base 892 | lin 446 | sna intouches 446 Loading the LM will be faster if you build a binary file. Reading /scratch/lm/lin_5g.arpa ----5---10---15---20---25---30---35---40---45---50---55---60---65---70---75---80---85---90---95--100 **************************************************************************************************** Found entries of length > 1 in alphabet. This is unusual unless style is BPE, but the alphabet was not recognized as BPE type. Is this correct? beam principal : 446 clips Loading weights: 0%| | 0/1096 [00:00 1 in alphabet. This is unusual unless style is BPE, but the alphabet was not recognized as BPE type. Is this correct? donneur mms1b_lin_best : 2225 hypotheses NOUVELLES Loading weights: 0%| | 0/797 [00:00 1 in alphabet. This is unusual unless style is BPE, but the alphabet was not recognized as BPE type. Is this correct? donneur rebuild_best : 2084 hypotheses NOUVELLES POOL EXTERNE TOTAL : 4309 hypotheses sur 446 clips (9.7/clip) Loading the LM will be faster if you build a binary file. Reading /scratch/lm/lin_5g.arpa ----5---10---15---20---25---30---35---40---45---50---55---60---65---70---75---80---85---90---95--100 **************************************************************************************************** calibration : ecart-type residuel median 1.050 (offset median -38.7) Loading weights: 0%| | 0/424 [00:00