Project2_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Project M2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "First we import the needed libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:03:37.890400Z",
     "start_time": "2022-11-18T21:03:37.871675Z"
    },
    "pycharm": {
     "is_executing": true,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import datetime\n",
    "\n",
    "sns.set_style(\"darkgrid\")\n",
    "%matplotlib inline\n",
    "\n",
    "#load the statistical libraries\n",
    "from scipy import stats\n",
    "\n",
    "from sklearn.linear_model import LinearRegression, LogisticRegression, Ridge\n",
    "# from sklearn.preprocessing import OneHotEncoder\n",
    "# from pandas.plotting import scatter_matrix\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import mean_squared_error, auc, roc_curve\n",
    "from statsmodels.stats import diagnostic\n",
    "from scipy import stats\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "from ipywidgets import interact\n",
    "from IPython.display import HTML,Javascript, display\n",
    "from sklearn import preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Data preprocessing\n",
    "In this project, we will investigate in to two beer rating datasets, BeerAdvocate and RateBeer. Both datasets contain ```user```, ```brewery```, ```beeer``` and ```rating```."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### Note of rating files\n",
    "Since the **rating** files are txt files of about several gigabytes in size, we cannot handle it simultaneously. Therefore, we **sliced** them to several small pieces and then **combined** them into '.csv' files.\n",
    "\n",
    "The slicing process is extremely time-consuming, thus we won't put the code here. The method is just parsing the file line by line and add some number restrictions for each bulk. And transfer the '.txt' file to '.csv' file.\n",
    "\n",
    "We also filtered out the beer data which has less than 120 ratings. In this way each beer has enough rating for data analysis.\n",
    "\n",
    "We put the processed rating data in the same directory as this ipython notebook and name them as **after_pro_data_ba.csv** and **after_pro_data_rb.csv** for **BeerAdvocate** and **RateBeer** respectively."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### Handle the missing value and initial data filtering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:03:55.686249Z",
     "start_time": "2022-11-18T21:03:37.893704Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Currently, there are 6264473 ratings on BeerAdvocate website\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>style</th>\n",
       "      <th>abv</th>\n",
       "      <th>date</th>\n",
       "      <th>user_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>appearance</th>\n",
       "      <th>aroma</th>\n",
       "      <th>palate</th>\n",
       "      <th>taste</th>\n",
       "      <th>overall</th>\n",
       "      <th>rating</th>\n",
       "      <th>text</th>\n",
       "      <th>review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1500112800</td>\n",
       "      <td>GregKobeArntz24</td>\n",
       "      <td>gregkobearntz24.1206160</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.75</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.50</td>\n",
       "      <td>4.54</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1483182000</td>\n",
       "      <td>awjess</td>\n",
       "      <td>awjess.1185716</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.11</td>\n",
       "      <td>Look - Has a nice foamy head,   Feel - Creamy ...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1468317600</td>\n",
       "      <td>fr0wn3r</td>\n",
       "      <td>fr0wn3r.862994</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.25</td>\n",
       "      <td>3.00</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.00</td>\n",
       "      <td>3.06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1459418400</td>\n",
       "      <td>Fatehunter</td>\n",
       "      <td>fatehunter.326927</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.25</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.50</td>\n",
       "      <td>3.52</td>\n",
       "      <td>Pours a finger of head on a hazy darker straw ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1450263600</td>\n",
       "      <td>MTLDOG</td>\n",
       "      <td>mtldog.920333</td>\n",
       "      <td>4.5</td>\n",
       "      <td>3.75</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.52</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             beer_name  beer_id                brewery_name  brewery_id  \\\n",
       "0  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "1  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "2  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "3  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "4  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "\n",
       "           style  abv        date        user_name                  user_id  \\\n",
       "0  Irish Red Ale  3.8  1500112800  GregKobeArntz24  gregkobearntz24.1206160   \n",
       "1  Irish Red Ale  3.8  1483182000           awjess           awjess.1185716   \n",
       "2  Irish Red Ale  3.8  1468317600          fr0wn3r           fr0wn3r.862994   \n",
       "3  Irish Red Ale  3.8  1459418400       Fatehunter        fatehunter.326927   \n",
       "4  Irish Red Ale  3.8  1450263600           MTLDOG            mtldog.920333   \n",
       "\n",
       "   appearance  aroma  palate  taste  overall  rating  \\\n",
       "0         4.5   4.75    4.25    4.5     4.50    4.54   \n",
       "1         4.5   4.00    4.25    4.0     4.25    4.11   \n",
       "2         3.0   3.25    3.00    3.0     3.00    3.06   \n",
       "3         4.0   3.25    4.00    3.5     3.50    3.52   \n",
       "4         4.5   3.75    4.00    3.0     3.75    3.52   \n",
       "\n",
       "                                                text  review  \n",
       "0                                                NaN   False  \n",
       "1  Look - Has a nice foamy head,   Feel - Creamy ...   False  \n",
       "2                                                NaN   False  \n",
       "3  Pours a finger of head on a hazy darker straw ...    True  \n",
       "4                                                NaN   False  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load ratings data from BeerAdvocate\n",
    "df_rating_ba = pd.read_csv('after_pro_data_ba.csv')\n",
    "df_rating_ba = df_rating_ba.drop(columns=['Unnamed: 0'])\n",
    "print(f'Currently, there are {len(df_rating_ba)} ratings on BeerAdvocate website')\n",
    "df_rating_ba.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:03:56.251689Z",
     "start_time": "2022-11-18T21:03:55.687205Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 2971642 rows have NAN values in BeerAdvocate\n",
      "There are 2971642 rows with 5 NAN values in BeerAdvocate\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count    2.971642e+06\n",
       "mean     3.902421e+00\n",
       "std      6.835105e-01\n",
       "min      1.000000e+00\n",
       "25%      3.500000e+00\n",
       "50%      4.000000e+00\n",
       "75%      4.250000e+00\n",
       "max      5.000000e+00\n",
       "Name: rating, dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The scoring attribute\n",
    "column_ = ['appearance', 'aroma', 'palate', 'taste', 'overall']\n",
    "print(f'There are {df_rating_ba [column_].isna().any(axis = 1).value_counts()[1]} rows have NAN values in BeerAdvocate')\n",
    "print(f'There are {np.sum(df_rating_ba [column_].isna().sum(axis = 1)==5)} rows with 5 NAN values in BeerAdvocate')\n",
    "compare_ba = df_rating_ba[df_rating_ba[column_].isna().any(axis = 1)]\n",
