Files
b2txt25/language_model/srilm-1.7.3/flm/test/tests/fngram-count-vocab/go.make-vocab
2025-07-02 12:18:09 -07:00

7 lines
191 B
Bash
Executable File

#!/bin/sh
gunzip -c -f ../fngram-count/ch_lm_train100.noamp.decomposed.txt | \
sed 's,:, ,g' | \
ngram-count -text - -write1 - -sort | \
${GAWK-gawk} '$2 > 1 { print $1 }' > train-gt1.vocab