debug

README.md

@@ -28,9 +28,9 @@ This dataset is intended for language detection for Bokmål to Nynorsk and vice
 ## Data Collection
 
 - **Period**: February 2011 to December 2022
-- **Source**: [Språkbanken](https://www.nb.no/sprakbanken/)
+- **Source**: [Omsetjingsminne frå Nynorsk Pressekontor - Språkbanken](https://www.nb.no/sprakbanken/ressurskatalog/oai-nb-no-sbr-80/)
 - **Size**: 800,000 sentence pairs
-- **Format**: JSON-lines (with `text` , `label`, `label_full` fields)
+- **Format**: JSON-lines (with `text` , `language` fields)
 
 ### Processing Steps
 

test.py

@@ -1,20 +0,0 @@
-# Python script to check the data types in a JSONL file
-import json
-import argparse
-
-def check_data_types(jsonl_file_path):
-    with open(jsonl_file_path, 'r') as f:
-        for i, line in enumerate(f):
-            data = json.loads(line)
-            if not isinstance(data.get('text', None), str):
-                print(f"Inconsistent data type at row {i+1}. 'text' is not a string.")
-            if not isinstance(data.get('language', None), str):
-                print(f"Inconsistent data type at row {i+1}. 'language' is not a string.")
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description='Check data types in a JSONL file.')
-    parser.add_argument('--jsonl_file_path', required=True, help='Path to the JSONL file.')
-
-    args = parser.parse_args()
-    check_data_types(args.jsonl_file_path)
-