    "compare_ba['rating'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Here we found if there is no overall score, there wouldn't be detailed scores about one beer. We also check the website manually, and found that we should rate the appearance, aroma, palate, taste, overall together.\n",
    "However, these data all have the rating values. We don't think that these data are useful since we want to find some relationship of the detailed score to the rating and other perspectives. And since we will still have very large amout data after deleting them, we will just simply delete them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:03:56.601440Z",
     "start_time": "2022-11-18T21:03:56.252579Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3292831, 17)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Delete all meaninigless lines\n",
    "df_rating_ba = df_rating_ba[~df_rating_ba[column_].isna().any(axis = 1)]\n",
    "df_rating_ba.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "The user_id in BeerAdvocate rating dataset are composed of both of name and id we want to keep the numeric id for simplicity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:03:58.052037Z",
     "start_time": "2022-11-18T21:03:56.604287Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>style</th>\n",
       "      <th>abv</th>\n",
       "      <th>date</th>\n",
       "      <th>user_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>appearance</th>\n",
       "      <th>aroma</th>\n",
       "      <th>palate</th>\n",
       "      <th>taste</th>\n",
       "      <th>overall</th>\n",
       "      <th>rating</th>\n",
       "      <th>text</th>\n",
       "      <th>review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1500112800</td>\n",
       "      <td>GregKobeArntz24</td>\n",
       "      <td>1206160</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.75</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.50</td>\n",
       "      <td>4.54</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1483182000</td>\n",
       "      <td>awjess</td>\n",
       "      <td>1185716</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.25</td>\n",
       "      <td>4.11</td>\n",
       "      <td>Look - Has a nice foamy head,   Feel - Creamy ...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1468317600</td>\n",
       "      <td>fr0wn3r</td>\n",
       "      <td>862994</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.25</td>\n",
       "      <td>3.00</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.00</td>\n",
       "      <td>3.06</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1459418400</td>\n",
       "      <td>Fatehunter</td>\n",
       "      <td>326927</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.25</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.50</td>\n",
       "      <td>3.52</td>\n",
       "      <td>Pours a finger of head on a hazy darker straw ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1450263600</td>\n",
       "      <td>MTLDOG</td>\n",
       "      <td>920333</td>\n",
       "      <td>4.5</td>\n",
       "      <td>3.75</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.52</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             beer_name  beer_id                brewery_name  brewery_id  \\\n",
       "0  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "1  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "2  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "3  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "4  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "\n",
       "           style  abv        date        user_name  user_id  appearance  \\\n",
       "0  Irish Red Ale  3.8  1500112800  GregKobeArntz24  1206160         4.5   \n",
       "1  Irish Red Ale  3.8  1483182000           awjess  1185716         4.5   \n",
       "2  Irish Red Ale  3.8  1468317600          fr0wn3r   862994         3.0   \n",
       "3  Irish Red Ale  3.8  1459418400       Fatehunter   326927         4.0   \n",
       "4  Irish Red Ale  3.8  1450263600           MTLDOG   920333         4.5   \n",
       "\n",
       "   aroma  palate  taste  overall  rating  \\\n",
       "0   4.75    4.25    4.5     4.50    4.54   \n",
       "1   4.00    4.25    4.0     4.25    4.11   \n",
       "2   3.25    3.00    3.0     3.00    3.06   \n",
       "3   3.25    4.00    3.5     3.50    3.52   \n",
       "4   3.75    4.00    3.0     3.75    3.52   \n",
       "\n",
       "                                                text  review  \n",
       "0                                                NaN   False  \n",
       "1  Look - Has a nice foamy head,   Feel - Creamy ...   False  \n",
       "2                                                NaN   False  \n",
       "3  Pours a finger of head on a hazy darker straw ...    True  \n",
       "4                                                NaN   False  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#spilt the user_id in BeerAdvocate\n",
    "def str_split(data):\n",
    "    return data.split('.')[1]\n",
    "df_rating_ba['user_id'] = df_rating_ba['user_id'].apply(str_split)\n",
    "df_rating_ba.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### RateBeer\n",
    "We will do the similar process to the RateBeer dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:11.659191Z",
     "start_time": "2022-11-18T21:03:58.052774Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Currently, there are 3443337 ratings on RateBeer website\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>style</th>\n",
       "      <th>abv</th>\n",
       "      <th>date</th>\n",
       "      <th>user_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>appearance</th>\n",
       "      <th>aroma</th>\n",
       "      <th>palate</th>\n",
       "      <th>taste</th>\n",
       "      <th>overall</th>\n",
       "      <th>rating</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1500199200</td>\n",
       "      <td>mcberko</td>\n",
       "      <td>95615</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>11</td>\n",
       "      <td>2.9</td>\n",
       "      <td>330mL can, Cambodian version (8%), pours black...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1497780000</td>\n",
       "      <td>Grumbo</td>\n",
       "      <td>118968</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>3.2</td>\n",
       "      <td>Can at Johnâs courtesy of Henrik. Sri Lanka ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1497434400</td>\n",
       "      <td>alex_leit</td>\n",
       "      <td>243248</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>13</td>\n",
       "      <td>3.2</td>\n",
       "      <td>Can, Myanmar, 8,0%, dark brown color, coffee f...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1491559200</td>\n",
       "      <td>gripweed57</td>\n",
       "      <td>60778</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>3.9</td>\n",
       "      <td>From old tasting notes. Pours an opaque black ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1486465200</td>\n",
       "      <td>fkoolen</td>\n",
       "      <td>261854</td>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>14</td>\n",
       "      <td>3.6</td>\n",
       "      <td>Big bottle in Nyaungshwe, Myanmar. Really dark...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         beer_name  beer_id  \\\n",
       "0  ABC Extra Stout     9226   \n",
       "1  ABC Extra Stout     9226   \n",
       "2  ABC Extra Stout     9226   \n",
       "3  ABC Extra Stout     9226   \n",
       "4  ABC Extra Stout     9226   \n",
       "\n",
       "                                        brewery_name  brewery_id  \\\n",
       "0  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "1  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "2  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "3  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "4  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "\n",
