2017-07-21 17:37:00 +02:00
|
|
|
SRC_LANG=pl
|
|
|
|
|
TRG_LANG=en
|
|
|
|
|
CORPUS_NAME=europarl_sample
|
|
|
|
|
DICTIONARY_NAME=classyf_popular_medicine
|
|
|
|
|
SEPARATOR=@\#@
|
|
|
|
|
CORPUS_CHUNK_SIZE=100000
|
|
|
|
|
|
|
|
|
|
all: corpora/$(CORPUS_NAME)/src_filtered.txt corpora/$(CORPUS_NAME)/trg_filtered.txt
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/src_filtered.txt corpora/$(CORPUS_NAME)/trg_filtered.txt: corpora/$(CORPUS_NAME)/src_clean.txt corpora/$(CORPUS_NAME)/trg_clean.txt corpora/$(CORPUS_NAME)/corpus_lines.txt
|
|
|
|
|
./compile.py corpora/$(CORPUS_NAME)/src_clean.txt corpora/$(CORPUS_NAME)/trg_clean.txt corpora/$(CORPUS_NAME)/corpus_lines.txt corpora/$(CORPUS_NAME)/src_filtered.txt corpora/$(CORPUS_NAME)/trg_filtered.txt
|
|
|
|
|
|
2017-07-21 17:55:28 +02:00
|
|
|
corpora/$(CORPUS_NAME)/corpus_lines.txt: corpora/$(CORPUS_NAME)/corpus.indexed dictionaries/$(DICTIONARY_NAME).lem
|
2017-07-21 17:37:00 +02:00
|
|
|
./get_corpus_lines.py dictionaries/$(DICTIONARY_NAME).lem corpora/$(CORPUS_NAME)/report.txt > $@
|
|
|
|
|
|
2017-07-21 17:55:28 +02:00
|
|
|
corpora/$(CORPUS_NAME)/corpus.indexed: corpora/$(CORPUS_NAME)/corpus.split
|
2017-07-21 17:37:00 +02:00
|
|
|
./load_corpus.sh corpora/$(CORPUS_NAME)/csv/
|
2017-07-21 17:55:28 +02:00
|
|
|
touch corpora/$(CORPUS_NAME)/corpus.indexed
|
2017-07-21 17:37:00 +02:00
|
|
|
|
2017-07-21 17:55:28 +02:00
|
|
|
corpora/$(CORPUS_NAME)/corpus.split: corpora/$(CORPUS_NAME)/src.csv
|
2017-07-21 17:37:00 +02:00
|
|
|
mkdir corpora/$(CORPUS_NAME)/csv
|
|
|
|
|
split -l $(CORPUS_CHUNK_SIZE) -d --additional-suffix=".csv" $< corpora/$(CORPUS_NAME)/csv/src
|
2017-07-21 17:55:28 +02:00
|
|
|
touch corpora/$(CORPUS_NAME)/corpus.split
|
2017-07-21 17:37:00 +02:00
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/src.csv: corpora/$(CORPUS_NAME)/src_clean.lem
|
|
|
|
|
./lem2csv.py $< > $@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
dictionaries/$(DICTIONARY_NAME).lem: dictionaries/$(DICTIONARY_NAME).txt
|
|
|
|
|
mono ../LemmaGenSentenceLemmatizer/LemmaGenSentenceLemmatizer/bin/Debug/LemmaGenSentenceLemmatizer.exe $(SRC_LANG) < $< | sort -u > $@
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/trg_clean.lem: corpora/$(CORPUS_NAME)/trg_clean.tok
|
|
|
|
|
mono ../LemmaGenSentenceLemmatizer/LemmaGenSentenceLemmatizer/bin/Debug/LemmaGenSentenceLemmatizer.exe $(TRG_LANG) < $< > $@
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/src_clean.lem: corpora/$(CORPUS_NAME)/src_clean.tok
|
|
|
|
|
mono ../LemmaGenSentenceLemmatizer/LemmaGenSentenceLemmatizer/bin/Debug/LemmaGenSentenceLemmatizer.exe $(SRC_LANG) < $< > $@
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/src_clean.txt corpora/$(CORPUS_NAME)/trg_clean.txt corpora/$(CORPUS_NAME)/src_clean.tok corpora/$(CORPUS_NAME)/trg_clean.tok: corpora/$(CORPUS_NAME)/src.txt corpora/$(CORPUS_NAME)/trg.txt corpora/$(CORPUS_NAME)/src.tok corpora/$(CORPUS_NAME)/trg.tok
|
|
|
|
|
../clean_corpus.py corpora/$(CORPUS_NAME)/src.txt corpora/$(CORPUS_NAME)/trg.txt corpora/$(CORPUS_NAME)/src.tok corpora/$(CORPUS_NAME)/trg.tok corpora/$(CORPUS_NAME)/src_clean.txt corpora/$(CORPUS_NAME)/trg_clean.txt corpora/$(CORPUS_NAME)/src_clean.tok corpora/$(CORPUS_NAME)/trg_clean.tok $(SEPARATOR)
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/trg.tok: corpora/$(CORPUS_NAME)/trg.txt
|
|
|
|
|
/usr/local/bin/concordia-sentence-tokenizer -c ../../concordia.cfg < $< > $@
|
|
|
|
|
|
|
|
|
|
corpora/$(CORPUS_NAME)/src.tok: corpora/$(CORPUS_NAME)/src.txt
|
|
|
|
|
/usr/local/bin/concordia-sentence-tokenizer -c ../../concordia.cfg < $< > $@
|
|
|
|
|
|
2017-07-21 17:55:28 +02:00
|
|
|
clean-filtering:
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/src_filtered.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/trg_filtered.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/corpus_lines.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/report.txt
|
|
|
|
|
rm -f dictionaries/$(DICTIONARY_NAME).lem
|
|
|
|
|
|
|
|
|
|
|
2017-07-21 17:37:00 +02:00
|
|
|
clean:
|
2017-07-21 17:55:28 +02:00
|
|
|
rm -f corpora/$(CORPUS_NAME)/src_filtered.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/trg_filtered.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/corpus.indexed
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/corpus.split
|
2017-07-21 17:37:00 +02:00
|
|
|
rm -f corpora/$(CORPUS_NAME)/report.txt
|
|
|
|
|
./clear_solr_index.sh
|
|
|
|
|
rm -rf corpora/$(CORPUS_NAME)/csv
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/src.csv
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/corpus_lines.txt
|
|
|
|
|
rm -f dictionaries/$(DICTIONARY_NAME).lem
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/*.lem
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/*.tok
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/src_clean.txt
|
|
|
|
|
rm -f corpora/$(CORPUS_NAME)/trg_clean.txt
|
