Script for downloading russian dataset

category_classification/datasets/ru/download_train_test.py

@@ -0,0 +1,20 @@
+#!/usr/bin/env python
+
+import os
+from pathlib import Path
+
+from kagglehub import dataset_download
+
+dataset = "hibiscus4000/Arxiv-papers-ru"
+
+test_dataset = "arxiv_test.csv"
+train_dataset = "arxiv_train.csv"
+
+dataset_path = Path(dataset_download(dataset))
+test_file_path = dataset_path / test_dataset
+train_file_path = dataset_path / train_dataset
+
+if not test_file_path.exists():
+    os.symlink(dataset_path / test_dataset, test_dataset)
+if not train_file_path.exists():
+    os.symlink(dataset_path / train_dataset, train_dataset)