       "           style  abv        date   user_name  user_id  appearance  aroma  \\\n",
       "0  Foreign Stout  7.0  1500199200     mcberko    95615           4      6   \n",
       "1  Foreign Stout  7.0  1497780000      Grumbo   118968           3      7   \n",
       "2  Foreign Stout  7.0  1497434400   alex_leit   243248           4      6   \n",
       "3  Foreign Stout  7.0  1491559200  gripweed57    60778           4      8   \n",
       "4  Foreign Stout  7.0  1486465200     fkoolen   261854           4      7   \n",
       "\n",
       "   palate  taste  overall  rating  \\\n",
       "0       2      6       11     2.9   \n",
       "1       4      6       12     3.2   \n",
       "2       3      6       13     3.2   \n",
       "3       4      8       15     3.9   \n",
       "4       4      7       14     3.6   \n",
       "\n",
       "                                                text  \n",
       "0  330mL can, Cambodian version (8%), pours black...  \n",
       "1  Can at Johnâs courtesy of Henrik. Sri Lanka ...  \n",
       "2  Can, Myanmar, 8,0%, dark brown color, coffee f...  \n",
       "3  From old tasting notes. Pours an opaque black ...  \n",
       "4  Big bottle in Nyaungshwe, Myanmar. Really dark...  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load rating data on RateBeer website\n",
    "df_rating_rb = pd.read_csv('after_pro_data_rb.csv')\n",
    "df_rating_rb = df_rating_rb.drop(columns=['Unnamed: 0'])\n",
    "print(f'Currently, there are {len(df_rating_rb)} ratings on RateBeer website')\n",
    "df_rating_rb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:11.811417Z",
     "start_time": "2022-11-18T21:04:11.660520Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Thera are 3443337 rows don't have NAN value\n",
      "Thera are 0 rows with NAN value\n",
      "Since we don't have nan values we won't do the missing values handling here.\n"
     ]
    }
   ],
   "source": [
    "print(f\"Thera are {df_rating_rb[column_].isna().any(axis = 1).value_counts()[0]} rows don't have NAN value\")\n",
    "print(f'Thera are {np.sum(df_rating_rb [column_].isna().sum(axis = 1)==5)} rows with NAN value')\n",
    "compare_rb = df_rating_rb[df_rating_rb[column_].isna().any(axis = 1)]\n",
    "print(\"Since we don't have nan values we won't do the missing values handling here.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### Datasets preparation and more filtering\n",
    "#### BeerAdvocate\n",
    "We first map the scale of each rating to 0-5 to make a consistency between the two datasets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:13.034755Z",
     "start_time": "2022-11-18T21:04:11.812444Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>style</th>\n",
       "      <th>abv</th>\n",
       "      <th>date</th>\n",
       "      <th>user_name</th>\n",
       "      <th>user_id</th>\n",
       "      <th>appearance</th>\n",
       "      <th>aroma</th>\n",
       "      <th>palate</th>\n",
       "      <th>taste</th>\n",
       "      <th>overall</th>\n",
       "      <th>rating</th>\n",
       "      <th>text</th>\n",
       "      <th>review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1500112800</td>\n",
       "      <td>GregKobeArntz24</td>\n",
       "      <td>1206160</td>\n",
       "      <td>4.375</td>\n",
       "      <td>4.6875</td>\n",
       "      <td>4.0625</td>\n",
       "      <td>4.375</td>\n",
       "      <td>4.3750</td>\n",
       "      <td>4.4250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1483182000</td>\n",
       "      <td>awjess</td>\n",
       "      <td>1185716</td>\n",
       "      <td>4.375</td>\n",
       "      <td>3.7500</td>\n",
       "      <td>4.0625</td>\n",
       "      <td>3.750</td>\n",
       "      <td>4.0625</td>\n",
       "      <td>3.8875</td>\n",
       "      <td>Look - Has a nice foamy head,   Feel - Creamy ...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1468317600</td>\n",
       "      <td>fr0wn3r</td>\n",
       "      <td>862994</td>\n",
       "      <td>2.500</td>\n",
       "      <td>2.8125</td>\n",
       "      <td>2.5000</td>\n",
       "      <td>2.500</td>\n",
       "      <td>2.5000</td>\n",
       "      <td>2.5750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1459418400</td>\n",
       "      <td>Fatehunter</td>\n",
       "      <td>326927</td>\n",
       "      <td>3.750</td>\n",
       "      <td>2.8125</td>\n",
       "      <td>3.7500</td>\n",
       "      <td>3.125</td>\n",
       "      <td>3.1250</td>\n",
       "      <td>3.1500</td>\n",
       "      <td>Pours a finger of head on a hazy darker straw ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Caffrey's Irish Ale</td>\n",
       "      <td>825</td>\n",
       "      <td>Thomas Caffrey Brewing Co.</td>\n",
       "      <td>297</td>\n",
       "      <td>Irish Red Ale</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1450263600</td>\n",
       "      <td>MTLDOG</td>\n",
       "      <td>920333</td>\n",
       "      <td>4.375</td>\n",
       "      <td>3.4375</td>\n",
       "      <td>3.7500</td>\n",
       "      <td>2.500</td>\n",
       "      <td>3.4375</td>\n",
       "      <td>3.1500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             beer_name  beer_id                brewery_name  brewery_id  \\\n",
       "0  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "1  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "2  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "3  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "4  Caffrey's Irish Ale      825  Thomas Caffrey Brewing Co.         297   \n",
       "\n",
       "           style  abv        date        user_name  user_id  appearance  \\\n",
       "0  Irish Red Ale  3.8  1500112800  GregKobeArntz24  1206160       4.375   \n",
       "1  Irish Red Ale  3.8  1483182000           awjess  1185716       4.375   \n",
       "2  Irish Red Ale  3.8  1468317600          fr0wn3r   862994       2.500   \n",
       "3  Irish Red Ale  3.8  1459418400       Fatehunter   326927       3.750   \n",
       "4  Irish Red Ale  3.8  1450263600           MTLDOG   920333       4.375   \n",
       "\n",
       "    aroma  palate  taste  overall  rating  \\\n",
       "0  4.6875  4.0625  4.375   4.3750  4.4250   \n",
       "1  3.7500  4.0625  3.750   4.0625  3.8875   \n",
       "2  2.8125  2.5000  2.500   2.5000  2.5750   \n",
       "3  2.8125  3.7500  3.125   3.1250  3.1500   \n",
       "4  3.4375  3.7500  2.500   3.4375  3.1500   \n",
       "\n",
       "                                                text  review  \n",
       "0                                                NaN   False  \n",
       "1  Look - Has a nice foamy head,   Feel - Creamy ...   False  \n",
       "2                                                NaN   False  \n",
       "3  Pours a finger of head on a hazy darker straw ...    True  \n",
       "4                                                NaN   False  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_list = ['appearance', 'aroma', 'palate', 'taste', 'overall','rating']\n",
    "list_value = df_rating_ba[column_list].values\n",
    "min_max_scaler = preprocessing.MinMaxScaler()\n",
    "list_value_scaler = min_max_scaler.fit_transform(list_value) * 5\n",
    "data_temp_ba = pd.DataFrame(list_value_scaler,columns=column_list,index = df_rating_ba.index)\n",
    "df_rating_ba[column_list] = data_temp_ba\n",
    "\n",
    "df_rating_ba.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "By doing some simple regression study we find there exist perfect linear relationship between 'overall', 'aroma', 'palate', 'taste', 'overall', 'rating'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:14.444050Z",
     "start_time": "2022-11-18T21:04:13.036157Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                 rating   R-squared:                       1.000\n",
