mt-summit-corpora/prepare-dataset.sh

17 lines
898 B
Bash
Raw Normal View History

2021-11-29 21:42:16 +01:00
#!/bin/bash
tar xvf mt-summit-corpora.tar.gz
cut -f1 mt-summit-corpora/train/train.en-pl > mt-summit-corpora/train/train.en
cut -f2 mt-summit-corpora/train/train.en-pl > mt-summit-corpora/train/train.pl
cut -f1 mt-summit-corpora/dev/dev.en-pl > mt-summit-corpora/dev/dev.en
cut -f2 mt-summit-corpora/dev/dev.en-pl > mt-summit-corpora/dev/dev.pl
cut -f1 mt-summit-corpora/test/constrained-scenario-test.tsv > mt-summit-corpora/test/constrained-scenario-test.en
cut -f2 mt-summit-corpora/test/constrained-scenario-test.tsv > mt-summit-corpora/test/constrained-scenario-test.pl
cut -f2 mt-summit-corpora/test/general-scenario-test.tsv > mt-summit-corpora/test/general-scenario-test.pl
cut -f1 mt-summit-corpora/test/general-scenario-test.tsv > mt-summit-corpora/test/general-scenario-test.en
rm mt-summit-corpora/train/train.en-pl
rm mt-summit-corpora/test/*.tsv
rm mt-summit-corpora/dev/dev.en-pl