Feature/hw 4

mypy.ini

@@ -1,2 +1,3 @@
 [mypy]
 allow_untyped_calls = true
+ignore_missing_imports = True

notebooks/online_LightFm.ipynb

@@ -0,0 +1,173 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import pickle\n",
+    "import time\n",
+    "from collections import defaultdict\n",
+    "from functools import reduce\n",
+    "from pathlib import Path\n",
+    "from typing import Optional\n",
+    "\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import scipy as sp\n",
+    "from lightfm import LightFM\n",
+    "from scipy.sparse import csr_matrix\n",
+    "from tqdm import tqdm\n",
+    "\n",
+    "from service.api.models_zoo import LightFMWrapper"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "os.environ['OPENBLAS_NUM_THREADS'] = \"1\"\n",
+    "datetime_col = 'last_watch_dt'\n",
+    "DATA_PATH = Path(\"../data/kion_train/\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CPU times: user 2.01 s, sys: 224 ms, total: 2.23 s\n",
+      "Wall time: 2.24 s\n"
+     ]
+    }
+   ],
+   "source": [
+    "%%time\n",
+    "users = pd.read_csv(DATA_PATH / 'users.csv')\n",
+    "items = pd.read_csv(DATA_PATH / 'items.csv')\n",
+    "interactions = pd.read_csv(DATA_PATH / 'interactions.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "interactions[datetime_col] = pd.to_datetime(interactions[datetime_col], format='%Y-%m-%d')\n",
+    "interactions.dropna(inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "interactions['watched'] = pd.cut(\n",
+    "    x=interactions['watched_pct'],\n",
+    "    bins=5,\n",
+    "    labels=[1, 2, 3, 4, 5]\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Удаляем те юзеры и айтемы, которых нет в основных табличках\n",
+    "interactions = interactions.merge(users.user_id.drop_duplicates(), on='user_id')\n",
+    "interactions = interactions.merge(items.item_id.drop_duplicates(), on='item_id')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model = LightFMWrapper(epochs=5)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Epoch: 100%|██████████| 5/5 [00:27<00:00,  5.43s/it]\n"
+     ]
+    }
+   ],
+   "source": [
+    "model.fit(train=interactions, item_features=items, user_features=users)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[3734, 2657, 11237, 13594, 12248, 11754, 1020, 10242, 142, 1451]"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.predict(user_id=176549)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "with open('../data/lightfm.pickle', 'wb') as f:\n",
+    "      pickle.dump(model, f)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.9.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}