      "Model:                            OLS   Adj. R-squared:                  1.000\n",
      "Method:                 Least Squares   F-statistic:                 4.904e+10\n",
      "Date:                Fri, 18 Nov 2022   Prob (F-statistic):               0.00\n",
      "Time:                        22:04:14   Log-Likelihood:             1.4883e+07\n",
      "No. Observations:             3292831   AIC:                        -2.977e+07\n",
      "Df Residuals:                 3292825   BIC:                        -2.977e+07\n",
      "Df Model:                           5                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.0003    8.3e-06     37.210      0.000       0.000       0.000\n",
      "appearance     0.0601   2.95e-06   2.04e+04      0.000       0.060       0.060\n",
      "aroma          0.2401   3.12e-06   7.69e+04      0.000       0.240       0.240\n",
      "palate         0.1001   3.46e-06    2.9e+04      0.000       0.100       0.100\n",
      "taste          0.4001   4.04e-06   9.89e+04      0.000       0.400       0.400\n",
      "overall        0.1999   4.09e-06   4.88e+04      0.000       0.200       0.200\n",
      "==============================================================================\n",
      "Omnibus:                   572478.286   Durbin-Watson:                   1.613\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           873283.288\n",
      "Skew:                           1.244   Prob(JB):                         0.00\n",
      "Kurtosis:                       2.584   Cond. No.                         48.7\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "The R-squared is 1.000 so it is of perfect linear relationship.\n"
     ]
    }
   ],
   "source": [
    "# Use linear regression to fit our data\n",
    "mod_ba = smf.ols(formula='rating ~appearance + aroma + palate + taste + overall',data = df_rating_ba)\n",
    "res_ba = mod_ba.fit()\n",
    "print(res_ba.summary())\n",
    "print(f'The R-squared is 1.000 so it is of perfect linear relationship.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We also find that there exists some ratings that the **overall** score exceeds the average of the other 4 features(can do this due to the perfect linear relationship). We consider this situation as invalid ratings. Because in common sense, if the detailed features of a beer is good/bad the overall will not be bad/good. And we set the difference threshold to 1.5(in scale of 5)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:15.788060Z",
     "start_time": "2022-11-18T21:04:14.445419Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3266748, 17)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rating_ba['average'] = df_rating_ba[['appearance','aroma','palate','taste']].mean(axis = 1)\n",
    "df_rating_ba['diff'] = np.abs(df_rating_ba['overall']-df_rating_ba['average'])\n",
    "df_rating_ba = df_rating_ba[df_rating_ba['diff'] <1.5]\n",
    "df_rating_ba = df_rating_ba.drop(columns=['average','diff'])\n",
    "df_rating_ba.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:15.938943Z",
     "start_time": "2022-11-18T21:04:15.789262Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nbr_ratings</th>\n",
       "      <th>nbr_reviews</th>\n",
       "      <th>user_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>joined</th>\n",
       "      <th>location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7820</td>\n",
       "      <td>465</td>\n",
       "      <td>184925</td>\n",
       "      <td>nmann08</td>\n",
       "      <td>1.199704e+09</td>\n",
       "      <td>United States, Washington</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2521</td>\n",
       "      <td>2504</td>\n",
       "      <td>163714</td>\n",
       "      <td>StJamesGate</td>\n",
       "      <td>1.191838e+09</td>\n",
       "      <td>United States, New York</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1797</td>\n",
       "      <td>1143</td>\n",
       "      <td>19527</td>\n",
       "      <td>mdagnew</td>\n",
       "      <td>1.116410e+09</td>\n",
       "      <td>Northern Ireland</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "      <td>10867</td>\n",
       "      <td>helloloser12345</td>\n",
       "      <td>1.101380e+09</td>\n",
       "      <td>Northern Ireland</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>604</td>\n",
       "      <td>604</td>\n",
       "      <td>3708</td>\n",
       "      <td>cypressbob</td>\n",
       "      <td>1.069326e+09</td>\n",
       "      <td>Northern Ireland</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nbr_ratings  nbr_reviews user_id        user_name        joined  \\\n",
       "0         7820          465  184925          nmann08  1.199704e+09   \n",
       "1         2521         2504  163714      StJamesGate  1.191838e+09   \n",
       "2         1797         1143   19527          mdagnew  1.116410e+09   \n",
       "3           31           31   10867  helloloser12345  1.101380e+09   \n",
       "4          604          604    3708       cypressbob  1.069326e+09   \n",
       "\n",
       "                    location  \n",
       "0  United States, Washington  \n",
       "1    United States, New York  \n",
       "2           Northern Ireland  \n",
       "3           Northern Ireland  \n",
       "4           Northern Ireland  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load user data from BeerAdvocate\n",
    "df_user_ba = pd.read_csv('BeerAdvocate/users.csv')\n",
    "df_user_ba['user_id'] = df_user_ba['user_id'].apply(str_split)\n",
    "df_user_ba.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:16.245092Z",
     "start_time": "2022-11-18T21:04:15.940142Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_id</th>\n",
       "      <th>beer_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>style</th>\n",
       "      <th>nbr_ratings</th>\n",
       "      <th>nbr_reviews</th>\n",
       "      <th>avg</th>\n",
       "      <th>ba_score</th>\n",
       "      <th>bros_score</th>\n",
       "      <th>abv</th>\n",
       "      <th>avg_computed</th>\n",
       "      <th>zscore</th>\n",
       "      <th>nbr_matched_valid_ratings</th>\n",
       "      <th>avg_matched_valid_ratings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>166064</td>\n",
       "      <td>Nashe Moskovskoe</td>\n",
       "      <td>39912</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>Euro Pale Lager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>166065</td>\n",
       "      <td>Nashe Pivovskoe</td>\n",
       "      <td>39912</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>Euro Pale Lager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>166066</td>\n",
       "      <td>Nashe Shakhterskoe</td>\n",
       "      <td>39912</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>Euro Pale Lager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>166067</td>\n",
       "      <td>Nashe Zhigulevskoe</td>\n",
       "      <td>39912</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>Euro Pale Lager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>166063</td>\n",
       "      <td>Zhivoe</td>\n",
       "      <td>39912</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>Euro Pale Lager</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   beer_id           beer_name  brewery_id            brewery_name  \\\n",
       "0   166064    Nashe Moskovskoe       39912  Abdysh-Ata (Абдыш Ата)   \n",
       "1   166065     Nashe Pivovskoe       39912  Abdysh-Ata (Абдыш Ата)   \n",
       "2   166066  Nashe Shakhterskoe       39912  Abdysh-Ata (Абдыш Ата)   \n",
       "3   166067  Nashe Zhigulevskoe       39912  Abdysh-Ata (Абдыш Ата)   \n",
       "4   166063              Zhivoe       39912  Abdysh-Ata (Абдыш Ата)   \n",
       "\n",
