zh, en = zh.strip(), en.strip()
zh = tools.zhsent_preprocess(zh)
with contextlib.redirect_stdout(outf):
print(zh)
print(en)
print()
sbc4_zh = gentask.slice_lines_grouped_by_n('sbc4_zh', sbc4(),
target_dir / 'ch.txt',
n=3,
s=0)
sbc4_tag = gentask.slice_lines_grouped_by_n('sbc4_tag', sbc4(),
target_dir / 'tag.txt',
n=3,
s=1)
sbc4_tag_lm = gentask.lm(
'sbc4_tag_lm', sbc4_tag, target_dir / 'tag.lm', target_dir / 'tag.blm')
sbc4_zhpreprocess_slash = gentask.transformat_line2slash(
'sbc4_zhpreprocess_slash', sbc4_zhpreprocess(),
target_dir / 'zhpreprcs.slash.txt')
sbc4_tok_tag_tm = gentask.phrasetable(
'sbc4_tok_tag_phrasetable', sbc4_zhpreprocess(),
target_dir / 'toktag.phrasetable.h5')
if __name__ == "__main__":
luigi.run(local_scheduler=True)
sbc4_test = gentask.slice_lines_grouped_by_n(
'sbc4_test', sbc4(), base_dir / 'sbc4_test.txt', n=3 * 10, s=slice(3 * 9, 3 * 10))
sbc4_test_zh = gentask.slice_lines_grouped_by_n(
'sbc4_test_zh', sbc4_test(), base_dir / 'sbc4_test.zh.txt', n=3, s=0)
sbc4_test_zh_untok = gentask.untok(
'sbc4_test_zh_untok', sbc4_test_zh(), base_dir / 'sbc4_test.zh.untok.txt')
sbc4_train_tag = gentask.slice_lines_grouped_by_n(
'sbc4_train_tag', sbc4_train(), base_dir / 'sbc4_train.tag.txt', n=3, s=1)
sbc4_train_tag_lm = gentask.lm(
'sbc4_train_tag_lm', sbc4_train_tag(), base_dir / 'sbc4_train.tag.lm', base_dir / 'sbc4_train.tag.blm')
sbc4_train_zh_to_tok_tag_phrasetable = gentask.phrasetable(
'sbc4_train_zh_to_tok_tag_phrasetable', sbc4_train(), base_dir / 'sbc4_train.zh2toktag.phrasetable.h5')
sbc4_train_toktag_sbc4_test = gentask.zhtoktag(
'sbc4_train_toktag_sbc4_test', sbc4_test_zh_untok(), base_dir / 'sbc4_test.zh.untok.tok.txt', tm=sbc4_train_zh_to_tok_tag_phrasetable(), lm=sbc4_train_tag_lm())
sbc4_test_slash = gentask.transformat_line2slash(
'sbc4_test_slash', sbc4_test(), base_dir / 'sbc4_test.slash.txt')
wdiff_sbc4_test = gentask.word_diff('wdiff_sbc4_test',
sbc4_test_slash(), sbc4_train_toktag_sbc4_test(), base_dir / 'sbc4-test.wdiff')