разделил данные по папкам

app.py

@@ -8,14 +8,14 @@ from transformers import AutoTokenizer, AutoModel
 import numpy as np
 from sklearn.metrics.pairwise import cosine_similarity
 
-movies = pd.read_csv('data.csv')
+movies = pd.read_csv('data/data.csv')
 
 toggle_state = st.sidebar.checkbox("режим разметки")
 input_search = st.text_input('Search')
 
 
 
-data = np.load('embeddings.npy')
+data = np.load('data/embeddings.npy')
 
 
 

catboost.ipynb

@@ -0,0 +1,309 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Результат:\n",
+      "[[1 2 3 7 8]\n",
+      " [4 5 6 7 8]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "\n",
+    "# Создаем матрицу (2D массив)\n",
+    "matrix = np.array([[1, 2, 3],\n",
+    "                   [4, 5, 6]])\n",
+    "\n",
+    "# Создаем вектор (1D массив)\n",
+    "vector = np.array([7, 8])\n",
+    "\n",
+    "# Сконкатенируем каждый вектор матрицы с вектором\n",
+    "result = np.column_stack((matrix, np.tile(vector, (matrix.shape[0], 1))))\n",
+    "\n",
+    "print(\"Результат:\")\n",
+    "print(result)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "137"
+      ]
+     },
+     "execution_count": 63,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 93,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X = np.load('X.npy')\n",
+    "Y = np.load('y.npy')\n",
+    "X=X[-2:]\n",
+    "Y=Y[-2:]\n",
+    "np.save('X',X)\n",
+    "np.save('y',Y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 104,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(29263, 624)"
+      ]
+     },
+     "execution_count": 104,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 106,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(29265, 624)"
+      ]
+     },
+     "execution_count": 106,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    ".shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 109,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(29265,)"
+      ]
+     },
+     "execution_count": 109,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Y.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "\n",
+    "dat = np.load('embeddings.npy')\n",
+    "data =np.column_stack((dat, dat))\n",
+    "datay = np.ones((data.shape[0]))*5\n",
+    "\n",
+    "data1 = np.column_stack((dat[1:], dat[:-1]))\n",
+    "datay1 = np.ones((data1.shape[0]))\n",
+    "\n",
+    "\n",
+    "X = np.load('X.npy') \n",
+    "Y = np.load('y.npy')\n",
+    "\n",
+    "\n",
+    "\n",
+    "X=np.concatenate((data,X))\n",
+    "Y=np.concatenate((datay,Y))\n",
+    "X = np.concatenate((data1,X))\n",
+    "Y = np.concatenate((datay1,Y))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 132,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(29263, 624)"
+      ]
+     },
+     "execution_count": 132,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "4.5227014967230694e-05"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "logreg = LinearRegression()\n",
+    "logreg.fit(X, Y)\n",
+    "\n",
+    "import pickle\n",
+    "with open('logreg.pkl', 'wb') as f:\n",
+    "    pickle.dump(logreg, f)\n",
+    "\n",
+    "logreg.score(X, Y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0:\ttest: 0.9786223\tbest: 0.9786223 (0)\ttotal: 51.1s\tremaining: 7m 39s\n",
+      "1:\ttest: 0.9950170\tbest: 0.9950170 (1)\ttotal: 1m 13s\tremaining: 4m 54s\n",
+      "2:\ttest: 0.9966407\tbest: 0.9966407 (2)\ttotal: 1m 35s\tremaining: 3m 42s\n",
+      "3:\ttest: 0.9982912\tbest: 0.9982912 (3)\ttotal: 1m 56s\tremaining: 2m 55s\n",
+      "4:\ttest: 0.9988039\tbest: 0.9988039 (4)\ttotal: 2m 18s\tremaining: 2m 18s\n",
+      "5:\ttest: 0.9992459\tbest: 0.9992459 (5)\ttotal: 2m 39s\tremaining: 1m 46s\n",
+      "6:\ttest: 0.9997030\tbest: 0.9997030 (6)\ttotal: 3m 1s\tremaining: 1m 17s\n",
+      "7:\ttest: 0.9998173\tbest: 0.9998173 (7)\ttotal: 3m 22s\tremaining: 50.7s\n",
+      "8:\ttest: 0.9998216\tbest: 0.9998216 (8)\ttotal: 3m 44s\tremaining: 24.9s\n",
+      "9:\ttest: 0.9998608\tbest: 0.9998608 (9)\ttotal: 4m 5s\tremaining: 0us\n",
+      "\n",
+      "bestTest = 0.9998607928\n",
+      "bestIteration = 9\n",
+      "\n"
+     ]
+    },
+    {
+     "ename": "",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31mThe Kernel crashed while executing code in the current cell or a previous cell. \n",
+      "\u001b[1;31mPlease review the code in the cell(s) to identify a possible cause of the failure. \n",
+      "\u001b[1;31mClick <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. \n",
+      "\u001b[1;31mView Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
+     ]
+    }
+   ],
+   "source": [
+    "from catboost import CatBoostRanker,Pool\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "\n",
+    "X, X_test, Y, Y_test = train_test_split(X, Y, test_size=0.1, random_state=42)\n",
+    "classes_test = np.ones(len(Y_test)).astype(int)\n",
+    "test_pool = Pool(data=X_test, label=Y_test, group_id=classes_test)\n",
+    "\n",
+    "\n",
+    "classes_train = np.ones(len(Y)).astype(int)\n",
+    "train_pool = Pool(data=X, label=Y, group_id=classes_train,)\n",
+    "\n",
+    "\n",
+    "cb = CatBoostRanker(iterations=10,)\n",
+    "cb.fit(train_pool,eval_set=test_pool)\n",
+    "cb.save_model('model.cbm')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[[0.82140051]\n",
+      " [0.91314228]\n",
+      " [0.92991252]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "from sklearn.metrics.pairwise import cosine_similarity\n",
+    "\n",
+    "# Пример данных\n",
+    "matrix = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])\n",
+    "vector = np.array([0.5, 0.7, 0.3])\n",
+    "\n",
+    "# Вычисление косинусного сходства между матрицей и вектором\n",
+    "similarity = cosine_similarity(matrix, vector.reshape(1, -1))\n",
+    "\n",
+    "print(similarity)\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "cv",
+   "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": "3.12.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