       "             style  nbr_ratings  nbr_reviews  avg  ba_score  bros_score  abv  \\\n",
       "0  Euro Pale Lager            0            0  NaN       NaN         NaN  4.7   \n",
       "1  Euro Pale Lager            0            0  NaN       NaN         NaN  3.8   \n",
       "2  Euro Pale Lager            0            0  NaN       NaN         NaN  4.8   \n",
       "3  Euro Pale Lager            0            0  NaN       NaN         NaN  4.0   \n",
       "4  Euro Pale Lager            0            0  NaN       NaN         NaN  4.5   \n",
       "\n",
       "   avg_computed  zscore  nbr_matched_valid_ratings  avg_matched_valid_ratings  \n",
       "0           NaN     NaN                          0                        NaN  \n",
       "1           NaN     NaN                          0                        NaN  \n",
       "2           NaN     NaN                          0                        NaN  \n",
       "3           NaN     NaN                          0                        NaN  \n",
       "4           NaN     NaN                          0                        NaN  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load beer data from BeerAdvocate\n",
    "df_beer_ba = pd.read_csv('BeerAdvocate/beers.csv')\n",
    "df_beer_ba.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:16.262288Z",
     "start_time": "2022-11-18T21:04:16.246057Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>location</th>\n",
       "      <th>name</th>\n",
       "      <th>nbr_beers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39912</td>\n",
       "      <td>Kyrgyzstan</td>\n",
       "      <td>Abdysh-Ata (Абдыш Ата)</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>39913</td>\n",
       "      <td>Kyrgyzstan</td>\n",
       "      <td>Arpa (АРПА)</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>39914</td>\n",
       "      <td>Kyrgyzstan</td>\n",
       "      <td>Bear Beer</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>39915</td>\n",
       "      <td>Kyrgyzstan</td>\n",
       "      <td>Blonder Pub</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>39916</td>\n",
       "      <td>Kyrgyzstan</td>\n",
       "      <td>Kellers Bier</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id    location                    name  nbr_beers\n",
       "0  39912  Kyrgyzstan  Abdysh-Ata (Абдыш Ата)          5\n",
       "1  39913  Kyrgyzstan             Arpa (АРПА)          1\n",
       "2  39914  Kyrgyzstan               Bear Beer          4\n",
       "3  39915  Kyrgyzstan             Blonder Pub          4\n",
       "4  39916  Kyrgyzstan            Kellers Bier          2"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load brewery data from BeerAdvocate\n",
    "df_brewery_ba = pd.read_csv('BeerAdvocate/breweries.csv')\n",
    "df_brewery_ba.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:17.157893Z",
     "start_time": "2022-11-18T21:04:16.265479Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# Merge data based on user id in order to get users' location\n",
    "#user_region = df_user_ba[['user_id', 'location']]\n",
    "region_ba = pd.merge(df_rating_ba, df_user_ba, how=\"left\", left_on='user_id', right_on='user_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:17.930433Z",
     "start_time": "2022-11-18T21:04:17.158727Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>location</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>United States, Pennsylvania</td>\n",
       "      <td>260952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>United States, California</td>\n",
       "      <td>235136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>United States, New York</td>\n",
       "      <td>209952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>United States, Illinois</td>\n",
       "      <td>201059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>United States, Massachusetts</td>\n",
       "      <td>178341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>United States, Ohio</td>\n",
       "      <td>136446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>United States, Texas</td>\n",
       "      <td>132515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>United States, New Jersey</td>\n",
       "      <td>109257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>United States, Florida</td>\n",
       "      <td>95612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>United States, Virginia</td>\n",
       "      <td>95248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>United States, Michigan</td>\n",
       "      <td>94560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>United States, Minnesota</td>\n",
       "      <td>93453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Canada</td>\n",
       "      <td>89853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>United States, North Carolina</td>\n",
       "      <td>84890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>United States, Washington</td>\n",
       "      <td>80398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>United States, Connecticut</td>\n",
       "      <td>72007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>United States, Maryland</td>\n",
       "      <td>70820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>United States, Colorado</td>\n",
       "      <td>69393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>United States, Wisconsin</td>\n",
       "      <td>66584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>United States, Georgia</td>\n",
       "      <td>61790</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         location   count\n",
       "0     United States, Pennsylvania  260952\n",
       "1       United States, California  235136\n",
       "2         United States, New York  209952\n",
       "3         United States, Illinois  201059\n",
       "4    United States, Massachusetts  178341\n",
       "5             United States, Ohio  136446\n",
       "6            United States, Texas  132515\n",
       "7       United States, New Jersey  109257\n",
       "8          United States, Florida   95612\n",
       "9         United States, Virginia   95248\n",
       "10        United States, Michigan   94560\n",
       "11       United States, Minnesota   93453\n",
       "12                         Canada   89853\n",
       "13  United States, North Carolina   84890\n",
       "14      United States, Washington   80398\n",
       "15     United States, Connecticut   72007\n",
       "16        United States, Maryland   70820\n",
       "17        United States, Colorado   69393\n",
       "18       United States, Wisconsin   66584\n",
       "19         United States, Georgia   61790"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_agg = region_ba[['location', 'beer_id']].groupby('location').agg('count')\n",
    "region_agg = pd.DataFrame(region_agg.to_records())\n",
    "region_agg.rename(columns={'beer_id':'count'}, inplace=True)\n",
    "region_agg = region_agg.sort_values(by=['count'], ascending=False)\n",
    "region_agg.reset_index(inplace=True, drop=True)\n",
    "# Print top 20 region\n",
    "region_agg.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### RateBeer\n",
    "First map to scale to 0-5 then filtered the invalid rating as previous steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:18.481486Z",
     "start_time": "2022-11-18T21:04:17.931464Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "list_value_rb = df_rating_rb[column_list].values\n",
    "list_value_scaler_rb = min_max_scaler.fit_transform(list_value_rb) * 5\n",
    "data_temp_rb = pd.DataFrame(list_value_scaler_rb,columns=column_list,index = df_rating_rb.index)\n",
    "df_rating_rb[column_list] = data_temp_rb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:19.726371Z",
