o eic @s$dZddlmZGdddeZdS)zLibrary for computing the BLEU score based on SacreBLEU SacreBLEU github: https://github.com/mjpost/sacrebleu Authors * Titouan Parcollet 2025 * Mirco Ravanelli 2021 ) MetricStatsc@s4eZdZdZd ddZddZd dd Zd d ZdS) BLEUStatsa+A class for tracking corpus-level BLEU (https://www.aclweb.org/anthology/P02-1040.pdf). Each hypothesis can be matched against one or multiple references. Arguments --------- max_ngram_order: int, default 4 The maximum length of the ngrams to use for BLEU scoring. Default is 4. Example ------- >>> bleu = BLEUStats() >>> bleu.append( ... ids=['utterance1', 'utterance2'], ... predict=[ ... 'The dog bit the man.', ... 'It was not surprising.'], ... targets=[ ... ['The dog bit the man.', 'It was not unexpected.'], ... ['The dog had bit the man.', 'No one was surprised.'] ... ] ... ) >>> stats = bleu.summarize() >>> stats['BLEU'] 74.19446627365011 cCsPzddlm}Wn tytdYnw|||d|_g|_d|_dS)Nr)BLEUz^Please install sacrebleu (https://pypi.org/project/sacrebleu/) in order to use the BLEU metric)max_ngram_order)sacrebleu.metricsr ImportErrorprintclearbleupredictstargets)selfrrrT/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/speechbrain/utils/bleu.py__init__'s   zBLEUStats.__init__cCsn|j||j||jdur||_dSt|jt|ks!Jtt|jD] }|j|||q(dS)aAdd stats to the relevant containers. * See MetricStats.append() Arguments --------- ids : list List of ids corresponding to utterances. predict : list[str] A str which represent the hypotheses. Of dimension [nb_hypotheses] targets : list[list[str]] List of list of reference. The dimensions are as follow: [nb_references, nb_hypotheses]. N)idsextendr r lenrange)rrpredictr irrrappend6s    zBLEUStats.appendNcCs|j|j|j}i}|j|d<|j|d<|j|j|d<|j|d<|j|d<|j|d<||_ ||_ |j d|j d<|durC|j |S|j S) zaSummarize the BLEU and return relevant statistics. * See MetricStats.summarize() rBPratiohyp_lenref_len precisions bleu_scoreN) r corpus_scorer r scorebpsys_lenrrscoressummary)rfieldr#detailsrrr summarizeNs      zBLEUStats.summarizecCs |js|t|j|ddS)zoWrite all relevant info (e.g., error rate alignments) to file. * See MetricStats.write_stats() )fileN)r$r'r r#)r filestreamrrr write_statsfszBLEUStats.write_stats)r)N)__name__ __module__ __qualname____doc__rrr'r*rrrrr s    rN)r.speechbrain.utils.metric_statsrrrrrrs