j$PdZ ddlZn #e$rYnwxYwd dZdZdZd d Zdd ZdS)a  Text Segmentation Metrics 1. Windowdiff Pevzner, L., and Hearst, M., A Critique and Improvement of an Evaluation Metric for Text Segmentation, Computational Linguistics 28, 19-36 2. Generalized Hamming Distance Bookstein A., Kulyukin V.A., Raita T. Generalized Hamming Distance Information Retrieval 5, 2002, pp 353-375 Baseline implementation in C++ http://digital.cs.usu.edu/~vkulyukin/vkweb/software/ghd/ghd.html Study describing benefits of Generalized Hamming Distance Versus WindowDiff for evaluating text segmentation tasks Begsten, Y. Quel indice pour mesurer l'efficacite en segmentation de textes ? TALN 2009 3. Pk text segmentation metric Beeferman D., Berger A., Lafferty J. (1999) Statistical Models for Text Segmentation Machine Learning, 34, 177-210 N1Fct|t|krtd|dkrtd|t|krtdd}|d||}|d||}tt||z dzD]}|dkrJ||||zdz |k||dz |kz z }||||zdz |k||dz |kz z }t ||z } |r|| z }l|t d| z }|t||z dzz S)a Compute the windowdiff score for a pair of segmentations. A segmentation is any sequence over a vocabulary of two items (e.g. "0", "1"), where the specified boundary value is used to mark the edge of a segmentation. From Pevzner & Hearst (2002), the WindowDiff metric is defined as:: WindowDiff(ref, hyp, k) = 1 / (N - k) * sum_{i=1}^{N-k} ( |b(ref, i, i+k) - b(hyp, i, i+k)| > 0 ) where ``b(seg, i, j)`` counts the number of boundaries in ``seg`` between positions ``i`` and ``j``, and ``N = len(seg)``. The weighted variant sums the absolute differences instead of thresholding at 1. >>> s1 = "000100000010" >>> s2 = "000010000100" >>> s3 = "100000010000" >>> '%.2f' % windowdiff(s1, s1, 3) '0.00' >>> '%.2f' % windowdiff(s1, s2, 3) '0.30' >>> '%.2f' % windowdiff(s2, s3, 3) '0.80' :param seg1: a segmentation :type seg1: str or list :param seg2: a segmentation :type seg2: str or list :param k: window width :type k: int :param boundary: boundary value :type boundary: str or int or bool :param weighted: use the weighted variant of windowdiff :type weighted: boolean :rtype: float !Segmentations have unequal lengthr%Window width k should not be negativezCWindow width k should be smaller or equal than segmentation lengthsN?)len ValueErrorcountrangeabsmin) seg1seg2kboundaryweightedwdcount1count2indiffs |/home/lighthouse/badge-speaker-benchmark/.benchmark-venv-community/lib/python3.11/site-packages/nltk/metrics/segmentation.py windowdiffr1sV 4yyCII<===1uu@AAA3t99}} Q    B "1"X^^H % %F "1"X^^H % %F 3t99q=1$ % %    q55 tAEAI(2tAE{h7NO OF tAEAI(2tAE{h7NO OFFVO$$  %KBB #a-- BB TQ$ %%ctj||f}|tj|z|dddf<|tj|z|dddf<|S)Nr)npemptyarange)nrowsncolsins_costdel_costmats r _init_matr%|sZ (E5> " "C29U+++C111I29U+++C1I Jrc@t|D]\}}t|D]x\}} |t|| z z|||fz} || kr |||f} n'|| kr||||dzfz} n|||dz|fz} t| | ||dz|dzf<ydS)Nr) enumerater r) r$rowvcolvr"r#shift_cost_coeffrrowijcolj shift_costtcosts r_ghd_auxr0sT?? 7 74  7 7GAt)Ct ,<,<>> # Same examples as Kulyukin C++ implementation >>> ghd('1100100000', '1100010000', 1.0, 1.0, 0.5) 0.5 >>> ghd('1100100000', '1100000001', 1.0, 1.0, 0.5) 2.0 >>> ghd('011', '110', 1.0, 1.0, 0.5) 1.0 >>> ghd('1', '0', 1.0, 1.0, 0.5) 1.0 >>> ghd('111', '000', 1.0, 1.0, 0.5) 3.0 >>> ghd('000', '111', 1.0, 2.0, 0.5) 6.0 :param ref: the reference segmentation :type ref: str or list :param hyp: the hypothetical segmentation :type hyp: str or list :param ins_cost: insertion cost :type ins_cost: float :param del_cost: deletion cost :type del_cost: float :param shift_cost_coeff: constant used to compute the cost of a shift. ``shift cost = shift_cost_coeff * |i - j|`` where ``i`` and ``j`` are the positions indicating the shift :type shift_cost_coeff: float :param boundary: boundary value :type boundary: str or int or bool :rtype: float c&g|] \}}|k |Sr4.0rvalrs r zghd.."CCCXa3(??q???rc&g|] \}}|k |Sr4r4r5s rr8zghd..r9rrgr)r;)r'r r%r0float) refhypr"r#r*rref_idxhyp_idx nref_bound nhyp_boundr$s ` rghdrCs\DCCC3CCCGCCCC3CCCGWJWJQ:??s aJ!OOH$$ qZ!^^H$$ JNJNHh G GC S'7Hh8HIII V  rc t|t|krtd|Pttt|t ||ddzz }|dkrtdd}|d||}|d||}t t||z dzD]i}|dkrJ||||zdz |k||dz |kz z }||||zdz |k||dz |kz z }|dk}|dk} || kr|dz }j|t||z dzz S)a Compute the Pk metric for a pair of segmentations A segmentation is any sequence over a vocabulary of two items (e.g. "0", "1"), where the specified boundary value is used to mark the edge of a segmentation. >>> '%.2f' % pk('0100'*100, '1'*400, 2) '0.50' >>> '%.2f' % pk('0100'*100, '0'*400, 2) '0.50' >>> '%.2f' % pk('0100'*100, '0100'*100, 2) '0.00' :param ref: the reference segmentation :type ref: str or list :param hyp: the segmentation to evaluate :type hyp: str or list :param k: window size, if None, set to half of the average reference segment length :type boundary: str or int or bool :param boundary: boundary value :type boundary: str or int or bool :rtype: float rNrr1rrr)r r introundmaxr r ) r=r>rrerr ref_count hyp_countrrhs rpkrMs2 3xx3s88<===y c#hh#cii&9&91"="="CDEE F F1uu@AAA C BQB h''IBQB h''I 3s88arQs@    D E&E&E&E&V 7 7 7 ====F5&5&5&5&5&5&s