data/X.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27096da97e7c093830ae02cb275c732f43a59c3aa96db08297466544f92a9b58
+size 37568

data/logreg.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecd4a43c1bd0b0f35262c896637f2beb4e020d3a26a28dec808b5c13ec67e093
+size 5407

data/y.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce8faa4674f06c82833ba7fddbd5e6ffb3a98bfcac2bad53facb1782a848f34a
+size 248

search.ipynb

@@ -0,0 +1,362 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>name</th>\n",
+       "      <th>description</th>\n",
+       "      <th>link</th>\n",
+       "      <th>year</th>\n",
+       "      <th>imdb</th>\n",
+       "      <th>kp</th>\n",
+       "      <th>country</th>\n",
+       "      <th>age</th>\n",
+       "      <th>actors</th>\n",
+       "      <th>genres</th>\n",
+       "      <th>poster</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>Мстители: Война бесконечности</td>\n",
+       "      <td>В то время как отважные Мстители с союзниками...</td>\n",
+       "      <td>https://www.lordfilm.bot/3670-mstiteli-vojna-b...</td>\n",
+       "      <td>2018.0</td>\n",
+       "      <td>8.4</td>\n",
+       "      <td>8.1</td>\n",
+       "      <td>США</td>\n",
+       "      <td>18+</td>\n",
+       "      <td>Роберт Дауни мл., Крис Хемсворт, Марк Руффало,...</td>\n",
+       "      <td>Фильмы, Фильмы Marvel, Боевики, Приключения, Ф...</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2020-10...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4286</th>\n",
+       "      <td>LEGO Мстители Марвел: Код красный</td>\n",
+       "      <td>Супергерои объединяются, чтобы противостоять ...</td>\n",
+       "      <td>https://www.lordfilm.bot/49932-lego-mstiteli-m...</td>\n",
+       "      <td>2023.0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>США</td>\n",
+       "      <td>0+</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Мультфильмы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2023-10...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13384</th>\n",
+       "      <td>Могучие рейнджеры: Потерянная галактика</td>\n",
+       "      <td>Казалось бы всё зло уже побеждено, однако в н...</td>\n",
+       "      <td>https://www.lordfilm.bot/18827-moguchie-rejndz...</td>\n",
+       "      <td>1999.0</td>\n",
+       "      <td>6.8</td>\n",
+       "      <td>4.2</td>\n",
+       "      <td>США, Франция, Япония</td>\n",
+       "      <td>0+</td>\n",
+       "      <td>Арчи Као, Регги Ролли, Дэнни Славин, Серина Ви...</td>\n",
+       "      <td>Сериалы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2021-03...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2609</th>\n",
+       "      <td>Стражи терракоты</td>\n",
+       "      <td>Стражи волшебной Терракоты и магические сущес...</td>\n",
+       "      <td>https://www.lordfilm.bot/46847-strazhi-terrako...</td>\n",
+       "      <td>2021.0</td>\n",
+       "      <td>6.2</td>\n",
+       "      <td>6.7</td>\n",
+       "      <td>Китай</td>\n",
+       "      <td>12+</td>\n",
+       "      <td>Тань Сяо</td>\n",
+       "      <td>Мультфильмы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2022-01...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1156</th>\n",
+       "      <td>Царство падальщиков</td>\n",
+       "      <td>Грузовой корабль «Деметра» терпит аварию на н...</td>\n",
+       "      <td>https://www.lordfilm.bot/49892-carstvo-padalsc...</td>\n",
+       "      <td>2023.0</td>\n",
+       "      <td>8.8</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>США</td>\n",
+       "      <td>0+</td>\n",
+       "      <td>Вунми Моссаку, Алиа Шокат, Сунита Мани, Боб Ст...</td>\n",
+       "      <td>Мультфильмы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2023-10...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3907</th>\n",
+       "      <td>Железный человек и Халк: Союз героев</td>\n",
+       "      <td>Когда питающийся электричеством и неуязвимый ...</td>\n",
+       "      <td>https://www.lordfilm.bot/21296-zheleznyj-chelo...</td>\n",
+       "      <td>2013.0</td>\n",
+       "      <td>4.6</td>\n",
+       "      <td>4.2</td>\n",
+       "      <td>США</td>\n",
+       "      <td>12+</td>\n",
+       "      <td>Адриан Пасдар, Фред Таташиор, Ди Брэдли Бейкер...</td>\n",
+       "      <td>Мультфильмы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2021-04...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27972</th>\n",
+       "      <td>Потерянное львиное королевство</td>\n",
+       "      <td>Мультфильм о борьбе добра со злом на простора...</td>\n",
+       "      <td>https://www.lordfilm.bot/24407-poterjannoe-lvi...</td>\n",
+       "      <td>2019.0</td>\n",
+       "      <td>3.8</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>США</td>\n",
+       "      <td>18+</td>\n",
+       "      <td>Kj Schrock, Сара Тейлор</td>\n",
+       "      <td>Мультфильмы</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2021-05...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>297</th>\n",
+       "      <td>Мир Юрского периода 3: Господство</td>\n",
+       "      <td>Катастрофическое извержение вулкана Сибо на И...</td>\n",
+       "      <td>https://www.lordfilm.bot/47499-mir-jurskogo-pe...</td>\n",
+       "      <td>2022.0</td>\n",
+       "      <td>5.6</td>\n",
+       "      <td>5.7</td>\n",
+       "      <td>США, Мальта</td>\n",
+       "      <td>12+</td>\n",
+       "      <td>Крис Пратт, Брайс Даллас Ховард, Лора Дерн, Сэ...</td>\n",
+       "      <td>Фильмы, 2022 год, Боевики, Приключения, Трилле...</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2022-06...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7626</th>\n",
+       "      <td>Лузеры</td>\n",
+       "      <td>«Лузеры» – сумасшедший экшн о предательстве и...</td>\n",
+       "      <td>https://www.lordfilm.bot/6139-luzery-2010.html</td>\n",
+       "      <td>2010.0</td>\n",
+       "      <td>6.2</td>\n",
+       "      <td>6.3</td>\n",
+       "      <td>США, Франция</td>\n",
+       "      <td>16+</td>\n",
+       "      <td>Джеффри Дин Морган, Зои Салдана, Крис Эванс, И...</td>\n",
+       "      <td>Фильмы, Боевики, Детективы, Комедии, Криминаль...</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2021-01...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25332</th>\n",
+       "      <td>Ancestral World</td>\n",
+       "      <td>Пытаясь спасти своего брата и королевство сво...</td>\n",
+       "      <td>https://www.lordfilm.bot/10306-ancestral-world...</td>\n",
+       "      <td>2020.0</td>\n",
+       "      <td>2.5</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>США</td>\n",
+       "      <td>0+</td>\n",
+       "      <td>Jennifer Mischiati, Джо Морелли, Райан А. Филл...</td>\n",
+       "      <td>Фильмы, Боевики, 2020 год, Фильмы про монстров</td>\n",
+       "      <td>https://www.lordfilm.bot/uploads/posts/2021-02...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                          name  \\\n",
+       "28               Мстители: Война бесконечности   \n",
+       "4286         LEGO Мстители Марвел: Код красный   \n",
+       "13384  Могучие рейнджеры: Потерянная галактика   \n",
+       "2609                          Стражи терракоты   \n",
+       "1156                       Царство падальщиков   \n",
+       "3907      Железный человек и Халк: Союз героев   \n",
+       "27972           Потерянное львиное королевство   \n",
+       "297          Мир Юрского периода 3: Господство   \n",