     "start_time": "2022-11-18T21:04:18.482268Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                 rating   R-squared:                       1.000\n",
      "Model:                            OLS   Adj. R-squared:                  1.000\n",
      "Method:                 Least Squares   F-statistic:                 1.529e+11\n",
      "Date:                Fri, 18 Nov 2022   Prob (F-statistic):               0.00\n",
      "Time:                        22:04:19   Log-Likelihood:             1.7073e+07\n",
      "No. Observations:             3443337   AIC:                        -3.415e+07\n",
      "Df Residuals:                 3443331   BIC:                        -3.415e+07\n",
      "Df Model:                           5                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept   5.049e-06   3.86e-06      1.309      0.190   -2.51e-06    1.26e-05\n",
      "appearance     0.0889   1.14e-06   7.81e+04      0.000       0.089       0.089\n",
      "aroma          0.2000   1.75e-06   1.14e+05      0.000       0.200       0.200\n",
      "palate         0.0889   1.29e-06   6.88e+04      0.000       0.089       0.089\n",
      "taste          0.2000   2.24e-06   8.92e+04      0.000       0.200       0.200\n",
      "overall        0.4222   2.25e-06   1.88e+05      0.000       0.422       0.422\n",
      "==================================================================================\n",
      "Omnibus:                 22573960.762   Durbin-Watson:                       2.000\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):   151634840746661408.000\n",
      "Skew:                         597.821   Prob(JB):                             0.00\n",
      "Kurtosis:                 1028054.953   Cond. No.                             31.4\n",
      "==================================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "mod_rb = smf.ols(formula='rating ~appearance + aroma + palate + taste + overall',data =df_rating_rb)\n",
    "res_rb = mod_rb.fit()\n",
    "print(res_rb.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:21.015154Z",
     "start_time": "2022-11-18T21:04:19.727389Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3412942, 16)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rating_rb['average'] = df_rating_rb[['appearance','aroma','palate','taste']].mean(axis = 1)\n",
    "df_rating_rb['diff'] = np.abs(df_rating_rb['overall']-df_rating_rb['average'])\n",
    "df_rating_rb = df_rating_rb[df_rating_rb['diff'] <1.5]\n",
    "df_rating_rb = df_rating_rb.drop(columns=['average','diff'])\n",
    "df_rating_rb.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:21.040767Z",
     "start_time": "2022-11-18T21:04:21.016792Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>location</th>\n",
       "      <th>name</th>\n",
       "      <th>nbr_beers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3198</td>\n",
       "      <td>Gabon</td>\n",
       "      <td>Sobraga</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13538</td>\n",
       "      <td>Northern Ireland</td>\n",
       "      <td>Ards Brewing Co.</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22304</td>\n",
       "      <td>Northern Ireland</td>\n",
       "      <td>Barrahooley Craft Brewery</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>22818</td>\n",
       "      <td>Northern Ireland</td>\n",
       "      <td>Boundary</td>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24297</td>\n",
       "      <td>Northern Ireland</td>\n",
       "      <td>Brewbot Belfast</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id          location                       name  nbr_beers\n",
       "0   3198             Gabon                    Sobraga          3\n",
       "1  13538  Northern Ireland           Ards Brewing Co.         13\n",
       "2  22304  Northern Ireland  Barrahooley Craft Brewery          4\n",
       "3  22818  Northern Ireland                   Boundary         66\n",
       "4  24297  Northern Ireland            Brewbot Belfast          3"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load brewery data on RateBeer website\n",
    "df_brewery_rb = pd.read_csv('RateBeer/breweries.csv')\n",
    "df_brewery_rb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:21.527223Z",
     "start_time": "2022-11-18T21:04:21.041943Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_id</th>\n",
       "      <th>beer_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>style</th>\n",
       "      <th>nbr_ratings</th>\n",
       "      <th>overall_score</th>\n",
       "      <th>style_score</th>\n",
       "      <th>avg</th>\n",
       "      <th>abv</th>\n",
       "      <th>avg_computed</th>\n",
       "      <th>zscore</th>\n",
       "      <th>nbr_matched_valid_ratings</th>\n",
       "      <th>avg_matched_valid_ratings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>410549</td>\n",
       "      <td>33 Export (Gabon)</td>\n",
       "      <td>3198</td>\n",
       "      <td>Sobraga</td>\n",
       "      <td>Pale Lager</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.72</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>105273</td>\n",
       "      <td>Castel Beer (Gabon)</td>\n",
       "      <td>3198</td>\n",
       "      <td>Sobraga</td>\n",
       "      <td>Pale Lager</td>\n",
       "      <td>10</td>\n",
       "      <td>5.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2.18</td>\n",
       "      <td>5.2</td>\n",
       "      <td>1.910000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>19445</td>\n",
       "      <td>Régab</td>\n",
       "      <td>3198</td>\n",
       "      <td>Sobraga</td>\n",
       "      <td>Pale Lager</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1.83</td>\n",
       "      <td>4.5</td>\n",
       "      <td>1.811111</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>155699</td>\n",
       "      <td>Ards Bally Black Stout</td>\n",
       "      <td>13538</td>\n",
       "      <td>Ards Brewing Co.</td>\n",
       "      <td>Stout</td>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.18</td>\n",
       "      <td>4.6</td>\n",
       "      <td>3.866667</td>\n",
       "      <td>0.823699</td>\n",
       "      <td>6</td>\n",
       "      <td>3.866667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>239097</td>\n",
       "      <td>Ards Belfast 366</td>\n",
       "      <td>13538</td>\n",
       "      <td>Ards Brewing Co.</td>\n",
       "      <td>Golden Ale/Blond Ale</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.79</td>\n",
       "      <td>4.2</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   beer_id               beer_name  brewery_id      brewery_name  \\\n",
       "0   410549       33 Export (Gabon)        3198           Sobraga   \n",
       "1   105273     Castel Beer (Gabon)        3198           Sobraga   \n",
       "2    19445                   Régab        3198           Sobraga   \n",
       "3   155699  Ards Bally Black Stout       13538  Ards Brewing Co.   \n",
       "4   239097        Ards Belfast 366       13538  Ards Brewing Co.   \n",
       "\n",
       "                  style  nbr_ratings  overall_score  style_score   avg  abv  \\\n",
       "0            Pale Lager            1            NaN          NaN  2.72  5.0   \n",
       "1            Pale Lager           10            5.0         58.0  2.18  5.2   \n",
       "2            Pale Lager           27            1.0         22.0  1.83  4.5   \n",
       "3                 Stout            6            NaN          NaN  3.18  4.6   \n",
       "4  Golden Ale/Blond Ale            1            NaN          NaN  2.79  4.2   \n",
       "\n",
       "   avg_computed    zscore  nbr_matched_valid_ratings  \\\n",
       "0      2.000000       NaN                          0   \n",
       "1      1.910000       NaN                          0   \n",
       "2      1.811111       NaN                          0   \n",
