n-sentences_bloc.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('./src')\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import joblib\n",
    "from icecream import ic\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "  \n",
    "from src.data import load_data_part1\n",
    "from src.eval import print_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([[0, 1], [2, 3], [4, 5], [6, 7], [8, 9]],\n",
       " array([ 1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1, -1, -1, -1]))"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X, y = load_data_part1()\n",
    "def create_n_sentence(y, n):\n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"\n",
    "    prev_y_value = y[0]\n",
    "    bloc_list = []\n",
    "    current_bloc = []\n",
    "    current_bloc_length = 0\n",
    "\n",
    "    for idx in range(len(y)):\n",
    "        if current_bloc_length < n:\n",
    "            current_bloc_length += 1\n",
    "            if prev_y_value == y[idx]:\n",
    "                current_bloc.append(idx)\n",
    "            else:\n",
    "                bloc_list.append(current_bloc)\n",
    "                current_bloc = [idx]\n",
    "                current_bloc_length = 1\n",
    "        else:\n",
    "            bloc_list.append(current_bloc)\n",
    "            current_bloc = [idx]\n",
    "            current_bloc_length = 1\n",
    "        prev_y_value = y[idx]\n",
    "    bloc_list.append(current_bloc)\n",
    "    return bloc_list\n",
    "    \n",
    "nested_idx = create_n_sentence(y, 2)\n",
    "nested_idx[0:5], y[0:14]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(\" Quand je dis chers amis, il ne s'agit pas là d'une formule diplomatique, mais de l'expression de ce que je ressens.\\n D'abord merci de cet exceptionnel accueil que les Congolais, les Brazavillois, nous ont réservé cet après-midi.\\n\",\n",
       " array([\" Quand je dis chers amis, il ne s'agit pas là d'une formule diplomatique, mais de l'expression de ce que je ressens.\\n\",\n",
       "        \" D'abord merci de cet exceptionnel accueil que les Congolais, les Brazavillois, nous ont réservé cet après-midi.\\n\"],\n",
       "       dtype='<U2530'),\n",
       " 1)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def create_nested_dataset(X, y, n):\n",
    "    def merge_string(l):\n",
    "        s = \"\"\n",
    "        for i in l:\n",
    "            s += i\n",
    "        return s\n",
    "    \n",
    "    new_X = []\n",
    "    new_y = []\n",
    "    nested_idx = create_n_sentence(y, n)\n",
    "    for idx_list in nested_idx:\n",
    "        new_X.append(merge_string(X[idx_list]))\n",
    "        new_y.append(y[idx_list[0]])\n",
    "    return new_X, new_y, nested_idx\n",
    "\n",
    "nested_X, nested_y, nested_idx = create_nested_dataset(X, y, 2)\n",
    "nested_X[0], X[nested_idx[0]], nested_y[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 21687/21687 [00:05<00:00, 4161.17it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, -1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, -1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, -1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, -1, 1, 1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, -1, -1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, 1, -1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, -1, 1, 1, -1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, 1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, 1, 1, -1]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 7229/7229 [00:01<00:00, 5681.85it/s]\n"
     ]
    }
   ],
   "source": [
    "d = joblib.load(\"./part1_speaker_recognition/gridsearch/results/part1_hrscv_LogisticRegression.pkl\")\n",
    "\n",
    "nested_X, nested_y, nested_idx = create_nested_dataset(X, y, 2)\n",
    "X_train, X_test, y_train, y_test = train_test_split(nested_X, nested_y)\n",
    "pipeline = d.best_estimator_\n",
    "pipeline.fit(X_train, y_train)\n",
    "print(y_train)\n",
    "y_hat = pipeline.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    oui :\n",
      "    =====\n",
      "    Macro F1-score : 0.8390282363847095\n",
      "    Micro F1-score : 0.9251625397703693\n",
      "    Macro ROC-AUC: 0.8440077843761223\n",
      "    Weighted ROC-AUC: 0.8440077843761223\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.71      0.73      0.72       954\n",
      "           1       0.96      0.95      0.96      6275\n",
      "\n",
      "    accuracy                           0.93      7229\n",
      "   macro avg       0.83      0.84      0.84      7229\n",
      "weighted avg       0.93      0.93      0.93      7229\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'-1': {'precision': 0.7092198581560284,\n",
       "  'recall': 0.7337526205450734,\n",
       "  'f1-score': 0.7212776919113859,\n",
       "  'support': 954},\n",
       " '1': {'precision': 0.9593079141300865,\n",
       "  'recall': 0.9542629482071713,\n",
       "  'f1-score': 0.9567787808580331,\n",
       "  'support': 6275},\n",
       " 'accuracy': 0.9251625397703693,\n",
       " 'macro avg': {'precision': 0.8342638861430575,\n",
       "  'recall': 0.8440077843761223,\n",
       "  'f1-score': 0.8390282363847095,\n",
       "  'support': 7229},\n",
       " 'weighted avg': {'precision': 0.9263041784267734,\n",
       "  'recall': 0.9251625397703693,\n",
       "  'f1-score': 0.9257000647347655,\n",
       "  'support': 7229},\n",
       " 'macro_auc': 0.8440077843761223,\n",
       " 'micro_auc': 0.8440077843761223}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_score(y_test, y_hat, \"oui\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 14551/14551 [00:03<00:00, 4136.53it/s]\n",
      "100%|██████████| 4851/4851 [00:01<00:00, 2690.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    oui :\n",
      "    =====\n",
      "    Macro F1-score : 0.8807504874116063\n",
      "    Micro F1-score : 0.9428983714698\n",
      "    Macro ROC-AUC: 0.8800425185132694\n",
      "    Weighted ROC-AUC: 0.8800425185132694\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.80      0.79      0.79       676\n",
