+ j Rt^RIt^RIHt^RIHtHt]P !]4t!RR]4t R] /t RR/t R^/t R#)z+XNLI utils (dataset loading and evaluation)N)logging) DataProcessor InputExamplec@a]tRt^toRtRRltRtRtRtRt Vt R#) XnliProcessorz Processor for the XNLI dataset. Adapted from https://github.com/google-research/bert/blob/f39e881b169b9d53bea03d2d341b31707a6c052b/run_classifier.py#L207 NcWnW nR#Nlanguagetrain_language)selfr r s&&&b/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/data/processors/xnli.py__init__XnliProcessor.__init__ s  ,c  lVPf VPM VPpVP\PP VRV R244p.p\ V4FwrVV^8XdKRV 2pV^,pV^,p V^,R8XdRMV^,p \V\4'g\RV R24h\V \4'g\RV R24h\V \4'g\RV R24hVP\WxWR 74K V#) See base class.z$XNLI-MT-1.0/multinli/multinli.train.z.tsvztrain- contradictory contradictionTraining input  is not a stringTraining label guidtext_atext_blabel) r r _read_tsvospathjoin enumerate isinstancestr TypeErrorappendr) r data_dirlglinesexamplesilinerrrrs && r get_train_examples XnliProcessor.get_train_examples$s"119T]]t?R?Rrww||H8\]_\``d6efg 'GAAvA3rGs_2 .   H %8:M8:x M    Ar