{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "81d6d69a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1cea66b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "tmdb = pd.read_csv(\"../datasets/raw/tmdb-movies/TMDB_movie_dataset_v11.csv\")\n",
    "KEEP = [\n",
    "    \"id\",\n",
    "    \"title\",\n",
    "    \"overview\",\n",
    "    \"genres\",\n",
    "    \"keywords\",\n",
    "    \"vote_average\",\n",
    "    \"vote_count\",\n",
    "    \"popularity\",\n",
    "    \"release_date\"\n",
    "]\n",
    "tmdb = tmdb[KEEP]\n",
    "tmdb = tmdb[tmdb[\"vote_count\"] > 30]\n",
    "tmdb[\"overview\"] = tmdb[\"overview\"].fillna(\"\")\n",
    "tmdb[\"genres\"] = tmdb[\"genres\"].fillna(\"\")\n",
    "tmdb[\"keywords\"] = tmdb[\"keywords\"].fillna(\"\")\n",
    "tmdb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "585e9712",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.hist(tmdb[\"popularity\"], bins=50)\n",
    "plt.title(\"Raw popularity\")\n",
    "plt.show()\n",
    "\n",
    "plt.hist(np.log1p(tmdb[\"popularity\"]), bins=50)\n",
    "plt.title(\"Log-transformed popularity\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dcd0d96",
   "metadata": {},
   "outputs": [],
   "source": [
    "tmdb[\"popularity_log\"] = np.log1p(tmdb[\"popularity\"])\n",
    "tmdb[\"content\"] = tmdb[\"overview\"] + \" \" + tmdb[\"genres\"] + \" \" + tmdb[\"keywords\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9be9b559",
   "metadata": {},
   "outputs": [],
   "source": [
    "tmdb.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fc5b3a32",
   "metadata": {},
   "outputs": [],
   "source": [
    "movie_lens_links = pd.read_csv(\"../datasets/raw/ml-32m/links.csv\")\n",
    "\n",
    "movie_lens_links = movie_lens_links.dropna(subset=[\"tmdbId\"])\n",
    "movie_lens_links[\"tmdbId\"] = movie_lens_links[\"tmdbId\"].astype(\"int64\")\n",
    "movie_lens_links.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0d725578",
   "metadata": {},
   "outputs": [],
   "source": [
    "movie_lens_ratings = pd.read_csv(\"../datasets/raw/ml-32m/ratings.csv\")\n",
    "ratings = pd.merge(\n",
    "    movie_lens_ratings[['userId', 'rating', 'movieId']],\n",
    "    movie_lens_links[['movieId', 'tmdbId']],\n",
    "    on='movieId',\n",
    "    how='left'\n",
    ")\n",
    "ratings = ratings[ratings[\"tmdbId\"].notna()]\n",
    "ratings[\"tmdbId\"] = ratings[\"tmdbId\"].astype(\"int64\")\n",
    "\n",
    "min_user_ratings = 5\n",
    "min_movie_ratings = 5\n",
    "\n",
    "user_counts = ratings.groupby('userId').size()\n",
    "movie_counts = ratings.groupby('tmdbId').size()\n",
    "\n",
    "rating = ratings[\n",
    "    ratings['userId'].isin(user_counts[user_counts >= min_user_ratings].index) &\n",
    "    ratings['tmdbId'].isin(movie_counts[movie_counts >= min_movie_ratings].index)\n",
    "]\n",
    "ratings.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "da16642e",
   "metadata": {},
   "outputs": [],
   "source": [
    "ratings.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cad9adc3-2c75-45a9-9d1b-52ec4fb413e6",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv (3.13.9)",
   "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": "3.13.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}