Ipynb preparing russian dataset

category_classification/datasets/prepare_ru_dataset.ipynb

@@ -0,0 +1,112 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pathlib import Path\n",
+    "\n",
+    "import pandas as pd\n",
+    "\n",
+    "from datasets_common import write_dataset, train_test_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset_dir = Path('ru')\n",
+    "parts_dir = dataset_dir / 'dataset_parts' "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dfs = []\n",
+    "for dataset_path in parts_dir.glob(\"*.csv\"):\n",
+    "    dfs.append(pd.read_csv(dataset_path))\n",
+    "df = pd.concat(dfs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "label_name = 'category'\n",
+    "df = df.rename(columns={'categories': label_name})\n",
+    "for column in df.columns:\n",
+    "    def transform_cell(value):\n",
+    "        prefixes = [\"{'translation_text': '\", \"{\\'translation_text\\': \\'\", \"\\'translation_text\\':\", \"{'translation_text': \\\"\"]\n",
+    "        suffix = \"\\'}\"\n",
+    "        for prefix in prefixes:\n",
+    "            if value.startswith(prefix):\n",
+    "                value = value[len(prefix):]\n",
+    "        if value.endswith(suffix):\n",
+    "            value = value[:-len(suffix)]\n",
+    "        return value\n",
+    "    df[column] = df[column].apply(transform_cell)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y = df[label_name]\n",
+    "X = df.drop(columns=label_name)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X_train, X_test, y_train, y_test = train_test_split(X, y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_filename = \"arxiv_train.csv\"\n",
+    "test_filename = \"arxiv_test.csv\"\n",
+    "write_dataset(dest_dir=dataset_dir, X=X_train, y=y_train, filename=train_filename, to_json=False)\n",
+    "write_dataset(dest_dir=dataset_dir, X=X_test, y=y_test, filename=test_filename, to_json=False)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}