      "           1       0.97      0.97      0.97      4175\n",
      "\n",
      "    accuracy                           0.94      4851\n",
      "   macro avg       0.88      0.88      0.88      4851\n",
      "weighted avg       0.94      0.94      0.94      4851\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'-1': {'precision': 0.7964338781575037,\n",
       "  'recall': 0.7928994082840237,\n",
       "  'f1-score': 0.7946627131208303,\n",
       "  'support': 676},\n",
       " '1': {'precision': 0.9664911440880805,\n",
       "  'recall': 0.967185628742515,\n",
       "  'f1-score': 0.9668382617023825,\n",
       "  'support': 4175},\n",
       " 'accuracy': 0.9428983714698,\n",
       " 'macro avg': {'precision': 0.8814625111227921,\n",
       "  'recall': 0.8800425185132694,\n",
       "  'f1-score': 0.8807504874116063,\n",
       "  'support': 4851},\n",
       " 'weighted avg': {'precision': 0.9427932030926013,\n",
       "  'recall': 0.9428983714698,\n",
       "  'f1-score': 0.9428451322772889,\n",
       "  'support': 4851},\n",
       " 'macro_auc': 0.8800425185132694,\n",
       " 'micro_auc': 0.8800425185132694}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nested_X, nested_y, _ = create_nested_dataset(X, y, 3)\n",
    "X_train, X_test, y_train, y_test = train_test_split(nested_X, nested_y)\n",
    "pipeline = d.best_estimator_\n",
    "pipeline.fit(X_train, y_train)\n",
    "y_hat = pipeline.predict(X_test)\n",
    "print_score(y_test, y_hat, \"oui\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10990/10990 [00:03<00:00, 2870.03it/s]\n",
      "100%|██████████| 3664/3664 [00:01<00:00, 2651.27it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    oui :\n",
      "    =====\n",
      "    Macro F1-score : 0.8915531822400085\n",
      "    Micro F1-score : 0.9475982532751092\n",
      "    Macro ROC-AUC: 0.8934753727791878\n",
      "    Weighted ROC-AUC: 0.8934753727791878\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.81      0.82      0.81       512\n",
      "           1       0.97      0.97      0.97      3152\n",
      "\n",
      "    accuracy                           0.95      3664\n",
      "   macro avg       0.89      0.89      0.89      3664\n",
      "weighted avg       0.95      0.95      0.95      3664\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'-1': {'precision': 0.8088803088803089,\n",
       "  'recall': 0.818359375,\n",
       "  'f1-score': 0.8135922330097087,\n",
       "  'support': 512},\n",
       " '1': {'precision': 0.9704386522568341,\n",
       "  'recall': 0.9685913705583756,\n",
       "  'f1-score': 0.9695141314703081,\n",
       "  'support': 3152},\n",
       " 'accuracy': 0.9475982532751092,\n",
       " 'macro avg': {'precision': 0.8896594805685716,\n",
       "  'recall': 0.8934753727791878,\n",
       "  'f1-score': 0.8915531822400085,\n",
       "  'support': 3664},\n",
       " 'weighted avg': {'precision': 0.9478628138810752,\n",
       "  'recall': 0.9475982532751092,\n",
       "  'f1-score': 0.9477259185849841,\n",
       "  'support': 3664},\n",
       " 'macro_auc': 0.8934753727791878,\n",
       " 'micro_auc': 0.8934753727791878}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nested_X, nested_y, _ = create_nested_dataset(X, y, 4)\n",
    "X_train, X_test, y_train, y_test = train_test_split(nested_X, nested_y)\n",
    "pipeline = d.best_estimator_\n",
    "pipeline.fit(X_train, y_train)\n",
    "y_hat = pipeline.predict(X_test)\n",
    "print_score(y_test, y_hat, \"oui\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43059/43059 [00:03<00:00, 14343.94it/s]\n",
      "100%|██████████| 14354/14354 [00:01<00:00, 9568.79it/s] \n",
      "100%|██████████| 21687/21687 [00:02<00:00, 7656.37it/s]\n",
      "100%|██████████| 7229/7229 [00:01<00:00, 6477.69it/s]\n",
      "100%|██████████| 14551/14551 [00:02<00:00, 6479.50it/s]\n",
      "100%|██████████| 4851/4851 [00:00<00:00, 5438.42it/s]\n",
      "100%|██████████| 10990/10990 [00:02<00:00, 5017.04it/s]\n",
      "100%|██████████| 3664/3664 [00:00<00:00, 3938.96it/s]\n",
      "100%|██████████| 8851/8851 [00:02<00:00, 3734.96it/s]\n",
      "100%|██████████| 2951/2951 [00:01<00:00, 2653.24it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f8f6a442e50>]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjUAAAGdCAYAAADqsoKGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAABUcElEQVR4nO3de1wU9f4/8NfuwrLIwiL3SyhCKio3BUS62EUKpVNmVJ6yVPLosZ9ZyjlHoUjNLljnHNOTlmapHc2jlci3i+JROpkmKoIEqCCCCiLLReUilwV25/eHtbYB6iI47PJ6Ph7zeMTMZ2beH0eZVzPzmZEIgiCAiIiIyMRJxS6AiIiIqDsw1BAREZFZYKghIiIis8BQQ0RERGaBoYaIiIjMAkMNERERmQWGGiIiIjILDDVERERkFizELuB20el0uHDhAmxtbSGRSMQuh4iIiG6CIAior6+Hh4cHpNLrX4vpM6HmwoUL8PLyErsMIiIi6oLS0lLccccd123TZ0KNra0tgKt/KHZ2diJXQ0RERDejrq4OXl5e+vP49fSZUPPrLSc7OzuGGiIiIhNzM4+O8EFhIiIiMgsMNURERGQWGGqIiIjILDDUEBERkVnoUqhZvXo1vL29oVAoEB4ejiNHjnTatrW1FUuXLoWvry8UCgWCgoKQmppq0Mbb2xsSiaTdNGfOHH2b+++/v93y2bNnd6V8IiIiMkNGh5pt27YhLi4OixcvRlZWFoKCghAVFYXKysoO2ycmJmLt2rX44IMPcOLECcyePRuTJk3CsWPH9G0yMjJQXl6un/bs2QMAeOqppwy2NXPmTIN27733nrHlExERkZmSCIIgGLNCeHg4wsLCsGrVKgBX39Tr5eWFuXPnIj4+vl17Dw8PvPbaawZXXWJiYmBtbY3Nmzd3uI958+bh22+/RWFhoX4I1/3334/g4GCsWLHCmHL16urqoFKpUFtbyyHdREREJsKY87dRV2paWlqQmZmJyMjIaxuQShEZGYn09PQO19FoNFAoFAbzrK2tceDAgU73sXnzZrzwwgvtxqR//vnncHJygr+/PxISEtDY2GhM+URERGTGjHr5XnV1NbRaLVxdXQ3mu7q6Ij8/v8N1oqKisHz5cowdOxa+vr5IS0tDcnIytFpth+1TUlJQU1OD6dOnG8x/9tlnMXDgQHh4eCAnJwcLFy5EQUEBkpOTO9yORqOBRqPR/1xXV2dET4mIiMjU9PgbhVeuXImZM2fCz88PEokEvr6+iI2Nxfr16zts/+mnn2LChAnw8PAwmD9r1iz9fwcEBMDd3R3jxo1DUVERfH19220nKSkJb7zxRvd2hoiIiHoto24/OTk5QSaToaKiwmB+RUUF3NzcOlzH2dkZKSkpaGhowLlz55Cfnw+lUgkfH592bc+dO4e9e/fiT3/60w1rCQ8PBwCcPn26w+UJCQmora3VT6WlpTfcJhEREZkuo0KNXC5HSEgI0tLS9PN0Oh3S0tIQERFx3XUVCgU8PT3R1taG7du3Y+LEie3abNiwAS4uLnjkkUduWEt2djYAwN3dvcPlVlZW+u888XtPRERE5s/o209xcXGYNm0aQkNDMXr0aKxYsQINDQ2IjY0FAEydOhWenp5ISkoCABw+fBhlZWUIDg5GWVkZlixZAp1OhwULFhhsV6fTYcOGDZg2bRosLAzLKioqwpYtWxAdHQ1HR0fk5ORg/vz5GDt2LAIDA7vadyIiIrpFjS1tOHr2MtKLL8JGLsNLDw4WrRajQ83kyZNRVVWFRYsWQa1WIzg4GKmpqfqHh0tKSiCVXrsA1NzcjMTERBQXF0OpVCI6OhqbNm2Cvb29wXb37t2LkpISvPDCC+32KZfLsXfvXn2A8vLyQkxMDBITE40tn4iIiG5Bc6sWWeeuhpj0oov4+XwNWrVX3w7joVJgzgN33tQXtXuC0e+pMVV8Tw0REZHxNG1aHCupQXrRRRwqvohjJTVo0eoM2njaW2OMjyPG+DjgiVF3QCbtvlBjzPm7x0c/ERERkeloadMh5/zVEJNefBGZ5y5D02YYYlztrBDh44gIX0dE+DjBy8FatKszv8VQQ0RE1Ie1aXXIKavVX4k5evYymloN3yXnpLRChO/VKzERPo4Y5GTTK0LM7zHUEBER9SFanYDjF2r1V2IyzlxCQ4thiHGwkesDTISvI3ydlb0yxPweQw0REZEZ0+kEnCivw6Hiq1diDp+5hPrmNoM2KmtLhA9yuHo7ydcRQ1xsIe3G52JuF4YaIiIiM6LTCThVWX/1SkzR1RBT29Rq0MZWYYHwQQ4Y88uVmGFudiYZYn6PoYaIiMiECYKAoqor+ttJh4ov4VJDi0EbG7kMYYOu3U4a4aHq1hFKvQVDDRERkQkRBAFnqhv074k5VHwJ1Vc0Bm2sLWUI9e7/y+gkR/h7qmApM+ojAiaJoYaIiKgXEwQBpZeakF5crb8aU1FnGGKsLKQIGdhffyUm8A57yC3MP8T8HkMNERFRL3P+cqM+wBwuvoSymiaD5XKZFCMH2P8yzNoRIwfYw8pCJlK1vQdDDRERkcjUtc0GV2JKLxmGGAupBMFe9vrbSaMG9ofCkiHm9xhqiIiIbrPK+mb9y+4OFV/CmeoGg+UyqQSBd6iujk7ycUSod3/0k/OUfSP8EyIiIuphF69ocKj4kv5qTFGVYYiRSgB/TxUifBwxxtcRYd4OUFrxFG0s/okRERF1s8sNLTh85upVmPSiiyioqDdYLpEAw9zs9LeTwgY5QGVtKVK15oOhhoiI6BbVNrXiyJlL+mdi8tV1EATDNn5utr98yfrqN5Ts+8nFKdaMMdQQEREZqb65FUfPXta/K+b4hVrofhdi7nRR6odYhw9ygKPSSpxi+xCGGiIiohtobGlDxtnL+isxeWW10P4uxfg42SD8lxAzxscBLrYKkartuxhqiIiIfqe5VYvMc9dCzM+lNWj7XYgZ4NBPfyVmjI8j3FQMMWJjqCEioj6vuVWLYyU1V7+dVHQR2aU1aNHqDNp42lvrPwAZ4esIT3trkaqlzjDUEBFRn9PSpsPP52v0X7LOKrkMTZthiHGzU+hHJ0X4OuKO/taQSMzvI5DmhKGGiIjMXqtWh5zztb+87O4ijp69jKZWrUEbJ6WVQYjxduzHEGNiGGqIiMjstGl1OH6hTj866ejZS2hoMQwxjjbyq8Orfwkyvs42DDEmjqGGiIhMnlYn4GR5HQ79EmKOnLmEek2bQRv7fpYIH+Twy5UYJwxxVTLEmBmGGiIiMjk6nYCCinr96KQjZy6htqnVoI2twgLhg64NsR7mZgeplCHGnDHUEBFRrycIAgorr+ivxBwqvojLjYYhRmllgTDv/r88F+OE4R52kDHE9CkMNURE1OsIgoDi6gb9lZjDxRdRfaXFoE0/uQyh3g4Y43P1llKApwoWMqlIFVNvwFBDRESiEwQBJZca9SEmvegiKus1Bm2sLKQI9e6vH50UeIc9LBli6DcYaoiISBSCICDnfC125pUjNU+NcxcbDZbLZVKMHGCvH2YdPMAeVhYykaolU8BQQ0REt41OJ+BY6WXszFUjNU+Nspom/TJLmQTBXvaI+GWY9agB/aGwZIihm8dQQ0REPUqrE5Bx9hJ25ZYj9bgaFXXXbiv1k8vwgJ8Lov3dcf9QZ9hY8bREXce/PURE1O3atDocPnMJO3PLsfu42uAhX1srC4wb5oIJAe64b4gzr8ZQt2GoISKibtHSpsPBomrsylXjvyfUBkOu7RQWeHiEG6ID3HD3nU58NoZ6BEMNERF1WXOrFgcKq7Ezrxx7T1SgrvnaW3wdbOSIGuGK8f7uiPBxhNyCI5WoZ3Xpb9jq1avh7e0NhUKB8PBwHDlypNO2ra2tWLp0KXx9faFQKBAUFITU1FSDNt7e3pBIJO2mOXPm6Ns0Nzdjzpw5cHR0hFKpRExMDCoqKrpSPhER3YKmFi1S89R4ZesxhL61F3/691EkZ5WhrrkNTkorPDdmALb8KRxHXh2HpCcCcd8QZwYaui2MvlKzbds2xMXFYc2aNQgPD8eKFSsQFRWFgoICuLi4tGufmJiIzZs3Y926dfDz88Pu3bsxadIkHDx4ECNHjgQAZGRkQKu99qGxvLw8PPTQQ3jqqaf08+bPn4/vvvsOX375JVQqFV566SU88cQT+Omnn7rSbyIiMkKDpg3/K6jErlw1vs+vNPjCtZudAuP93RAd4I6Qgf35Fl8SjUQQBMGYFcLDwxEWFoZVq1YBAHQ6Hby8vDB37lzEx8e3a+/h4YHXXnvN4KpLTEwMrK2tsXnz5g73MW/ePHz77bcoLCyERCJBbW0tnJ2dsWXLFjz55JMAgPz8fAwbNgzp6ekYM2bMDeuuq6uDSqVCbW0t7OzsjOkyEVGfVNfciu9PVmJnbjn2naqCpk2nX+Zpb43oADeM93fHSC97flOJeowx52+jrtS0tLQgMzMTCQkJ+nlSqRSRkZFIT0/vcB2NRgOFQmEwz9raGgcOHOh0H5s3b0ZcXJz+66mZmZlobW1FZGSkvp2fnx8GDBjQaajRaDTQaK4NG6yrq7v5jhIR9VE1jS3Yc6ICqXlq7C+sRov2WpAZ6NgP0QHumODvhgBPFb9wTb2OUaGmuroaWq0Wrq6uBvNdXV2Rn5/f4TpRUVFYvnw5xo4dC19fX6SlpSE5OdngdtNvpaSkoKamBtOnT9fPU6vVkMvlsLe3b7dftVrd4XaSkpLwxhtv3HzniIj6qItXNNhzogI789Q4eLoabbprF/B9nW1+CTLuGOZuyyBDvVqPj35auXIlZs6cCT8/P0gkEvj6+iI2Nhbr16/vsP2nn36KCRMmwMPD45b2m5CQgLi4OP3PdXV18PLyuqVtEhGZi8r6Zuw+XoFdueU4VHwRv8kx8HOzxQR/d0QHuGGwq614RRIZyahQ4+TkBJlM1m7UUUVFBdzc3Dpcx9nZGSkpKWhubsbFixfh4eGB+Ph4+Pj4tGt77tw57N27F8nJyQbz3dzc0NLSgpqaGoOrNdfbr5WVFaysrIzpHhGRWSuvbUJqnhq78tTIOHsJv32i0t/TDhP8r95a8nFWilck0S0wKtTI5XKEhIQgLS0Njz/+OICrDwqnpaXhpZdeuu66CoUCnp6eaG1txfbt2/H000+3a7Nhwwa4uLjgkUceMZgfEhICS0tLpKWlISYmBgBQUFCAkpISREREGNMFIqI+5fzlRqTmqbEztxxZJTUGy4K97DHB3w0T/N0xwLGfOAUSdSOjbz/FxcVh2rRpCA0NxejRo7FixQo0NDQgNjYWADB16lR4enoiKSkJAHD48GGUlZUhODgYZWVlWLJkCXQ6HRYsWGCwXZ1Ohw0bNmDatGmwsDAsS6VSYcaMGYiLi4ODgwPs7Owwd+5cRERE3NTIJyKivuRsdQN25amxK68cOedr9fMlEiBkQH9MCHDHeH83eNpbi1glUfczOtRMnjwZVVVVWLRoEdRqNYKDg5Gamqp/eLikpARS6bWXLDU3NyMxMRHFxcVQKpWIjo7Gpk2b2j30u3fvXpSUlOCFF17ocL/vv/8+pFIpYmJioNFoEBUVhQ8//NDY8omIzNLpyivYlVuOXXlqnCi/NtpTKgFGD3JAdIA7oka4wdVOcZ2tEJk2o99TY6r4nhoiMieCIKCgoh67cq9ekTlVcUW/TCaV4C5fR4z3d8PDw93gbMvnC8l09dh7aoiISDyCIOD4hTrsyivHrlw1iqsb9MssZRLcfacTov3d8dBwV/S3kYtYKZE4GGqIiHoxQRDw8/la7Motx868cpReatIvk1tIMXawM6ID3DBumCtU1pYiVkokPoYaIqJeRqcTkFVyGTtz1dh9XI2ymmtBRmEpxQNDXTAhwB0P+rlAacVf40S/4r8GIqJeQKsTkHH2kv5h38r6a5956SeX4UE/F0QHuOP+oc7oJ+evbqKO8F8GEZFI2rQ6HCq+hJ155fjvcTWqr7Tol9laWSByuCsm+Lth7BBnKCxlIlZKZBoYaoiIbqOWNh1+KqrGrtxy7DlRgcuNrfplKmtLPDzcFdEB7rjrTkdYWTDIEBmDoYaIqIc1t2qxv7Aau/KuBpn65jb9MkcbOR4e4YYJ/m6I8HWEpUx6nS0R0fUw1BAR9YCmFi32narEzlw10k5WoKFFq1/mbGuF8SPcMCHADaO9HWDBIEPULRhqiIi6SYOmDd/nV2JXXjn+l1+FptZrQcZdpcB4fzdEB7hj1ID+kEklIlZKZJ4YaoiIbkFdcyvSTlZgV64a+05VQdOm0y+7o781on/5zlLwHfaQMsgQ9SiGGiIiI9U0tmDPiQrsylPjQGE1WrTXgoy3Yz9EB7hjgr87/D3tIJEwyBDdLgw1REQ34eIVDf57ogI7c8uRXnQRbbprn80b7KLEBH83TAhwh5+bLYMMkUgYaoiIOlFZ14zdx9XYlafGoeKL+E2OgZ+b7S9XZNww2NVWvCKJSI+hhojoN8prm5Cap8auXDUyzl2C8JsgE+CpwoQAN0zwd8cgJxvxiiSiDjHUEFGfV3qpEal5auzMK8exkhqDZSMH2F+9teTvDi+HfuIUSEQ3haGGiPqks9UN2JlXjl25auSW1ernSyRA6MD+mOB/ddSSh721iFUSkTEYaoiozzhdWY9duWrszFPjZHmdfr5UAoQPckR0gBuiRrjBxU4hYpVE1FUMNURktgRBQEFFPXbmqrErtxyFlVf0y2RSCe7ydUR0gDseGu4KJ6WViJUSUXdgqCEisyIIAo5fqMPO3HLsylPjTHWDfpmlTIJ77nTChAB3PDTMFf1t5CJWSkTdjaGGiEyeIAjILq3RP+xbeqlJv0xuIcV9Q5wRHeCGB/1cobK2FLFSIupJDDVEZNKuaNowb+sx7D1ZqZ+nsJTiQT8XTPB3xwN+LlBa8VcdUV/Af+lEZLLKa5vwwsajOFleB7mFFFEj3BDt74b7hjqjn5y/3oj6Gv6rJyKTlFdWixmfZaCiTgMnpRU+mRaKYC97scsiIhEx1BCRyfnvcTVe2ZqNplYthrgqsX56GO7ozxfjEfV1DDVEZDIEQcCnB87g7Z0nIQjAvYOdsHrKKNgp+PAvETHUEJGJaNPqsOSb49h8qAQAMCV8AN54bAQsZFKRKyOi3oKhhoh6vfrmVszZcgw/nqqCRAK8Fj0MM+4ZBIlEInZpRNSLMNQQUa92/nIjZmw8ioKKelhbyrDyj8F4eISb2GURUS/EUENEvdbPpTWY8dlRVF/RwMXWCp9OC0PAHSqxyyKiXoqhhoh6pV255Zj/RTaaW3UY5m6H9dND4a7iF7OJqHMMNUTUqwiCgLU/FmPZrnwAwIN+LvjXMyP5VmAiuiH+liCiXqNVq8PrKXnYmlEKAJh+lzcSHxnGEU5EdFO69Jti9erV8Pb2hkKhQHh4OI4cOdJp29bWVixduhS+vr5QKBQICgpCampqu3ZlZWV47rnn4OjoCGtrawQEBODo0aP65dOnT4dEIjGYxo8f35XyiagXqm1qxfQNR7A1oxRSCbDk0eFYwiHbRGQEo6/UbNu2DXFxcVizZg3Cw8OxYsUKREVFoaCgAC4uLu3aJyYmYvPmzVi3bh38/Pywe/duTJo0CQcPHsTIkSMBAJcvX8bdd9+NBx54ALt27YKzszMKCwvRv39/g22NHz8eGzZs0P9sZWVlbPlE1AuVXmpE7MYMnK68Ahu5DB88OxIP+rmKXRYRmRiJIAiCMSuEh4cjLCwMq1atAgDodDp4eXlh7ty5iI+Pb9few8MDr732GubMmaOfFxMTA2tra2zevBkAEB8fj59++gn79+/vdL/Tp09HTU0NUlJSjClXr66uDiqVCrW1tbCzs+vSNoio+2Weu4xZ/z6Kiw0tcFcp8Om0MAz34L9RIrrKmPO3Udd1W1pakJmZicjIyGsbkEoRGRmJ9PT0DtfRaDRQKBQG86ytrXHgwAH9z19//TVCQ0Px1FNPwcXFBSNHjsS6devabeuHH36Ai4sLhg4dihdffBEXL17stFaNRoO6ujqDiYh6l29+voBn1h3CxYYW+HvaIWXO3Qw0RNRlRoWa6upqaLVauLoaXhZ2dXWFWq3ucJ2oqCgsX74chYWF0Ol02LNnD5KTk1FeXq5vU1xcjI8++giDBw/G7t278eKLL+Lll1/GZ599pm8zfvx4/Pvf/0ZaWhreffdd7Nu3DxMmTIBWq+1wv0lJSVCpVPrJy8vLmK4SUQ8SBAGrvi/E3P8cQ0ubDpHDXPHFnyPgaqe48cpERJ0w6vbThQsX4OnpiYMHDyIiIkI/f8GCBdi3bx8OHz7cbp2qqirMnDkT33zzDSQSCXx9fREZGYn169ejqakJACCXyxEaGoqDBw/q13v55ZeRkZHR6RWg4uJi+Pr6Yu/evRg3bly75RqNBhqNRv9zXV0dvLy8ePuJSGQtbTokJOdie9Z5AMCf7hmEhOhhkEn5yQMiaq/Hbj85OTlBJpOhoqLCYH5FRQXc3Dp+bbmzszNSUlLQ0NCAc+fOIT8/H0qlEj4+Pvo27u7uGD58uMF6w4YNQ0lJSae1+Pj4wMnJCadPn+5wuZWVFezs7AwmIhJXTWMLnv/0MLZnnYdMKsFbj/sj8Q/DGWiIqFsYFWrkcjlCQkKQlpamn6fT6ZCWlmZw5aYjCoUCnp6eaGtrw/bt2zFx4kT9srvvvhsFBQUG7U+dOoWBAwd2ur3z58/j4sWLcHd3N6YLRCSSs9UNeOLDgzh85hKUVhZYPz0Mz43p/N84EZGxjH4BRFxcHNatW4fPPvsMJ0+exIsvvoiGhgbExsYCAKZOnYqEhAR9+8OHDyM5ORnFxcXYv38/xo8fD51OhwULFujbzJ8/H4cOHcI777yD06dPY8uWLfj444/1I6auXLmCv/3tbzh06BDOnj2LtLQ0TJw4EXfeeSeioqJu9c+AiHpYxtlLmPThTyiuboCnvTW2v3gX7hviLHZZRGRmjH5PzeTJk1FVVYVFixZBrVYjODgYqamp+oeHS0pKIJVey0rNzc1ITExEcXExlEoloqOjsWnTJtjb2+vbhIWFYceOHUhISMDSpUsxaNAgrFixAlOmTAEAyGQy5OTk4LPPPkNNTQ08PDzw8MMP48033+S7aoh6uZRjZVjwVQ5atDoE3aHCummhcLHlA8FE1P2Mfk+NqeJ7aohuL0EQsGJvIVamFQIAJvi7YfnTwbCWy0SujIhMiTHnb377iYi6naZNi4Vf5SAl+wIAYPZ9vlgQNRRSPhBMRD2IoYaIutWlhhb8edNRZJy9DItfRjj9cfQAscsioj6AoYaIuk1R1RW8sDED5y42wlZhgTXPheDuO53ELouI+giGGiLqFulFFzF7cyZqm1rh5WCNDdPDcKeLrdhlEVEfwlBDRLfsy6OleHVHLlq1AkYNsMfHU0PhpOTIRCK6vRhqiKjLdDoB/9xTgNX/KwIA/CHQHf94KggKS45wIqLbj6GGiLqkuVWLv375M77Nufpx2rkP3on5kUM4womIRMNQQ0RGq76iwcx/H8WxkhpYyiRIeiIQT4bcIXZZRNTHMdQQkVEKK+oRuzED5y83QWVtibXPh2CMj6PYZRERMdQQ0c07UFiNFz/PRH1zG7wd+2H99DD4OCvFLouICABDDRHdpP8cKUFiSh60OgFh3v2x9vlQONjIxS6LiEiPoYaIrkunE/Buaj7W/lgMAJg00hPLYgJgZcERTkTUuzDUEFGnmlq0mL8tG6nH1QCA+ZFD8PK4OyGRcIQTEfU+DDVE1KHK+mbM/Owofj5fC7lMir8/FYiJwZ5il0VE1CmGGiJqJ19dhxkbj6Kspgn9+1ni46mhCPN2ELssIqLrYqghIgM/FFTipS3HcEXTBh8nG6yfHgZvJxuxyyIiuiGGGiLS23ToHBb/Xx50AjDGxwFrnguBfT+OcCIi08BQQ0TQ6gS8s/MkPj1wBgDwZMgdeGdSAOQWUpErIyK6eQw1RH1cg6YNr2zNxt6TFQCAv0UNxf+735cjnIjI5DDUEPVh6tpmzPgsA8cv1EFuIcXyp4Pwh0APscsiIuoShhqiPur4hVrM2HgU6rpmONrIsW5aKEYN6C92WUREXcZQQ9QHpZ2swNz/HENjixZ3uiixYXoYvBz6iV0WEdEtYagh6mM2/HQGb357AjoBuOdOJ6yeMgoqa0uxyyIiumUMNUR9RJtWhze/PYHP0s8BAJ4Z7YWlE/1hKeMIJyIyDww1RH3AFU0bXtqShR8KqiCRAPHj/TBrrA9HOBGRWWGoITJzF2qa8MLGDOSr66GwlGLF5GCM93cXuywiom7HUENkxnLO12DGZ0dRVa+Bs60VPpkaiiAve7HLIiLqEQw1RGZq93E1Xtl6DM2tOvi52eLT6WHwtLcWuywioh7DUENkZgRBwCf7z+CdXSchCMB9Q5yx6tmRsFVwhBMRmTeGGiIz0qrVYfHXx7HlcAkA4LkxA7Dk0RGw4AgnIuoDGGqIzERdcyvmfJ6F/YXVkEiAxEeG44W7vTnCiYj6DIYaIjNQeqkRMz7LwKmKK7C2lOFfz4zEQ8NdxS6LiOi26tI16dWrV8Pb2xsKhQLh4eE4cuRIp21bW1uxdOlS+Pr6QqFQICgoCKmpqe3alZWV4bnnnoOjoyOsra0REBCAo0eP6pcLgoBFixbB3d0d1tbWiIyMRGFhYVfKJzIrx0ouY9KHP+FUxRW42lnhy9kRDDRE1CcZHWq2bduGuLg4LF68GFlZWQgKCkJUVBQqKys7bJ+YmIi1a9figw8+wIkTJzB79mxMmjQJx44d07e5fPky7r77blhaWmLXrl04ceIE/vnPf6J//2sf13vvvffwr3/9C2vWrMHhw4dhY2ODqKgoNDc3d6HbROZhZ245/vjxIVRfacFwdzukzLkb/p4qscsiIhKFRBAEwZgVwsPDERYWhlWrVgEAdDodvLy8MHfuXMTHx7dr7+Hhgddeew1z5szRz4uJiYG1tTU2b94MAIiPj8dPP/2E/fv3d7hPQRDg4eGBv/zlL/jrX/8KAKitrYWrqys2btyIP/7xjzesu66uDiqVCrW1tbCzszOmy0S9jiAI+GhfEd5LLQAAjPNzwb+eGQkbK95RJiLzYsz526grNS0tLcjMzERkZOS1DUiliIyMRHp6eofraDQaKBQKg3nW1tY4cOCA/uevv/4aoaGheOqpp+Di4oKRI0di3bp1+uVnzpyBWq022K9KpUJ4ePh191tXV2cwEZmDljYdFm7P0Qea2Lu98fHUUAYaIurzjAo11dXV0Gq1cHU1vF/v6uoKtVrd4TpRUVFYvnw5CgsLodPpsGfPHiQnJ6O8vFzfpri4GB999BEGDx6M3bt348UXX8TLL7+Mzz77DAD02zZmv0lJSVCpVPrJy8vLmK4S9Uq1ja2YvuEIvjh6HlIJsHTiCCx+dARkUo5wIiLq8ZdXrFy5EoMHD4afnx/kcjleeuklxMbGQiq9tmudTodRo0bhnXfewciRIzFr1izMnDkTa9as6fJ+ExISUFtbq59KS0u7oztEoim52IgnPvoJB4suwkYuw6fTwjA1wlvssoiIeg2jQo2TkxNkMhkqKioM5ldUVMDNza3DdZydnZGSkoKGhgacO3cO+fn5UCqV8PHx0bdxd3fH8OHDDdYbNmwYSkquvkDs120bs18rKyvY2dkZTESmKvPcJTz+4U8oqmqAh0qBr168Cw/4uYhdFhFRr2JUqJHL5QgJCUFaWpp+nk6nQ1paGiIiIq67rkKhgKenJ9ra2rB9+3ZMnDhRv+zuu+9GQUGBQftTp05h4MCBAIBBgwbBzc3NYL91dXU4fPjwDfdLZOr+L7sMz6w7jEsNLQjwVCFlzt0Y5s6QTkT0e0Y/WRgXF4dp06YhNDQUo0ePxooVK9DQ0IDY2FgAwNSpU+Hp6YmkpCQAwOHDh1FWVobg4GCUlZVhyZIl0Ol0WLBggX6b8+fPx1133YV33nkHTz/9NI4cOYKPP/4YH3/8MQBAIpFg3rx5eOuttzB48GAMGjQIr7/+Ojw8PPD44493wx8DUe8jCAI++P40lu85BQB4eLgrVvwxGP3kfCCYiKgjRv92nDx5MqqqqrBo0SKo1WoEBwcjNTVV/xBvSUmJwfMyzc3NSExMRHFxMZRKJaKjo7Fp0ybY29vr24SFhWHHjh1ISEjA0qVLMWjQIKxYsQJTpkzRt1mwYAEaGhowa9Ys1NTU4J577kFqamq7kVVE5kDTpkXC9lwkHysDAMwa64P48X6Q8oFgIqJOGf2eGlPF99SQqbjc0II/b87EkTOXIJNK8OZEfzwbPkDssoiIRGHM+ZvXsYl6kTPVDXhhYwbOVDfA1soCq6eMwtghzmKXRURkEhhqiHqJw8UX8efNmahpbIWnvTU2xIZhiKut2GUREZkMhhqiXiA56zwWbs9Bq1ZAsJc91k0NhbOtldhlERGZFIYaIhEJgoD395zCv74/DQB4JMAd/3w6CApLmciVERGZHoYaIpE0t2qx4KscfP3zBQDA/7vfF399eChHOBERdRFDDZEILl7RYNamTGSeuwwLqQTvTArA02H8PhkR0a1gqCG6zU5XXsELGzNQcqkRdgoLrHkuBHfd6SR2WUREJo+hhug2Oni6GrM3Z6KuuQ0DHPph/fQw3OmiFLssIiKzwFBDdJt8kVGKV3fkok0nIGRgf3z8fAgclRzhRETUXRhqiHqYTifg7/8twEc/FAEAHgvywHtPBnKEExFRN2OoIepBza1a/OWLn/FdbjkA4OVxgzE/cjAkEo5wIiLqbgw1RD2kql6Dmf8+iuzSGljKJHg3JhBPjLpD7LKIiMwWQw1RDzhVUY/YDRkoq2mCfT9LrH0uBOE+jmKXRURk1hhqiLrZ/sIq/L/NWajXtGGQkw3WTw/DICcbscsiIjJ7DDVE3WjL4RK8/n950OoEjB7kgLXPhaC/jVzssoiI+gSGGqJuoNUJWLbrJNbtPwMAeGKkJ5JiAmBlwRFORES3C0MN0S1qbGnDvK3Z+O+JCgDAXx4agpcevJMjnIiIbjOGGqJbUFHXjD99dhS5ZbWQW0jx9ycDMTHYU+yyiIj6JIYaoi46WV6HFzZmoLy2GQ42cnz8fAhCvR3ELouIqM9iqCHqgv/lV+KlLVloaNHCx9kGG6aHYaAjRzgREYmJoYbISP9OP4slXx+HTgAifByx5rkQqPpZil0WEVGfx1BDdJO0OgFvfnsCGw+eBQA8HXoH3no8AHILqbiFERERAIYaopvSoGnDy/85hrT8SgDAgvFD8eJ9vhzhRETUizDUEN1AeW0TZmw8ihPldbCykGL508F4JNBd7LKIiOh3GGqIriOvrBYzPstARZ0GTko51k0NxcgB/cUui4iIOsBQQ9SJPScq8PJ/jqGpVYvBLkqsnx4GL4d+YpdFRESdYKgh+h1BEPDpgTN4e+dJCAJw72AnrJ4yCnYKjnAiIurNGGqIfqNNq8OSb45j86ESAMCz4QPwxmMjYCnjCCciot6OoYboF/XNrXhpyzHsO1UFiQR4dcIw/OneQRzhRERkIhhqiACU1TRhxsYM5KvrYW0pw4o/BiNqhJvYZRERkREYaqjPK666gskfH0JVvQbOtlZYPy0MAXeoxC6LiIiMxFBDfZpWJ+CvX/6MqnoN/NxssX56GDzsrcUui4iIuqBLTz+uXr0a3t7eUCgUCA8Px5EjRzpt29raiqVLl8LX1xcKhQJBQUFITU01aLNkyRJIJBKDyc/Pz6DN/fff367N7Nmzu1I+kd6/088iq6QGSisLbIhloCEiMmVGX6nZtm0b4uLisGbNGoSHh2PFihWIiopCQUEBXFxc2rVPTEzE5s2bsW7dOvj5+WH37t2YNGkSDh48iJEjR+rbjRgxAnv37r1WmEX70mbOnImlS5fqf+7Xj+8Moa4rvdSI91ILAAAJ0X5wVzHQEBGZMqOv1CxfvhwzZ85EbGwshg8fjjVr1qBfv35Yv359h+03bdqEV199FdHR0fDx8cGLL76I6Oho/POf/zRoZ2FhATc3N/3k5OTUblv9+vUzaGNnZ2ds+UQArr6L5tUduWhq1WL0IAc8EzZA7JKIiOgWGRVqWlpakJmZicjIyGsbkEoRGRmJ9PT0DtfRaDRQKBQG86ytrXHgwAGDeYWFhfDw8ICPjw+mTJmCkpKSdtv6/PPP4eTkBH9/fyQkJKCxsbHTWjUaDerq6gwmol9tzyrD/sJqWFlIseyJAEilHLZNRGTqjAo11dXV0Gq1cHV1NZjv6uoKtVrd4TpRUVFYvnw5CgsLodPpsGfPHiQnJ6O8vFzfJjw8HBs3bkRqaio++ugjnDlzBvfeey/q6+v1bZ599lls3rwZ//vf/5CQkIBNmzbhueee67TWpKQkqFQq/eTl5WVMV8mMVdVr8Oa3JwAA8x8aAh9npcgVERFRd+jx0U8rV67EzJkz4efnB4lEAl9fX8TGxhrcrpowYYL+vwMDAxEeHo6BAwfiiy++wIwZMwAAs2bN0rcJCAiAu7s7xo0bh6KiIvj6+rbbb0JCAuLi4vQ/19XVMdgQAGDJ18dR29QKf087/OmeQWKXQ0RE3cSoKzVOTk6QyWSoqKgwmF9RUQE3t45fVObs7IyUlBQ0NDTg3LlzyM/Ph1KphI+PT6f7sbe3x5AhQ3D69OlO24SHhwNAp22srKxgZ2dnMBHtPq7Gd7nlkEkleDcmEBb8/AERkdkw6je6XC5HSEgI0tLS9PN0Oh3S0tIQERFx3XUVCgU8PT3R1taG7du3Y+LEiZ22vXLlCoqKiuDu7t5pm+zsbAC4bhui36ptasXrKXkAgNn3+WCEB1+wR0RkToy+/RQXF4dp06YhNDQUo0ePxooVK9DQ0IDY2FgAwNSpU+Hp6YmkpCQAwOHDh1FWVobg4GCUlZVhyZIl0Ol0WLBggX6bf/3rX/Hoo49i4MCBuHDhAhYvXgyZTIZnnnkGAFBUVIQtW7YgOjoajo6OyMnJwfz58zF27FgEBgZ2x58D9QFJO0+isl4DH2cbzH1wsNjlEBFRNzM61EyePBlVVVVYtGgR1Go1goODkZqaqn94uKSkBFLptQtAzc3NSExMRHFxMZRKJaKjo7Fp0ybY29vr25w/fx7PPPMMLl68CGdnZ9xzzz04dOgQnJ2dAVy9QrR37159gPLy8kJMTAwSExNvsfvUVxw8XY2tGaUAgHdjAqGwlIlcERERdTeJIAiC2EXcDnV1dVCpVKitreXzNX1MU4sWUSt+RMmlRkyNGIilE/3FLomIiG6SMedvPiVJZm/5ngKUXGqEh0qBBeP9brwCERGZJIYaMms/l9bg0wNnAABvTwqA0orfcCUiMlcMNWS2Wtp0WLg9BzoBeDzYAw/4tf82GRERmQ+GGjJba/YVIV9dDwcbORY9OkLscoiIqIcx1JBZKqyox6rvr76YccljI+BgIxe5IiIi6mkMNWR2tDoBC7fnoEWrwzg/FzwayBc0EhH1BQw1ZHb+nX4WWSU1UFpZ4K1J/pBI+AVuIqK+gKGGzErppUb8fXcBACAh2g/uKmuRKyIiotuFoYbMhiAIeHVHLhpbtBg9yAHPhA0QuyQiIrqNGGrIbGzPKsP+wmpYWUix7IkASKW87URE1Jcw1JBZqKrX4M1vTwAA5kUOgY+zUuSKiIjodmOoIbOw5OvjqG1qhb+nHWbeO0jscoiISAQMNWTydh9X47vccsikErwbEwgLGf9aExH1RfztTyattqkVr6fkAQD+PNYHIzxUIldERERiYaghk5a08yQq6zXwcbLBy+MGi10OERGJiKGGTNbB09XYmlEKAHj3yUAoLGUiV0RERGJiqCGT1NSiRXxyLgDg+TEDEebtIHJFREQkNoYaMknL9xSg5FIjPFQKLBg/VOxyiIioF2CoIZPzc2kNPj1wBgDw9qQA2CosRa6IiIh6A4YaMiktbTos3J4DnQA8HuyBB/xcxC6JiIh6CYYaMilr9hUhX10PBxs5Fj06QuxyiIioF2GoIZNRWFGPVd+fBgAsfnQ4HGzkIldERES9CUMNmQStTsDC7Tlo0eowzs8FjwV5iF0SERH1Mgw1ZBL+nX4WWSU1UFpZ4K1J/pBI+AVuIiIyxFBDvV7ppUb8fXcBACB+gh/cVdYiV0RERL0RQw31aoIg4NUduWhs0WL0IAc8O3qA2CUREVEvxVBDvdr2rDLsL6yGlYUUy54IgFTK205ERNQxhhrqtarqNXjz2xMAgHmRQ+DjrBS5IiIi6s0YaqjXWvL1cdQ2tcLf0w4z7x0kdjlERNTLMdRQr7T7uBrf5ZZDJpXg3ZhAWMj4V5WIiK6PZwrqdWqbWvF6Sh4A4M9jfTDCQyVyRUREZAoYaqjXSdp5EpX1Gvg42eDlcYPFLoeIiExEl0LN6tWr4e3tDYVCgfDwcBw5cqTTtq2trVi6dCl8fX2hUCgQFBSE1NRUgzZLliyBRCIxmPz8/AzaNDc3Y86cOXB0dIRSqURMTAwqKiq6Uj71YgdPV2NrRikAYFlMIBSWMpErIiIiU2F0qNm2bRvi4uKwePFiZGVlISgoCFFRUaisrOywfWJiItauXYsPPvgAJ06cwOzZszFp0iQcO3bMoN2IESNQXl6unw4cOGCwfP78+fjmm2/w5ZdfYt++fbhw4QKeeOIJY8unXqypRYv45FwAwPNjBmL0IAeRKyIiIlMiEQRBMGaF8PBwhIWFYdWqVQAAnU4HLy8vzJ07F/Hx8e3ae3h44LXXXsOcOXP082JiYmBtbY3NmzcDuHqlJiUlBdnZ2R3us7a2Fs7OztiyZQuefPJJAEB+fj6GDRuG9PR0jBkz5oZ119XVQaVSoba2FnZ2dsZ0mW6Tt787gXX7z8BDpcDu+WNhq7AUuyQiIhKZMedvo67UtLS0IDMzE5GRkdc2IJUiMjIS6enpHa6j0WigUCgM5llbW7e7ElNYWAgPDw/4+PhgypQpKCkp0S/LzMxEa2urwX79/PwwYMCA6+63rq7OYKLe6+fSGnx64AwA4O1JAQw0RERkNKNCTXV1NbRaLVxdXQ3mu7q6Qq1Wd7hOVFQUli9fjsLCQuh0OuzZswfJyckoLy/XtwkPD8fGjRuRmpqKjz76CGfOnMG9996L+vp6AIBarYZcLoe9vf1N7zcpKQkqlUo/eXl5GdNVuo1a2nRYuD0HOgF4PNgDD/i5iF0SERGZoB4f/bRy5UoMHjwYfn5+kMvleOmllxAbGwup9NquJ0yYgKeeegqBgYGIiorCzp07UVNTgy+++KLL+01ISEBtba1+Ki0t7Y7uUA9Ys68I+ep6ONjIsejREWKXQ0REJsqoUOPk5ASZTNZu1FFFRQXc3Nw6XMfZ2RkpKSloaGjAuXPnkJ+fD6VSCR8fn073Y29vjyFDhuD06dMAADc3N7S0tKCmpuam92tlZQU7OzuDiXqfwop6rPr+6nFe/OhwONjIRa6IiIhMlVGhRi6XIyQkBGlpafp5Op0OaWlpiIiIuO66CoUCnp6eaGtrw/bt2zFx4sRO2165cgVFRUVwd3cHAISEhMDS0tJgvwUFBSgpKbnhfqn30uoELNyegxatDuP8XPBYkIfYJRERkQmzMHaFuLg4TJs2DaGhoRg9ejRWrFiBhoYGxMbGAgCmTp0KT09PJCUlAQAOHz6MsrIyBAcHo6ysDEuWLIFOp8OCBQv02/zrX/+KRx99FAMHDsSFCxewePFiyGQyPPPMMwAAlUqFGTNmIC4uDg4ODrCzs8PcuXMRERFxUyOfqHfalH4WWSU1UFpZ4K1J/pBI+AVuIiLqOqNDzeTJk1FVVYVFixZBrVYjODgYqamp+oeHS0pKDJ6XaW5uRmJiIoqLi6FUKhEdHY1NmzYZPPR7/vx5PPPMM7h48SKcnZ1xzz334NChQ3B2dta3ef/99yGVShETEwONRoOoqCh8+OGHt9B1ElPppUa8t7sAABA/wQ/uKmuRKyIiIlNn9HtqTBXfU9N7CIKAqeuPYH9hNUYPcsDWmWMglfIqDRERtddj76kh6g7JWWXYX1gNuYUUy54IYKAhIqJuwVBDt1VVvQZLvz0BAJgfOQQ+zkqRKyIiInPBUEO31ZKvj6O2qRX+nnaYee8gscshIiIzwlBDt83u42p8l1sOmVSCd2MCYSHjXz8iIuo+PKvQbVHb1IrXU/IAAH8e64MRHiqRKyIiInPDUEO3RdLOk6is18DHyQYvjxssdjlERGSGGGqoxx08XY2tGVe/vbUsJhAKS5nIFRERkTliqKEe1dSiRXxyLgDg+TEDMXqQg8gVERGRuWKooR61fE8BSi41wl2lwILxQ8Uuh4iIzBhDDfWYn0tr8OmBMwCAtyf5w1ZhKXJFRERkzhhqqEe0tOmwcHsOdALweLAHHvRzFbskIiIycww11CPW7itCvroeDjZyLHp0hNjlEBFRH8BQQ92usKIeH3x/GgCw+NHhcLCRi1wRERH1BQw11K20OgELt+egRavDOD8XPBbkIXZJRETURzDUULfalH4WWSU1UFpZ4K1J/pBI+AVuIiK6PRhqqNuUXmrEe7sLAADxE/zgrrIWuSIiIupLGGqoWwiCgFd35KKxRYvRgxzw7OgBYpdERER9DEMNdYvkrDLsL6yG3EKKZU8EQCrlbSciIrq9GGrollXVa7D02xMAgPmRQ+DjrBS5IiIi6osYauiWLfn6OGqbWjHCww4z7x0kdjlERNRHMdTQLdl9XI3vcsshk0rwbkwgLGT8K0VEROLgGYi6rLapFa+n5AEA/jzWB/6eKpErIiKivoyhhrps2a6TqKzXwMfJBi+PGyx2OURE1Mcx1FCXHDxdjf8cKQUALIsJhMJSJnJFRETU1zHUkNGaWrSIT84FADw3ZgBGD3IQuSIiIiKGGuqC9/eeQsmlRrirFFg43k/scoiIiAAw1JCRfi6twSf7iwEAb0/yh63CUuSKiIiIrmKooZvW0qbDwu050AnAxGAPPOjnKnZJREREegw1dNPW7itCvroeDjZyLPrDcLHLISIiMsBQQzelsKIeH3x/GgCw+NHhcFRaiVwRERGRIYYauiGtTsDC7Tlo0erwoJ8LHgvyELskIiKidroUalavXg1vb28oFAqEh4fjyJEjnbZtbW3F0qVL4evrC4VCgaCgIKSmpnbaftmyZZBIJJg3b57B/Pvvvx8SicRgmj17dlfKJyNtSj+LrJIaKK0s8Nbj/pBI+AVuIiLqfYwONdu2bUNcXBwWL16MrKwsBAUFISoqCpWVlR22T0xMxNq1a/HBBx/gxIkTmD17NiZNmoRjx461a5uRkYG1a9ciMDCww23NnDkT5eXl+um9994ztnwy0vnLjXhvdwEAIH6CHzzsrUWuiIiIqGNGh5rly5dj5syZiI2NxfDhw7FmzRr069cP69ev77D9pk2b8OqrryI6Oho+Pj548cUXER0djX/+858G7a5cuYIpU6Zg3bp16N+/f4fb6tevH9zc3PSTnZ2dseWTEQRBwKs78tDYosXoQQ54dvQAsUsiIiLqlFGhpqWlBZmZmYiMjLy2AakUkZGRSE9P73AdjUYDhUJhMM/a2hoHDhwwmDdnzhw88sgjBtv+vc8//xxOTk7w9/dHQkICGhsbjSmfjJScVYYfT1VBbiHFsicCIJXythMREfVeFsY0rq6uhlarhaur4ftJXF1dkZ+f3+E6UVFRWL58OcaOHQtfX1+kpaUhOTkZWq1W32br1q3IyspCRkZGp/t+9tlnMXDgQHh4eCAnJwcLFy5EQUEBkpOTO2yv0Wig0Wj0P9fV1RnT1T6vql6Dpd+eAADMixwMH2elyBURERFdn1GhpitWrlyJmTNnws/PDxKJBL6+voiNjdXfriotLcUrr7yCPXv2tLui81uzZs3S/3dAQADc3d0xbtw4FBUVwdfXt137pKQkvPHGG93foT5iyTfHUdvUihEedph5r4/Y5RAREd2QUbefnJycIJPJUFFRYTC/oqICbm5uHa7j7OyMlJQUNDQ04Ny5c8jPz4dSqYSPz9UTZWZmJiorKzFq1ChYWFjAwsIC+/btw7/+9S9YWFgYXNH5rfDwcADA6dOnO1yekJCA2tpa/VRaWmpMV/u03cfV+C6nHDKpBO/GBMJSxpH/RETU+xl1tpLL5QgJCUFaWpp+nk6nQ1paGiIiIq67rkKhgKenJ9ra2rB9+3ZMnDgRADBu3Djk5uYiOztbP4WGhmLKlCnIzs6GTCbrcHvZ2dkAAHd39w6XW1lZwc7OzmCiG6ttasXrKXkAgFljfeDvqRK5IiIioptj9O2nuLg4TJs2DaGhoRg9ejRWrFiBhoYGxMbGAgCmTp0KT09PJCUlAQAOHz6MsrIyBAcHo6ysDEuWLIFOp8OCBQsAALa2tvD39zfYh42NDRwdHfXzi4qKsGXLFkRHR8PR0RE5OTmYP38+xo4d2+nwb+qaZbtOorJeAx8nG7wybrDY5RAREd00o0PN5MmTUVVVhUWLFkGtViM4OBipqan6h4dLSkoglV67ANTc3IzExEQUFxdDqVQiOjoamzZtgr29/U3vUy6XY+/evfoA5eXlhZiYGCQmJhpbPl3HwdPV+M+Rq7fplsUEQmHZ8VUyIiKi3kgiCIIgdhG3Q11dHVQqFWpra3krqgNNLVpErfgRJZca8dyYAXjr8QCxSyIiIjLq/M0nQAkA8P7eUyi51Ah3lQILx/uJXQ4REZHRGGoIP5fW4JP9xQCAtyf5w1ZhKXJFRERExmOo6eNa2nRYuD0HOgGYGOyBB/1cb7wSERFRL8RQ08et3VeEfHU9HGzkWPSH4WKXQ0RE1GUMNX3Y6cp6fPD91ZcXLn50OByVViJXRERE1HUMNX2UVidgwVc5aNHq8KCfCx4L8hC7JCIiolvCUNNHbUo/i6ySGiitLPDW4/6QSPgFbiIiMm0MNX3Q+cuNeG93AQBg4QQ/eNhbi1wRERHRrWOo6WMEQcCrO/LQ2KLFaG8HTBk9QOySiIiIugVDTR+TnFWGH09VQW4hxbKYAEilvO1ERETmgaGmD6mq12DptycAAPMiB8PHWSlyRURERN2HoaYPWfLNcdQ2tWKEhx1m3usjdjlERETdiqGmj9h9XI3vcsohk0rwbkwgLGU89EREZF54ZusDapta8XpKHgBg1lgf+HuqRK6IiIio+zHU9AHLdp1EZb0GPk42eGXcYLHLISIi6hEMNWbuYFE1/nOkFACQ9EQAFJYykSsiIiLqGQw1ZqypRYuE5FwAwHNjBiDcx1HkioiIiHoOQ40Ze3/vKZy72Ah3lQILx/uJXQ4REVGPYqgxUz+X1uCT/cUAgLcn+cNWYSlyRURERD2LocYMtbTpsHB7DnQCMDHYAw/6uYpdEhERUY9jqDFDa/cVIV9dj/79LLHoD8PFLoeIiOi2YKgxM6cr6/HB96cBAEseGwFHpZXIFREREd0eDDVmRKsTsOCrHLRodXjQzwWPBXmIXRIREdFtw1BjRjaln0VWSQ2UVhZ463F/SCT8AjcREfUdDDVm4vzlRry3uwAAsHCCHzzsrUWuiIiI6PZiqDEDgiDg1R15aGzRYrS3A6aMHiB2SURERLcdQ40ZSM4qw4+nqiC3kGJZTACkUt52IiKivoehxsRV1Wvw5ncnAADzIgfDx1kpckVERETiYKgxcUu+OY6axlaM8LDDzHt9xC6HiIhINAw1Juy/x9X4LqccMqkE78YEwlLGw0lERH0Xz4ImqrapFa//Xx4AYNZYH/h7qkSuiIiISFwMNSZq2a6TqKjTYJCTDV4ZN1jscoiIiETXpVCzevVqeHt7Q6FQIDw8HEeOHOm0bWtrK5YuXQpfX18oFAoEBQUhNTW10/bLli2DRCLBvHnzDOY3Nzdjzpw5cHR0hFKpRExMDCoqKrpSvsk7WFSN/xwpBQAseyIACkuZyBURERGJz+hQs23bNsTFxWHx4sXIyspCUFAQoqKiUFlZ2WH7xMRErF27Fh988AFOnDiB2bNnY9KkSTh27Fi7thkZGVi7di0CAwPbLZs/fz6++eYbfPnll9i3bx8uXLiAJ554wtjyTV5TixYJybkAgOfGDEC4j6PIFREREfUOEkEQBGNWCA8PR1hYGFatWgUA0Ol08PLywty5cxEfH9+uvYeHB1577TXMmTNHPy8mJgbW1tbYvHmzft6VK1cwatQofPjhh3jrrbcQHByMFStWAABqa2vh7OyMLVu24MknnwQA5OfnY9iwYUhPT8eYMWNuWHddXR1UKhVqa2thZ2dnTJd7lXd2nsTHPxbDXaXAf+ePha3CUuySiIiIeowx52+jrtS0tLQgMzMTkZGR1zYglSIyMhLp6ekdrqPRaKBQKAzmWVtb48CBAwbz5syZg0ceecRg27/KzMxEa2urwTI/Pz8MGDDguvutq6szmEzdz6U1+GR/MQDg7Un+DDRERES/YVSoqa6uhlarhaurq8F8V1dXqNXqDteJiorC8uXLUVhYCJ1Ohz179iA5ORnl5eX6Nlu3bkVWVhaSkpI63IZarYZcLoe9vf1N7zcpKQkqlUo/eXl5GdHT3qelTYeF23OgE4CJwR540M/1xisRERH1IT0++mnlypUYPHgw/Pz8IJfL8dJLLyE2NhZS6dVdl5aW4pVXXsHnn3/e7orOrUhISEBtba1+Ki0t7bZti2HtviLkq+vRv58lFv1huNjlEBER9TpGhRonJyfIZLJ2o44qKirg5ubW4TrOzs5ISUlBQ0MDzp07h/z8fCiVSvj4XH37bWZmJiorKzFq1ChYWFjAwsIC+/btw7/+9S9YWFhAq9XCzc0NLS0tqKmpuen9WllZwc7OzmAyVacr6/HB96cBAEseGwFHpZXIFREREfU+RoUauVyOkJAQpKWl6efpdDqkpaUhIiLiuusqFAp4enqira0N27dvx8SJEwEA48aNQ25uLrKzs/VTaGgopkyZguzsbMhkMoSEhMDS0tJgvwUFBSgpKbnhfk2dVidgwVc5aNHq8MBQZzwW5CF2SURERL2ShbErxMXFYdq0aQgNDcXo0aOxYsUKNDQ0IDY2FgAwdepUeHp66p+POXz4MMrKyhAcHIyysjIsWbIEOp0OCxYsAADY2trC39/fYB82NjZwdHTUz1epVJgxYwbi4uLg4OAAOzs7zJ07FxERETc18smUbUo/i6ySGtjIZXh7UgAkEn6Bm4iIqCNGh5rJkyejqqoKixYtglqtRnBwMFJTU/UPD5eUlOiflwGuvjQvMTERxcXFUCqViI6OxqZNm9o99Hsj77//PqRSKWJiYqDRaBAVFYUPP/zQ2PJNyvnLjXhvdwEAID56GDzsrUWuiIiIqPcy+j01psrU3lMjCAKmbcjAj6eqMNrbAVtnjYFUyqs0RETUt/TYe2ro9knOKsOPp6ogt5AiKSaAgYaIiOgGGGp6oap6Dd787gQAYF7kYPg6K0WuiIiIqPdjqOmFlnxzHDWNrRjuboeZ9/qIXQ4REZFJYKjpZf57XI3vcsohk0rw3pOBsJTxEBEREd0MnjF7kdqmVrz+f3kAgFljfeDvqRK5IiIiItPBUNOLLNt1EhV1GgxyssEr4waLXQ4REZFJYajpJQ4WVeM/R65+n2rZEwFQWMpEroiIiMi0MNT0Ak0tWiQk5wIApoQPQLiPo8gVERERmR6Gml7g/b2ncO5iI9xVCsRP8BO7HCIiIpPEUCOynPM1+GR/MQDg7Un+sFVYilwRERGRaWKoEVFLmw4LvsqBTgAeC/LAg36uYpdERERkshhqRLR2XxHy1fXo388Six8dLnY5REREJo2hRiSnK+vxwfenAQBLHhsBR6WVyBURERGZNoYaEWh1AhZ8lYMWrQ4PDHXGY0EeYpdERERk8hhqRLAp/SyySmpgI5fh7UkBkEj4BW4iIqJbxVBzm52/3Ij3dhcAAOIn+MHD3lrkioiIiMwDQ81tJAgCXt2Rh8YWLcK8+2NK+ECxSyIiIjIbDDW3UXJWGX48VQW5hRTLYgIhlfK2ExERUXdhqLlNquo1ePO7EwCAV8YNhq+zUuSKiIiIzAtDzW2y5JvjqGlsxXB3O8wa6yN2OURERGaHoeY2+O9xNb7LKYdMKsF7TwbCUsY/diIiou7Gs2sPq21qxev/lwcAmHmvD/w9VSJXREREZJ4YanrYsl0nUVGnwSAnG8yLHCx2OURERGaLoaYHHSyqxn+OlAIAlj0RAIWlTOSKiIiIzBdDTQ9patEiITkXADAlfADCfRxFroiIiMi8MdT0kPf3nsK5i41wVykQP8FP7HKIiIjMHkNND8g5X4NP9hcDAN563B+2CkuRKyIiIjJ/DDXdrKVNhwVf5UAnAI8FeWDcMFexSyIiIuoTGGq62dp9RchX16N/P0ssfnS42OUQERH1GQw13eh0ZT0++P40AGDxoyPgqLQSuSIiIqK+g6Gmm2h1AhZ8lYMWrQ4PDHXGxGAPsUsiIiLqUxhqusmm9LPIKqmBjVyGtycFQCLhF7iJiIhupy6FmtWrV8Pb2xsKhQLh4eE4cuRIp21bW1uxdOlS+Pr6QqFQICgoCKmpqQZtPvroIwQGBsLOzg52dnaIiIjArl27DNrcf//9kEgkBtPs2bO7Un63O3+5Ee/tLgAAxE/wg4e9tcgVERER9T1Gh5pt27YhLi4OixcvRlZWFoKCghAVFYXKysoO2ycmJmLt2rX44IMPcOLECcyePRuTJk3CsWPH9G3uuOMOLFu2DJmZmTh69CgefPBBTJw4EcePHzfY1syZM1FeXq6f3nvvPWPL73aCIODVHXlobNEizLs/poQPFLskIiKiPkkiCIJgzArh4eEICwvDqlWrAAA6nQ5eXl6YO3cu4uPj27X38PDAa6+9hjlz5ujnxcTEwNraGps3b+50Pw4ODvj73/+OGTNmALh6pSY4OBgrVqwwply9uro6qFQq1NbWws7Orkvb6Mju42r8eVMm5BZS7HrlXvg6K7tt20RERH2dMedvo67UtLS0IDMzE5GRkdc2IJUiMjIS6enpHa6j0WigUCgM5llbW+PAgQMdttdqtdi6dSsaGhoQERFhsOzzzz+Hk5MT/P39kZCQgMbGxk5r1Wg0qKurM5h6wjg/Fywc74e/PTyUgYaIiEhEFsY0rq6uhlarhaur4QvlXF1dkZ+f3+E6UVFRWL58OcaOHQtfX1+kpaUhOTkZWq3WoF1ubi4iIiLQ3NwMpVKJHTt2YPjwa+95efbZZzFw4EB4eHggJycHCxcuREFBAZKTkzvcb1JSEt544w1jutclFjIpXrzft8f3Q0RERNdnVKjpipUrV2LmzJnw8/ODRCKBr68vYmNjsX79eoN2Q4cORXZ2Nmpra/HVV19h2rRp2Ldvnz7YzJo1S982ICAA7u7uGDduHIqKiuDr2z5UJCQkIC4uTv9zXV0dvLy8eqiXREREJDajbj85OTlBJpOhoqLCYH5FRQXc3Nw6XMfZ2RkpKSloaGjAuXPnkJ+fD6VSCR8fH4N2crkcd955J0JCQpCUlISgoCCsXLmy01rCw8MBAKdPn+5wuZWVlX401a8TERERmS+jQo1cLkdISAjS0tL083Q6HdLS0to9//J7CoUCnp6eaGtrw/bt2zFx4sTrttfpdNBoNJ0uz87OBgC4u7vffAeIiIjIbBl9+ykuLg7Tpk1DaGgoRo8ejRUrVqChoQGxsbEAgKlTp8LT0xNJSUkAgMOHD6OsrAzBwcEoKyvDkiVLoNPpsGDBAv02ExISMGHCBAwYMAD19fXYsmULfvjhB+zevRsAUFRUhC1btiA6OhqOjo7IycnB/PnzMXbsWAQGBnbHnwMRERGZOKNDzeTJk1FVVYVFixZBrVYjODgYqamp+oeHS0pKIJVeuwDU3NyMxMREFBcXQ6lUIjo6Gps2bYK9vb2+TWVlJaZOnYry8nKoVCoEBgZi9+7deOihhwBcvUK0d+9efYDy8vJCTEwMEhMTb7H7REREZC6Mfk+Nqeqp99QQERFRz+mx99QQERER9VYMNURERGQWGGqIiIjILDDUEBERkVlgqCEiIiKzwFBDREREZoGhhoiIiMxCj3/Qsrf49XU8dXV1IldCREREN+vX8/bNvFavz4Sa+vp6AOCXuomIiExQfX09VCrVddv0mTcK63Q6XLhwAba2tpBIJN267bq6Onh5eaG0tNQs31Zs7v0DzL+P7J/pM/c+sn+mr6f6KAgC6uvr4eHhYfAZpo70mSs1UqkUd9xxR4/uw87Ozmz/sgLm3z/A/PvI/pk+c+8j+2f6eqKPN7pC8ys+KExERERmgaGGiIiIzAJDTTewsrLC4sWLYWVlJXYpPcLc+weYfx/ZP9Nn7n1k/0xfb+hjn3lQmIiIiMwbr9QQERGRWWCoISIiIrPAUENERERmgaGGiIiIzAJDzQ38+OOPePTRR+Hh4QGJRIKUlJQbrvPDDz9g1KhRsLKywp133omNGzf2eJ23wtg+/vDDD5BIJO0mtVp9ewo2UlJSEsLCwmBrawsXFxc8/vjjKCgouOF6X375Jfz8/KBQKBAQEICdO3fehmqN15X+bdy4sd3xUygUt6li43z00UcIDAzUv9ArIiICu3btuu46pnLsfmVsH03p+HVk2bJlkEgkmDdv3nXbmdpx/NXN9M/UjuGSJUva1evn53fddcQ4fgw1N9DQ0ICgoCCsXr36ptqfOXMGjzzyCB544AFkZ2dj3rx5+NOf/oTdu3f3cKVdZ2wff1VQUIDy8nL95OLi0