Stable tested version
diff --git a/systems/parse_turku.py b/systems/parse_turku.py
index 284dbd4..668a48c 100644
--- a/systems/parse_turku.py
+++ b/systems/parse_turku.py
@@ -6,6 +6,7 @@
import my_utils.file_utils as fu
import argparse
+
if __name__ == "__main__":
"""
@@ -21,6 +22,7 @@
parser = argparse.ArgumentParser()
parser.add_argument("-i", "--input_file", help="Input Corpus", required=True)
+ parser.add_argument("-o", "--output_file", help="Output Parsed Corpus", default=None)
parser.add_argument("-n", "--corpus_name", help="Corpus Name", default="Corpus")
parser.add_argument("-gtt", "--gld_token_type", help="CoNLL Format of the Gold Data", default="CoNLL09_Token")
parser.add_argument("-c", "--comment_str", help="CoNLL Format of comentaries inside the file", default="#")
@@ -29,6 +31,8 @@
file_has_next, chunk_ix = True, 0
CHUNK_SIZE = 10000
+ output_file = args.input_file if not args.output_file else args.output_file
+
# =====================================================================================
# LOGGING INFO ...
# =====================================================================================
@@ -48,7 +52,7 @@
raw_text, file_has_next, n_sents = fu.get_file_chunk(line_generator, chunk_size=CHUNK_SIZE, token_class=get_token_type(args.gld_token_type), comment_str=args.comment_str)
total_processed_sents += n_sents
if len(raw_text) > 0:
- fu.turku_parse_file(raw_text, args.input_file, chunk_ix)
+ fu.turku_parse_file(raw_text, output_file, chunk_ix)
now = time.time()
elapsed = (now - start)
logger.info(f"Time Elapsed: {elapsed}. Processed {total_processed_sents}. [{total_processed_sents/elapsed} Sents/sec]\n") # Toks/Sec???