Trends in Anime & Manga.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trends in Anime & Manga"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mangaki.algo.dataset import Dataset\n",
    "from mangaki.algo.fit_algo import get_dataset_backup\n",
    "from mangaki.utils.values import rating_values\n",
    "\n",
    "dataset = Dataset()\n",
    "dataset.load_csv('balse/ratings.csv', convert=lambda x: rating_values[x], title_filename='balse/works.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mangaki.algo.balse import MangakiBALSE\n",
    "from mangaki.settings import DATA_DIR\n",
    "from scipy.sparse import load_npz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os.path\n",
    "import numpy as np\n",
    "\n",
    "T = load_npz(os.path.join(DATA_DIR, 'balse/tag-matrix.npz'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<9979x503 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 210009 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "tags = json.load(open('../data/illustration2vec/mangaki_i2v.json'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'general': [['1girl', 0.5351361632347107],\n",
       "  ['1boy', 0.5314481258392334],\n",
       "  ['black hair', 0.4160270392894745],\n",
       "  ['short hair', 0.4073851704597473],\n",
       "  ['blue eyes', 0.2690363824367523],\n",
       "  ['multiple boys', 0.16339154541492462],\n",
       "  ['weapon', 0.16306668519973755],\n",
       "  ['long hair', 0.16088248789310455],\n",
       "  ['2boys', 0.15612725913524628],\n",
       "  ['male', 0.1489870250225067],\n",
       "  ['brown hair', 0.13320335745811462],\n",
       "  ['scan', 0.11500567942857742],\n",
       "  ['2girls', 0.10508275777101517]],\n",
       " 'character': [],\n",
       " 'copyright': [],\n",
       " 'rating': [['safe', 0.9664666652679443],\n",
       "  ['questionable', 0.0323437824845314],\n",
       "  ['explicit', 0.00026572716888040304]]}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tags['1.jpg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "work_ids = list(Work.objects.values_list('id', flat=True))\n",
    "encode_work = dict(zip(work_ids, range(len(work_ids))))\n",
    "decode_work = {v: k for k, v in encode_work.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "503\n"
     ]
    }
   ],
   "source": [
    "all_tags = set()\n",
    "for _, value in tags.items():\n",
    "    for tag, _ in value['general']:\n",
    "        all_tags.add(tag)\n",
    "print(len(all_tags))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "encode_tag = dict(zip(list(all_tags), range(len(all_tags))))\n",
    "decode_tag = {v: k for k, v in encode_tag.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = []\n",
    "rows = []\n",
    "cols = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.sparse import coo_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "for work_id in range(len(encode_work)):\n",
    "    poster = '{:d}.jpg'.format(work_id)\n",
    "    if poster in tags:\n",
    "        for tag, value in tags[poster]['general']:\n",
    "            rows.append(work_id)\n",
    "            cols.append(encode_tag[tag])\n",
    "            data.append(value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(307075, 307075, 307075)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(rows), len(cols), len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "T = coo_matrix((data, (rows, cols))).tocsr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<12817x503 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 307075 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.sparse import save_npz\n",
    "\n",
    "save_npz('tags2018.npz', T)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Work.objects.get(id=1346).date.month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16796"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Work.objects.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12624"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Work.objects.filter(date__isnull=False).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_encoded = T.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter, defaultdict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "trends = defaultdict(Counter)\n",
    "popular = defaultdict(Counter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "217 in 2000\n",
      "349 in 2001\n",
      "343 in 2002\n",
      "395 in 2003\n",
      "427 in 2004\n",
      "446 in 2005\n",
      "504 in 2006\n",
      "493 in 2007\n",
      "501 in 2008\n",
      "546 in 2009\n",
      "549 in 2010\n",
      "617 in 2011\n",
      "658 in 2012\n",
      "652 in 2013\n",
      "741 in 2014\n",
      "730 in 2015\n",
      "493 in 2016\n",
      "320 in 2017\n",
      "139 in 2018\n"
     ]
    }
   ],
   "source": [
    "for year in range(2000, 2019):\n",
    "    ids = Work.objects.filter(date__year=year).values_list('id', flat=True).popular()\n",
    "    print(len(ids), 'in', year)\n",
    "    encoded = [encode_work[work_id] for work_id in ids if encode_work[work_id] < max_encoded]\n",
    "    weights = T[encoded].mean(axis=0).A1\n",
    "    weights2 = T[encoded[:50]].mean(axis=0).A1\n",
    "    trends[year] = Counter(dict(zip(decode_tag.values(), weights)))\n",
    "    popular[year] = Counter(dict(zip(decode_tag.values(), weights2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# interesting = ['1girl', 'multiple girls', '1boy', 'multiple boys', 'solo', 'short hair', 'long hair', 'blonde hair']\n",
    "interesting = ['1girl', 'multiple girls', '1boy', 'multiple boys', 'solo']\n",
    "years = range(2000, 2019)\n",
    "\n",
    "ax = plt.subplot(111)\n",
    "for tag in interesting:\n",
    "    plt.plot(years, [popular[year][tag] for year in years], label=tag)\n",
    "\n",
    "plt.legend()\n",
    "#leg = plt.legend(loc='best', ncol=2, mode=\"expand\", shadow=True, fancybox=True)\n",
    "#leg.get_frame().set_alpha(0.5)\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "performer = Counter()\n",
    "\n",
    "for tag in all_tags:\n",
    "    for year in range(2005, 2015):\n",
    "        performer[tag] += abs(trends[year][tag] - trends[year - 1][tag])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('1girl', 0.2423617688245796),\n",
       " ('solo', 0.21679440158562094),\n",
       " ('multiple girls', 0.21360942023217894),\n",
       " ('1boy', 0.17883022408636914),\n",
       " ('brown hair', 0.15948363945518376),\n",
       " ('red hair', 0.15740664805737725),\n",
       " ('school uniform', 0.15301911138722404),\n",
       " ('black hair', 0.13427295391641358),\n",
       " ('blue hair', 0.13193226431427715),\n",
       " ('breasts', 0.12835214056855682),\n",
       " ('blonde hair', 0.127904875969778),\n",
       " ('6+girls', 0.10539020168310947),\n",
       " ('brown eyes', 0.10447416588166036),\n",
       " ('open mouth', 0.10430087209069533),\n",
       " ('short hair', 0.10365397457312392),\n",
       " ('multiple boys', 0.10195656762683869),\n",
       " ('blush', 0.10007566893056125),\n",
       " ('glasses', 0.09947027753818952),\n",
       " ('underwear', 0.09936063911499088),\n",
       " ('male', 0.09684066599388592)]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performer.most_common(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Django Shell-Plus",
   "language": "python",
   "name": "django_extensions"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}