kMV3pp9+/Zhzpw5OHToEPbs2YPW1lY8/PDDaGho6HSdgwcP4plnnsGMGTNw7NgxPP7443j88ceRl5d3Gyu/OV3pH3D1rZ+/PX7nzp27TRUb54477sCyZcuQmZmJo0eP4sEHH8TEiRNx/PjxDtub0rH7lbF9BEzn+P1eRkYG1q5di8DAwOu2M8XjCNx8/wDTO4YjRowwqPfAgQOdthXt+Al00wAIO3bsuG6bBQsWCCNGjDCYN3nyZCEqKqoHK+s+N9PH//3vfwIA4fLly7elpu5WWVkpABD27dvXaZunn35aeOSRRwzmhYeHC3/+8597urxbdjP927Bhg6BSqW5fUd2sf//+wieffNLhMlM+dr91vT6a6vGrr68XBg8eLOzZs0e47777hFdeeaXTtqZ4HI3pn6kdw8WLFwtBQUE33V6s48crNd0sPT0dkZGRBvOioqKQnp4uUkU9Jzg4GO7u7njooYfw008/iV3OTautrQUAODg4dNrGlI/jzfQPAK5cuYKBAwfCy8vrhlcFegutVoutW7eioaEBERERHbYx5WMH3FwfAdM8fnPmzMEjjzzS7vh0xBSPozH9A0zvGBYWFsLDwwM+Pj6YMmUKSkpKOm0r1vHrMx+0vF3UajVcXV0N5rm6uqKurg5NTU2wtrYWqbLu4+7ujjVr1iA0NBQajQaffPIJ7r//fhw+fBijRo0Su7zr0ul0mDdvHu6++274+/t32q6z49hbnxv61c32b+jQoVi/fj0CAwNRW1uLf/zjH7jrrrtw/PjxHv/wa1fk5uYiIiICzc3NUCqV2LFjB4YPH95hW1M9dsb00dSOHwBs3boVWVlZyMjIuKn2pnYcje2fqR3D8PBwbNy4EUOHDkV5eTneeOMN3HvvvcjLy4OtrW279mIdP4YaMtrQoUMxdOhQ/c933XUXioqK8P7772PTpk0iVnZjc+bMQV5e3nXvBZuym+1fRESEwVWAu+66C8OGDcPatWvx5ptv9nSZRhs6dCiys7NRW1uLr776CtOmTcO+ffs6PembImP6aGrHr7S0FK+88gr27NnTqx+G7aqu9M/UjuGECRP0/x0YGIjw8HAMHDgQX3zxBWbMmCFiZYYYarqZm5sbKioqDOZVVFTAzs7OLK7SdGb06NG9Pii89NJL+Pbbb/Hjjz/e8P+EOjuObm5uPVniLTGmf79naWmJkSNH4vTp0z1U3a2Ry+W48847AQAhISHIyMjAypUrsXbt2nZtTfHYAcb18fd6+/HLzMxEZWWlwZVcrVaLH3/8EatWrYJGo4FMJjNYx5SOY1f693u9/Rj+nr29PYYMGdJpvWIdPz5T080iIiKQlpZmMG/Pnj3XvTduDrKzs+Hu7i52GR0SBAEvvfQSduzYge+//x6DBg264TqmdBy70r/f02q1yM3N7bXH8Pd0Oh00Gk2Hy0zp2F3P9fr4e739+I0bNw65ubnIzs7WT6GhoZgyZQqys7M7POGb0nHsSv9+r7cfw9+7cuUKioqKOq1XtOPXo48hm4H6+nrh2LFjwrFjxwQAwvLly4Vjx44J586dEwRBEOLj44Xnn39e3764uFjo16+f8Le//U04efKksHr1akEmkwmpqalideGGjO3j+++/L6SkpAiFhYVCbm6u8MorrwhSqVTYu3evWF24rhdffFFQqVTCDz/8IJSXl+unxsZGfZvnn39eiI+P1//8008/CRYWFsI//vEP4eTJk8LixYsFS0tLITc3V4wuXFdX+vfGG28Iu3fvFoqKioTMzEzhj3/8o6BQKITjx4+L0YXrio+PF/bt2yecOXNGyMnJEeLj4wWJRCL897//FQTBtI/dr4ztoykdv878fnSQORzH37pR/0ztGP7lL38RfvjhB+HMmTPCTz/9JERGRgpOTk5CZWWlIAi95/gx1NzAr8OXfz9NmzZNEARBmDZtmnDfffe1Wyc4OFiQy+WCj4+PsGHDhttetzGM7eO7774r+Pr6CgqFQnBwcBDuv/9+4fvvvxen+JvQUd8AGByX++67T9/fX33xxRfCkCFDBLlcLowYMUL47rvvbm/hN6kr/Zs3b54wYMAAQS6XC66urkJ0dLSQlZV1+4u/CS+88IIwcOBAQS6XC87OzsK4ceP0J3tBMO1j9ytj+2hKx68zvz/pm8Nx/K0b9c/UjuHkyZMFd3d3QS6XC56ensLkyZOF06dP65f3luMnEQRB6NlrQUREREQ9j8/UEBERkVlgqCEiIiKzwFBDREREZoGhhoiIiMwCQw0RERGZBYYaIiIiMgsMNURERGQWGGqIiIjILDDUEBERkVlgqCEiIiKzwFBDREREZoGhhoiIiMzC/wf4a/ZTeIdA3AAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "score_list = []\n",
    "param_list = []\n",
    "for i in range(1, 6):\n",
    "    nested_X, nested_y, _ = create_nested_dataset(X, y, i)\n",
    "    X_train, X_test, y_train, y_test = train_test_split(nested_X, nested_y)\n",
    "    pipeline = d.best_estimator_\n",
    "    pipeline.fit(X_train, y_train)\n",
    "    y_hat = pipeline.predict(X_test)\n",
    "    score_list.append(f1_score(y_test, y_hat))\n",
    "    param_list.append(i)\n",
    "plt.plot(param_list, score_list)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ca marche bien mais on triche dans le train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def unnest_result(y_hat_nested, nested_idx):\n",
    "    \"\"\"\n",
    "    Unnest un y_hat nested pour pouvoir l'utiliser avec un y_true non nested\n",
    "    \"\"\"\n",
    "    l = []\n",
    "    for i, idx_list in enumerate(nested_idx):\n",
    "        for _ in idx_list:\n",
    "            l.append(y_hat_nested[i])\n",
    "    return l"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 43059/43059 [00:03<00:00, 13430.23it/s]\n",
      "100%|██████████| 14354/14354 [00:01<00:00, 12902.46it/s]\n",
      "100%|██████████| 43059/43059 [00:02<00:00, 17731.70it/s]\n",
      "100%|██████████| 14354/14354 [00:00<00:00, 14762.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 1 :\n",
      "    =====\n",
      "    Macro F1-score : 0.7864642547401168\n",
      "    Micro F1-score : 0.8962658492406298\n",
      "    Macro ROC-AUC: 0.8127118735513237\n",
      "    Weighted ROC-AUC: 0.8127118735513237\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.58      0.70      0.63      1836\n",
      "           1       0.95      0.92      0.94     12518\n",
      "\n",
      "    accuracy                           0.90     14354\n",
      "   macro avg       0.77      0.81      0.79     14354\n",
      "weighted avg       0.91      0.90      0.90     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 25060/25060 [00:02<00:00, 10543.49it/s]\n",
      "100%|██████████| 7177/7177 [00:00<00:00, 7726.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 2 :\n",
      "    =====\n",
      "    Macro F1-score : 0.6694454679439941\n",
      "    Micro F1-score : 0.8489619618224885\n",
      "    Macro ROC-AUC: 0.673579587877117\n",
      "    Weighted ROC-AUC: 0.673579587877117\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.41      0.44      0.43      1836\n",
      "           1       0.92      0.91      0.91     12518\n",
      "\n",
      "    accuracy                           0.85     14354\n",
      "   macro avg       0.67      0.67      0.67     14354\n",
      "weighted avg       0.85      0.85      0.85     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 19245/19245 [00:02<00:00, 8225.60it/s]\n",
      "100%|██████████| 4785/4785 [00:00<00:00, 4989.33it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 3 :\n",
      "    =====\n",
      "    Macro F1-score : 0.6117293464081286\n",
      "    Micro F1-score : 0.8489619618224885\n",
      "    Macro ROC-AUC: 0.5989828242102614\n",
      "    Weighted ROC-AUC: 0.5989828242102614\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.37      0.26      0.31      1836\n",
      "           1       0.90      0.93      0.92     12518\n",
      "\n",
      "    accuracy                           0.85     14354\n",
      "   macro avg       0.63      0.60      0.61     14354\n",
      "weighted avg       0.83      0.85      0.84     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 16377/16377 [00:02<00:00, 6711.76it/s]\n",
      "100%|██████████| 3589/3589 [00:00<00:00, 3791.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 4 :\n",
      "    =====\n",
      "    Macro F1-score : 0.5638298173478391\n",