+       "7626                                    Лузеры   \n",
+       "25332                          Ancestral World   \n",
+       "\n",
+       "                                             description  \\\n",
+       "28      В то время как отважные Мстители с союзниками...   \n",
+       "4286    Супергерои объединяются, чтобы противостоять ...   \n",
+       "13384   Казалось бы всё зло уже побеждено, однако в н...   \n",
+       "2609    Стражи волшебной Терракоты и магические сущес...   \n",
+       "1156    Грузовой корабль «Деметра» терпит аварию на н...   \n",
+       "3907    Когда питающийся электричеством и неуязвимый ...   \n",
+       "27972   Мультфильм о борьбе добра со злом на простора...   \n",
+       "297     Катастрофическое извержение вулкана Сибо на И...   \n",
+       "7626    «Лузеры» – сумасшедший экшн о предательстве и...   \n",
+       "25332   Пытаясь спасти своего брата и королевство сво...   \n",
+       "\n",
+       "                                                    link    year  imdb   kp  \\\n",
+       "28     https://www.lordfilm.bot/3670-mstiteli-vojna-b...  2018.0   8.4  8.1   \n",
+       "4286   https://www.lordfilm.bot/49932-lego-mstiteli-m...  2023.0   NaN  NaN   \n",
+       "13384  https://www.lordfilm.bot/18827-moguchie-rejndz...  1999.0   6.8  4.2   \n",
+       "2609   https://www.lordfilm.bot/46847-strazhi-terrako...  2021.0   6.2  6.7   \n",
+       "1156   https://www.lordfilm.bot/49892-carstvo-padalsc...  2023.0   8.8  NaN   \n",
+       "3907   https://www.lordfilm.bot/21296-zheleznyj-chelo...  2013.0   4.6  4.2   \n",
+       "27972  https://www.lordfilm.bot/24407-poterjannoe-lvi...  2019.0   3.8  NaN   \n",
+       "297    https://www.lordfilm.bot/47499-mir-jurskogo-pe...  2022.0   5.6  5.7   \n",
+       "7626      https://www.lordfilm.bot/6139-luzery-2010.html  2010.0   6.2  6.3   \n",
+       "25332  https://www.lordfilm.bot/10306-ancestral-world...  2020.0   2.5  NaN   \n",
+       "\n",
+       "                    country  age  \\\n",
+       "28                      США  18+   \n",
+       "4286                    США   0+   \n",
+       "13384  США, Франция, Япония   0+   \n",
+       "2609                  Китай  12+   \n",
+       "1156                    США   0+   \n",
+       "3907                    США  12+   \n",
+       "27972                   США  18+   \n",
+       "297             США, Мальта  12+   \n",
+       "7626           США, Франция  16+   \n",
+       "25332                   США   0+   \n",
+       "\n",
+       "                                                  actors  \\\n",
+       "28     Роберт Дауни мл., Крис Хемсворт, Марк Руффало,...   \n",
+       "4286                                                 NaN   \n",
+       "13384  Арчи Као, Регги Ролли, Дэнни Славин, Серина Ви...   \n",
+       "2609                                            Тань Сяо   \n",
+       "1156   Вунми Моссаку, Алиа Шокат, Сунита Мани, Боб Ст...   \n",
+       "3907   Адриан Пасдар, Фред Таташиор, Ди Брэдли Бейкер...   \n",
+       "27972                            Kj Schrock, Сара Тейлор   \n",
+       "297    Крис Пратт, Брайс Даллас Ховард, Лора Дерн, Сэ...   \n",
+       "7626   Джеффри Дин Морган, Зои Салдана, Крис Эванс, И...   \n",
+       "25332  Jennifer Mischiati, Джо Морелли, Райан А. Филл...   \n",
+       "\n",
+       "                                                  genres  \\\n",
+       "28     Фильмы, Фильмы Marvel, Боевики, Приключения, Ф...   \n",
+       "4286                                         Мультфильмы   \n",
+       "13384                                            Сериалы   \n",
+       "2609                                         Мультфильмы   \n",
+       "1156                                         Мультфильмы   \n",