       "3      3.866667  0.823699                          6   \n",
       "4      3.000000       NaN                          0   \n",
       "\n",
       "   avg_matched_valid_ratings  \n",
       "0                        NaN  \n",
       "1                        NaN  \n",
       "2                        NaN  \n",
       "3                   3.866667  \n",
       "4                        NaN  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load beer date on RateBeer website\n",
    "df_beer_rb = pd.read_csv('RateBeer/beers.csv')\n",
    "df_beer_rb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:21.569802Z",
     "start_time": "2022-11-18T21:04:21.528069Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nbr_ratings</th>\n",
       "      <th>user_id</th>\n",
       "      <th>user_name</th>\n",
       "      <th>joined</th>\n",
       "      <th>location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1890</td>\n",
       "      <td>175852</td>\n",
       "      <td>Manslow</td>\n",
       "      <td>1.337508e+09</td>\n",
       "      <td>Poland</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>89</td>\n",
       "      <td>442761</td>\n",
       "      <td>MAGICuenca91</td>\n",
       "      <td>1.484046e+09</td>\n",
       "      <td>Spain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>169</td>\n",
       "      <td>288889</td>\n",
       "      <td>Sibarh</td>\n",
       "      <td>1.384600e+09</td>\n",
       "      <td>Poland</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3371</td>\n",
       "      <td>250510</td>\n",
       "      <td>fombe89</td>\n",
       "      <td>1.363950e+09</td>\n",
       "      <td>Spain</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13043</td>\n",
       "      <td>122778</td>\n",
       "      <td>kevnic2008</td>\n",
       "      <td>1.296644e+09</td>\n",
       "      <td>Germany</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nbr_ratings  user_id     user_name        joined location\n",
       "0         1890   175852       Manslow  1.337508e+09   Poland\n",
       "1           89   442761  MAGICuenca91  1.484046e+09    Spain\n",
       "2          169   288889        Sibarh  1.384600e+09   Poland\n",
       "3         3371   250510       fombe89  1.363950e+09    Spain\n",
       "4        13043   122778    kevnic2008  1.296644e+09  Germany"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load user date on RateBeer website\n",
    "df_user_rb = pd.read_csv('RateBeer/users.csv')\n",
    "df_user_rb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:22.033892Z",
     "start_time": "2022-11-18T21:04:21.571011Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewery_name</th>\n",
       "      <th>brewery_id</th>\n",
       "      <th>style</th>\n",
       "      <th>abv</th>\n",
       "      <th>date</th>\n",
       "      <th>user_name_x</th>\n",
       "      <th>user_id</th>\n",
       "      <th>appearance</th>\n",
       "      <th>aroma</th>\n",
       "      <th>palate</th>\n",
       "      <th>taste</th>\n",
       "      <th>overall</th>\n",
       "      <th>rating</th>\n",
       "      <th>text</th>\n",
       "      <th>nbr_ratings</th>\n",
       "      <th>user_name_y</th>\n",
       "      <th>joined</th>\n",
       "      <th>location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1500199200</td>\n",
       "      <td>mcberko</td>\n",
       "      <td>95615</td>\n",
       "      <td>3.75</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>1.25</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>2.631579</td>\n",
       "      <td>2.666667</td>\n",
       "      <td>330mL can, Cambodian version (8%), pours black...</td>\n",
       "      <td>15893.0</td>\n",
       "      <td>mcberko</td>\n",
       "      <td>1.253441e+09</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1497780000</td>\n",
       "      <td>Grumbo</td>\n",
       "      <td>118968</td>\n",
       "      <td>2.50</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>3.75</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>2.894737</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>Can at Johnâs courtesy of Henrik. Sri Lanka ...</td>\n",
       "      <td>3050.0</td>\n",
       "      <td>Grumbo</td>\n",
       "      <td>1.292238e+09</td>\n",
       "      <td>England</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1497434400</td>\n",
       "      <td>alex_leit</td>\n",
       "      <td>243248</td>\n",
       "      <td>3.75</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>2.50</td>\n",
       "      <td>2.777778</td>\n",
       "      <td>3.157895</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>Can, Myanmar, 8,0%, dark brown color, coffee f...</td>\n",
       "      <td>4300.0</td>\n",
       "      <td>alex_leit</td>\n",
       "      <td>1.360148e+09</td>\n",
       "      <td>Belarus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1491559200</td>\n",
       "      <td>gripweed57</td>\n",
       "      <td>60778</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.888889</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.888889</td>\n",
       "      <td>3.684211</td>\n",
       "      <td>3.777778</td>\n",
       "      <td>From old tasting notes. Pours an opaque black ...</td>\n",
       "      <td>2252.0</td>\n",
       "      <td>gripweed57</td>\n",
       "      <td>1.189678e+09</td>\n",
       "      <td>United States, New Jersey</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ABC Extra Stout</td>\n",
       "      <td>9226</td>\n",
       "      <td>Singapore Brewery (Asia Pacific Breweries-Hein...</td>\n",
       "      <td>547</td>\n",
       "      <td>Foreign Stout</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1486465200</td>\n",
       "      <td>fkoolen</td>\n",
       "      <td>261854</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>3.75</td>\n",
       "      <td>3.333333</td>\n",
       "      <td>3.421053</td>\n",
       "      <td>3.444444</td>\n",
       "      <td>Big bottle in Nyaungshwe, Myanmar. Really dark...</td>\n",
       "      <td>1200.0</td>\n",
       "      <td>fkoolen</td>\n",
       "      <td>1.369908e+09</td>\n",
       "      <td>Netherlands</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         beer_name  beer_id  \\\n",
       "0  ABC Extra Stout     9226   \n",
       "1  ABC Extra Stout     9226   \n",
       "2  ABC Extra Stout     9226   \n",
       "3  ABC Extra Stout     9226   \n",
       "4  ABC Extra Stout     9226   \n",
       "\n",
       "                                        brewery_name  brewery_id  \\\n",
       "0  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "1  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "2  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "3  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "4  Singapore Brewery (Asia Pacific Breweries-Hein...         547   \n",
       "\n",
       "           style  abv        date user_name_x  user_id  appearance     aroma  \\\n",
       "0  Foreign Stout  7.0  1500199200     mcberko    95615        3.75  2.777778   \n",
       "1  Foreign Stout  7.0  1497780000      Grumbo   118968        2.50  3.333333   \n",
       "2  Foreign Stout  7.0  1497434400   alex_leit   243248        3.75  2.777778   \n",
       "3  Foreign Stout  7.0  1491559200  gripweed57    60778        3.75  3.888889   \n",
       "4  Foreign Stout  7.0  1486465200     fkoolen   261854        3.75  3.333333   \n",
       "\n",
       "   palate     taste   overall    rating  \\\n",
       "0    1.25  2.777778  2.631579  2.666667   \n",
       "1    3.75  2.777778  2.894737  3.000000   \n",
       "2    2.50  2.777778  3.157895  3.000000   \n",
       "3    3.75  3.888889  3.684211  3.777778   \n",
       "4    3.75  3.333333  3.421053  3.444444   \n",
       "\n",
       "                                                text  nbr_ratings user_name_y  \\\n",
