$Alignment.__new__..r) frozenset__new__max_len_index)clsZpairsr rrrrAs $zAlignment.__new__cCstdd|DS)a Read a giza-formatted string and return an Alignment object. >>> Alignment.fromstring('0-0 2-1 9-2 21-3 10-4 7-5') Alignment([(0, 0), (2, 1), (7, 5), (9, 2), (10, 4), (21, 3)]) :type s: str :param s: the positional alignments in giza format :rtype: Alignment :return: An Alignment object corresponding to the string representation ``s``. cSsg|] }t|qSr) _giza2pair)rarrr sz(Alignment.fromstring..)r split)rEr!rrr fromstringszAlignment.fromstringcCs|js||j|S)zN Look up the alignments that map from a given index or slice. )rD _build_index __getitem__)r keyrrrrLszAlignment.__getitem__cCstdd|DS)zI Return an Alignment object, being the inverted mapping. css*|]"}|d|df|ddVqdS)rrNr)rr?rrrrsz#Alignment.invert..)r )r rrrr9szAlignment.invertNcCsZt}|js||s*ttt|j}x&|D]}|dd|j|Dq0Wt|S)z Work out the range of the mapping from the given positions. If no positions are specified, compute the range of the entire mapping. css|]\}}|VqdS)Nr)r_frrrrsz"Alignment.range..)setrDrKlistr rupdatesorted)r Z positionsZimager?rrrr s zAlignment.rangecCs dt|S)zM Produce a Giza-formatted string representing the alignment. z Alignment(%r))rT)r rrrrszAlignment.__repr__cCsdddt|DS)zM Produce a Giza-formatted string representing the alignment. rcss|]}d|ddVqdS)z%d-%dNrNr)rr?rrrrsz$Alignment.__str__..)rrT)r rrrr8szAlignment.__str__cCs@ddt|jdD|_x |D]}|j|d|q WdS)z Build a list self._index such that self._index[i] is a list of the alignments originating from word i. cSsg|]}gqSrr)rrOrrrrHsz*Alignment._build_index..rrN)r rCrDappend)r r?rrrrKs zAlignment._build_index)N) r:r;r<r=rA classmethodrJrLr9r rr8rKrrrrr s  r cCs|d\}}t|t|fS)N-)rIint) pair_stringr$jrrrrF srFcCs |d\}}}t|t|fS)NrW)rIrX)rYr$rZr?rrr _naacl2pairsr[csPt|tksttfdd|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|dkoknVqdS)rNr)rpair) num_wordsrrr"sz#_check_alignment..z&Alignment is outside boundary of wordsc3s6|].}|ddkp,d|dko(knVqdS)rNrr)rr\)num_motsrrr$sz%Alignment is outside boundary of motsN)r r r all IndexError)r]r^r r)r^r]rrs 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|_dS)N)dict src_phrases)r rrrr1szPhraseTable.__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) )rf)r src_phraserrrtranslations_for4s zPhraseTable.translations_forcCsLt||d}||jkr 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 )rbrccSs|jS)N)rc)errrPsz!PhraseTable.add..T)rMreverseN)rarfrUsort)r rgrbrcentryrrraddCs   zPhraseTable.addcCs ||jkS)N)rf)r rgrrr __contains__RszPhraseTable.__contains__N)r:r;r<r=rrhrnrorrrrrd+s rd)Z __future__rrr, collectionsrZ nltk.compatrobjectrr@r rFr[rrardrrrr s    b