      "    Micro F1-score : 0.8534206492963634\n",
      "    Macro ROC-AUC: 0.5536670767080153\n",
      "    Weighted ROC-AUC: 0.5536670767080153\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.34      0.15      0.21      1836\n",
      "           1       0.88      0.96      0.92     12518\n",
      "\n",
      "    accuracy                           0.85     14354\n",
      "   macro avg       0.61      0.55      0.56     14354\n",
      "weighted avg       0.81      0.85      0.83     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 14670/14670 [00:02<00:00, 6456.95it/s]\n",
      "100%|██████████| 2871/2871 [00:00<00:00, 3053.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 5 :\n",
      "    =====\n",
      "    Macro F1-score : 0.5244672416273637\n",
      "    Micro F1-score : 0.8619896892852167\n",
      "    Macro ROC-AUC: 0.5267427540507247\n",
      "    Weighted ROC-AUC: 0.5267427540507247\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.33      0.08      0.12      1836\n",
      "           1       0.88      0.98      0.93     12518\n",
      "\n",
      "    accuracy                           0.86     14354\n",
      "   macro avg       0.60      0.53      0.52     14354\n",
      "weighted avg       0.81      0.86      0.82     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 13577/13577 [00:02<00:00, 6024.27it/s]\n",
      "100%|██████████| 2393/2393 [00:00<00:00, 2706.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 6 :\n",
      "    =====\n",
      "    Macro F1-score : 0.5102856831798325\n",
      "    Micro F1-score : 0.8621986902605545\n",
      "    Macro ROC-AUC: 0.5184965893122617\n",
      "    Weighted ROC-AUC: 0.5184965893122617\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.30      0.06      0.10      1836\n",
      "           1       0.88      0.98      0.93     12518\n",
      "\n",
      "    accuracy                           0.86     14354\n",
      "   macro avg       0.59      0.52      0.51     14354\n",
      "weighted avg       0.80      0.86      0.82     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 12804/12804 [00:02<00:00, 5725.66it/s]\n",
      "100%|██████████| 2051/2051 [00:00<00:00, 2158.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 7 :\n",
      "    =====\n",
      "    Macro F1-score : 0.4868368968037807\n",
      "    Micro F1-score : 0.8671450466768845\n",
      "    Macro ROC-AUC: 0.5076215739531154\n",
      "    Weighted ROC-AUC: 0.5076215739531154\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.28      0.02      0.05      1836\n",
      "           1       0.87      0.99      0.93     12518\n",
      "\n",
      "    accuracy                           0.87     14354\n",
      "   macro avg       0.58      0.51      0.49     14354\n",
      "weighted avg       0.80      0.87      0.82     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 12248/12248 [00:02<00:00, 5517.11it/s]\n",
      "100%|██████████| 1795/1795 [00:00<00:00, 1990.35it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 8 :\n",
      "    =====\n",
      "    Macro F1-score : 0.4761414171331845\n",
      "    Micro F1-score : 0.869862059356277\n",
      "    Macro ROC-AUC: 0.5036020026586552\n",
      "    Weighted ROC-AUC: 0.5036020026586552\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.28      0.01      0.02      1836\n",
      "           1       0.87      1.00      0.93     12518\n",
      "\n",
      "    accuracy                           0.87     14354\n",
      "   macro avg       0.58      0.50      0.48     14354\n",
      "weighted avg       0.80      0.87      0.81     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 11788/11788 [00:02<00:00, 5167.95it/s]\n",
      "100%|██████████| 1595/1595 [00:00<00:00, 1761.53it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    n = 9 :\n",
      "    =====\n",
      "    Macro F1-score : 0.48018294396982275\n",
      "    Micro F1-score : 0.869862059356277\n",
      "    Macro ROC-AUC: 0.5054611120335301\n",
      "    Weighted ROC-AUC: 0.5054611120335301\n",
      "    Classification report :\n",
      "                  precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.32      0.02      0.03      1836\n",
      "           1       0.87      1.00      0.93     12518\n",
      "\n",
      "    accuracy                           0.87     14354\n",
      "   macro avg       0.60      0.51      0.48     14354\n",
      "weighted avg       0.80      0.87      0.82     14354\n",
      "\n",
      "    =====\n",
      "    \n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "score_list = []\n",
    "param_list = []\n",
    "\n",
    "# Baseline \n",
    "pipeline = d.best_estimator_\n",
    "pipeline.fit(X_train, y_train)\n",
    "y_hat = pipeline.predict(X_test)\n",
    "score_list.append(f1_score(y_test, y_hat, pos_label=-1))\n",
    "param_list.append(0)\n",
    "\n",
    "for i in range(1, 10):\n",
    "    nested_X_train, nested_y_train, _  = create_nested_dataset(X_train, y_train, i)\n",
    "    nested_X_test, nested_y_test, nested_idx = create_nested_dataset(X_test, [1 for i in range(len(X_test))], i) # sans regarder les y \n",
    "\n",
    "\n",
    "    pipeline = d.best_estimator_\n",
    "    pipeline.fit(nested_X_train, nested_y_train)\n",
    "    y_hat_nested = pipeline.predict(nested_X_test)\n",
    "    \n",
    "    y_hat = unnest_result(y_hat_nested, nested_idx)\n",
    "    print_score(y_test, y_hat, f'n = {i}')\n",
    "    score_list.append(f1_score(y_test, y_hat, pos_label=-1))\n",
    "    param_list.append(i)\n",
    "\n",
    "# plt.plot(param_list, score_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.DataFrame({\n",
    "    \"Taille d'un bloc de phrase\": param_list,\n",
    "    \"f1_score sur Mitterand\": score_list\n",
    "})\n",
    "sns.lineplot(data=df, x=\"Taille d'un bloc de phrase\", y=\"f1_score sur Mitterand\", marker=\"o\") # performance en test des bloc de phrase.\n",
    "plt.savefig('./part1_speaker_recognition/reports/figures/n-sentence_bloc_perf.png')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.10"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}