+       "3907                                         Мультфильмы   \n",
+       "27972                                        Мультфильмы   \n",
+       "297    Фильмы, 2022 год, Боевики, Приключения, Трилле...   \n",
+       "7626   Фильмы, Боевики, Детективы, Комедии, Криминаль...   \n",
+       "25332     Фильмы, Боевики, 2020 год, Фильмы про монстров   \n",
+       "\n",
+       "                                                  poster  \n",
+       "28     https://www.lordfilm.bot/uploads/posts/2020-10...  \n",
+       "4286   https://www.lordfilm.bot/uploads/posts/2023-10...  \n",
+       "13384  https://www.lordfilm.bot/uploads/posts/2021-03...  \n",
+       "2609   https://www.lordfilm.bot/uploads/posts/2022-01...  \n",
+       "1156   https://www.lordfilm.bot/uploads/posts/2023-10...  \n",
+       "3907   https://www.lordfilm.bot/uploads/posts/2021-04...  \n",
+       "27972  https://www.lordfilm.bot/uploads/posts/2021-05...  \n",
+       "297    https://www.lordfilm.bot/uploads/posts/2022-06...  \n",
+       "7626   https://www.lordfilm.bot/uploads/posts/2021-01...  \n",
+       "25332  https://www.lordfilm.bot/uploads/posts/2021-02...  "
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "\n",
+    "from transformers import AutoTokenizer, AutoModel\n",
+    "import numpy as np\n",
+    "from sklearn.metrics.pairwise import cosine_similarity\n",
+    "import torch\n",
+    "import pandas as pd\n",
+    "\n",
+    "\n",
+    "data = np.load('embeddings.npy')\n",
+    "movies = pd.read_csv('data.csv')\n",
+    "\n",
+    "def get_embeddings():\n",
+    "    tokenizer = AutoTokenizer.from_pretrained(\"cointegrated/rubert-tiny2\")\n",
+    "    model = AutoModel.from_pretrained(\"cointegrated/rubert-tiny2\")\n",
+    "    # model.cuda()  \n",
+    "    return model, tokenizer\n",
+    "\n",
+    "def embed_bert_cls(text ):\n",
+    "    model, tokenizer = get_embeddings()\n",
+    "    t = tokenizer(text, padding=True, truncation=True, return_tensors='pt')\n",
+    "    with torch.no_grad():\n",
+    "        model_output = model(**{k: v.to(model.device) for k, v in t.items()})\n",
+    "    embeddings = model_output.last_hidden_state[:, 0, :]\n",
+    "    embeddings = torch.nn.functional.normalize(embeddings)\n",
+    "    return embeddings[0].cpu().numpy()\n",
+    "\n",
+    "def top_indices(array, n):\n",
+    "\n",
+    "    sorted_indices = np.argsort(array)[::-1]\n",
+    "    # Выбираем первые n индексов\n",
+    "    top_n_indices = sorted_indices[:n]\n",
+    "    return top_n_indices\n",
+    "\n",
+    "\n",
+    "def predict_rating(input_search):\n",
+    "\n",
+    "    emb = embed_bert_cls(input_search)\n",
+    "    X=np.column_stack((data, np.tile(emb, (data.shape[0], 1))))\n",
+    "\n",
+    "\n",
+    "    # from catboost import CatBoostRanker\n",
+    "    # cb= CatBoostRanker()\n",
+    "    # cb.load_model('model.cbm')\n",
+    "    # y = cb.predict(X)\n",
+    "\n",
+    "    # import pickle\n",
+    "    # with open('logreg.pkl', 'rb') as f:\n",
+    "    #     logreg = pickle.load(f)\n",
+    "    # y = logreg.predict(X)\n",
+    "\n",
+    "    y= cosine_similarity(data, emb.reshape(1, -1)).reshape(-1)\n",
+    "\n",
+    "    return top_indices(y, 10)\n",
+    "\n",
+    "\n",
+    "preds=predict_rating(\"Пока Мстители и их союзники продолжают защищать мир от различных опасностей, с которыми не смог бы справиться один супергерой, новая угроза возникает из космоса: Танос. Межгалактический тиран преследует цель \")\n",
+    "\n",
+    "movies.iloc[preds]"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "cv",
+   "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": "3.12.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}