       "0  330mL can, Cambodian version (8%), pours black...      15893.0     mcberko   \n",
       "1  Can at Johnâs courtesy of Henrik. Sri Lanka ...       3050.0      Grumbo   \n",
       "2  Can, Myanmar, 8,0%, dark brown color, coffee f...       4300.0   alex_leit   \n",
       "3  From old tasting notes. Pours an opaque black ...       2252.0  gripweed57   \n",
       "4  Big bottle in Nyaungshwe, Myanmar. Really dark...       1200.0     fkoolen   \n",
       "\n",
       "         joined                   location  \n",
       "0  1.253441e+09                     Canada  \n",
       "1  1.292238e+09                    England  \n",
       "2  1.360148e+09                    Belarus  \n",
       "3  1.189678e+09  United States, New Jersey  \n",
       "4  1.369908e+09                Netherlands  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge data based on user id in order to get users' location\n",
    "region_rb = pd.merge(df_rating_rb, df_user_rb, how=\"left\", left_on='user_id', right_on='user_id')\n",
    "region_rb.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Data Analysis and visualization\n",
    "In this section we will explore more of the datasets we prepared above, and visualize them."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Overall Rating Distribution of the two website"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:22.827786Z",
     "start_time": "2022-11-18T21:04:22.034690Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#creating subplots\n",
    "ax=plt.subplots()\n",
    "\n",
    "#plotting columns\n",
    "ax=sns.histplot(data=df_rating_ba, x='rating', color = 'purple', binwidth=0.25, label='x')\n",
    "ax=sns.histplot(data=df_rating_rb, x='rating', color = 'pink', binwidth=0.25)\n",
    "\n",
    "# setting figure legend\n",
    "plt.legend(title='Website', loc='upper left', labels=['BeerAdvocate', 'RateBeer'])\n",
    "\n",
    "#renaming the axes\n",
    "ax.set(xlabel=\"Rating\", ylabel=\"Count\")\n",
    "\n",
    "# setting figute title\n",
    "ax.set_title(\"Rating Distribution on two website\")\n",
    "\n",
    "# visualizing illustration\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We can see that in this plot, rating on both websites are distributed like a normal distribution. There concentration of **rating** between 3-5 contributes to the majority."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Rating Distribution of Brewery\n",
    "##### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:24.754679Z",
     "start_time": "2022-11-18T21:04:22.828671Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAESCAYAAADwnNLKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAfH0lEQVR4nO3de5gcVbnv8W9PT4YYGDIgAcQNRM6GFxTlEhUEQgIC2eEgePAoRC6CIBdhE0FEuSkHUOCAQYIX3MmBcJGwD4GooAgSuSubBwyXcHkR5OIRhRCYEAjJMD19/lhroJn0LTNd1T1Tv8/z5El3ra5Vb6+qqbfWquqqXLFYREREsqut2QGIiEhzKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhkXHuzA8gaMysCi4ACUATGAG8Ax7r7gzXmPRLocPefmtkxQJe7n9+AmMYDzwKPxUl5YDlwkrvfN9T6k2Zmc4A9gcVxUgdwN3CWu//TzDYC5rn7TlXq+Ahwkbt/oUzZu/Ob2VnAeu5+/GrGOAu4zN0fMrPZwHXufvvq1JEUM7sT2BRYCuQI7TfX3c9OYFnrAX8D5rj7sVU+dzKwtbsf1ugYKiyv4vrPAiWC5tjN3V/tfxM3+kuBz9SYbxdCEsHdL2twTG+7+7YlMX0JmANs3uDlJOVid78IwMxywKnA78xsgru/BFRMAtGmgJUrqHP+WvYEfh7rO3KIdSXhW+4+D8DMuoAnzGxBAgcCRwC/Ar5sZqe7+2sNrn+wKq7/LFAiaDIzawc2AV6L7zcg7DA2ADYEXgC+BOwM7AvsaWZvA+OIR6Zm9jxhp/3ZWNdV7n5mrO87hD++ZYSj5M+7+/g6Qvsg8I9Yx2TgEuAtYC3gU8BewBmEo8flwMnA0zHeDdz9LTP7ObClu0+K9fwlfoc3gB/HWEcRjo5/EHsm9wBPAuOBK4GPuvtBcf5dgEvdfbtqgbt7EfiBmR0W2+spYJG7r2VmWwL/BxhNOPqdHdt7NvBhM7sVOHpAHF8Bfu/ua8VFbGVmdwPrAguBr7v7srge/md/z67/PfA/gI2AX5jZocAFwI/dfZ6ZfR74HmGYdhmhF/ZA7HmMBz5E2En9HTjY3f9R+l3NbBQwg7DuC8B/ASeWxDOHMttFDZ3x/1fjMj5MmfUVy3aK32fNuPz/5e43x7Y/Ik5f6u67mVlbbNuvE7ajo4DzS77HTELCfAV4GVhqZlsAfwQ2cvceM8sDLwJ7xPb6WWynHHClu18Y69sHODe261vAMe7+iJmdBuwHfCDGdjLwa0rWv7tPqfS96mi7YUnnCJrjDjN71MxeIuw8AQ6P/x8I/MndPwNsRtjJHuLu8wkb7MXu/pMyda7l7hMJR64nm9lHzGwKcBhhxz2B9/7Ay/mAmT0c/71A2PGfV1K+NTDN3T9B2CH8ANg77pSPAm4EVgAPALvFeSYDZmZrmdlHgXfc/UngauByd58AfBrYI/ZAAP4FOMfdtwBmAfuY2bqx7ChgdXpCjwAfHzDtW8BNcdl7A7sShuiOBJ519yll4vjHgDr+FfhCrDtHSIgVufvpwEvAQe7+X/3TY1K6DPiCu28DfBf4lZmtHT8yEfiiu29J3JmVqf4MQpLZJv5rAy4sKV9lu6gQ5oVx3T9OGCa8nfe2zbLry8zWAa4gbJ/bE3awPzOzTeJ8HwMmu3v/9vBvhKHQ2wlJ/vh4IAQhOWwBfJSQDDaJbfc08DjhAALCAchzcTv6BXCHu3+ccKB0sJkdGA+mrgEOj9vrhcD5ZrYpIYFMjtNPB8529wIl67+O7zXiKBE0x25xQ9yH8Idxh7u/AuDulwB/NLOTgJ8SdsBrVazpPb+K8/+dcES1LmFHd727d8ej5HIJpN/b7r5t/Ldp/7wlO46/ufsL8fWehCPVBWb2MOEPso+wg5wPTDWzrQhHsfcBkwh/TDeY2Zrx/Tlx3vsJf/Tbxrp7gT/F7/IKcDNwSPzjnBKXVa8iIZGWmg+cYmY3AvsDJ7h7X5l5342jjBvdfXFs0ysI7TEYuwML3P2vAO7+B8K6mxDL73T3N+LrhYR1OtBUwrmHd+L3uDRO61duuyjnW3Hdfwz4MLAl8J0a6+szhO3gl7Hst4Q2/0Ss89GS+AGOBX7h7r2Eg5oxwBdj2R7Ate7e4+5v8f71PJtwQAPhgGlWjGtn4jbt7ksJvZ+pcfoid18Yy25096lx+z0UOMjMzick1nJ/W7W+14ijoaEmcvc/m9mJwBwzW+juz5vZBYSjrsuBOwhd8Vwd1b1d8roY5+kdMG9hNWK73cyeibG8DLxZUpwn7MAO6J9gZhsTjnqXEIagngZ+D7xOOIr7NOEPLx9j2sndl8d51yP0JtYDVsYdRb+fELr/vcAN7l4aR0XxPMEEwpBG6fe62cw2J+y8Pwt8z8wmlKliYBylStuxDXgnvu5v934dNcLMx3lKtRHWOZRfp7XqKJ2/3jrex90Xm9m1hMT7Eyqvr92AJ919h/5544n1xcBBlGwz8Wh8b2B7M+s/IdsOnAjMje9LYytt++uBGfHgYhIhKeRZ9bv0f/deStokbgsfj8v7FXAxcBtwF2HbGihf5XuNSOoRNJm7zyUMp1wcJ00BfuTuVxOO4PYkbJgQNvBRq1RS2W+AL5jZ2Pj+CFbd8ZQVx2bHE45EB1oA7BWHNjCzvYFHgQ+4+/8jjC0fQ/hju5UwjPJBd38kHiHeD5wU5+0i9Br2KxeHu/+R0Ns4mTqHheI48neBV9397gFl1wIHuPt1hOGIN4D/xuq17b5mtk5czteAW+L0xcAn43ImE44q+5WrfwEwxcw2i/PsDmxMGOev1++AY81sVByDP46QgActjtfvBTxQY33dD2xuZrvGsm2BvxB6FAMdDdzr7h929/HxPNUEQmLYidCGh5rZaDMbDbx7kOHuK4DrCEf8N7j7cndfFpd/XFz2WMLR/u8J7beVmX0sVrEfYahoV+BBd59BSAKfp/zf1up8rxFBiaA1HA/sHcf0zwYuMrNHCd3newlDLhD+WI4xs1PrqTQONcwC/mRmDwJjWXWopF/pOYKHgXnAUXGMdmC9TxDG668zs0eAc4B9S47W5xNOZi909+cIR6XzS6r4MrCjmT1G+KOd6+7VhnyuAF5y90erfObEGPtCQvLahHAEOtA5hKGBR+Ky5xN6ME8AK8zsAWofNT9BGLJ6DOgmnvAEvg1Mj+13CPBQyTw3AteY2V79E2I7fh240cwWxXo+F4c56nUu8E/gYcLJ7VHA9NWYv9+FJe33OOGk//djWdn15e6LCUn+wtieVxPG1Z8vrdjMOggHIf+7dLq7/4XQGziRcML+QcJVcXcBzw2IbxahVzm7ZNpBwGdjXA8Q2niOu78cy66M6+Ikwrm3ucB6ZvYkYR2+CaxrZp28f/2/Ws/3Gklyug31yGVmnyR06WfG9ycBO5QO6bS6eDJxPnCNu/9ns+MRGYnUIxjZngYmmtmieNT0WWIXfziIVxotjv+ub3I4IiOWegQiIhmnHoGISMYpEYiIZNyw+x1BX19fsVBo3HBWPp+jkfUlSbEmY7jEOlziBMWalKHEOmpU/lXC1XyrGHaJoFAo0t1d6QrI1dfVNaah9SVJsSZjuMQ6XOIExZqUocQ6blznC5XKNDQkIpJxSgQiIhmnRCAiknFKBCIiGadEICKScUoEIiIZp0QgIpJxSgQiIhmnRCAiknHD7pfFItLiOtpZsryHQn7V48yOfBv0VHoCqDSLEoGINFRPoY+Trn+U3t5VH5E9c9p2NR/kLOnT0JCISMapRyAiqRnV3kZPhTINGzWPEoGIpKant8j0uQvLlmnYqHk0NCQiknHqEYhI6+top6fQV7lYw0pDokQgIi2vp9DHCRWGlEDDSkOloSERkYxTIhARybjEhobMbAfgAnefbGbbApcCBWAlcKi7v2xmXwOOBnqBc9395qTiERGR8hLpEZjZKcBsYHScdAnw7+4+GbgR+LaZbQicAOwMTAHOM7M1kohHREQqS2po6Flg/5L3B7r7w/F1O7AC+DRwn7uvdPelwDPAJxKKR0REKkhkaMjdbzCz8SXv/wFgZjsBxwO7EnoBS0tmWwaMrVV3Pp+jq2tMw2LN59saWl+SFGsyhkusrRLnWz0FVpS5j1C/XBFyQHt7vkxhhelAx6h82RvV9ddZaT6IbdM5umJ5Na3SrvVIKtbULh81swOA04H/7u6LzewNoLPkI51Ad616CoUi3d3LGxZXV9eYhtaXJMWajOESa6vE2ZNvq3op5yXTtqcIZW86R7HCdGDlO31Mn/vninVWmg+gUOgbdNu0SrvWYyixjhvXWbEslURgZgcTTgpPdvfX4uQHgO+b2WhgDWArYFEa8YiIyHsSTwRmlgdmAi8CN5oZwF3u/j0zmwncQzhXcbq7r0g6HhEReb/EEoG7Pw/sGN+uW+Ezs4BZScUgIiK16QdlIiIZp0QgIpJxSgQiIhmnRCAiknFKBCIiGadEICKScUoEIiIZp0QgIpJxSgQiIhmnRCAiknFKBCIiGadEICKScUoEIiIZp0QgIpJxSgQiIhmnRCAiknFKBCIiGadEICKScUoEIiIZp0QgIpJxSgQiIhmnRCAiknFKBCIiGadEICKSce1JVWxmOwAXuPtkM/tXYA5QBBYBx7l7n5l9DTga6AXOdfebk4pHRETKS6RHYGanALOB0XHSDOAMd58I5ID9zGxD4ARgZ2AKcJ6ZrZFEPCIiUllSQ0PPAvuXvJ8A3BVf3wLsAXwauM/dV7r7UuAZ4BMJxSMiIhUkMjTk7jeY2fiSSTl3L8bXy4CxwNrA0pLP9E+vKp/P0dU1plGhks+3NbS+JCnWZAyXWNOM862eAit6C2XLckVob89XnjkXuv1lP5OrMu9gy4ht0zm6Ynk1w2X9Q3KxJnaOYIC+ktedQDfwRnw9cHpVhUKR7u7lDQusq2tMQ+tLkmJNxnCJNc04e/JtnDB3YdmyS6ZtT2+FJAFAMZwMLPuZYoXpQykDCoW+QbfNcFn/MLRYx43rrFiW1lVDC81scnw9FbgHeACYaGajzWwssBXhRLKIiKQorR7BN4FZZtYBPAnMc/eCmc0kJIU24HR3X5FSPCIiEiWWCNz9eWDH+PppYFKZz8wCZiUVg4iI1KYflImIZJwSgYhIxikRiIhknBKBiEjGKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhknBKBiEjGKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhknBKBiEjGKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhkXHuzAxARGapR7W30VCjryLdBT2+q8Qw3SgQiMuz19BaZPndh2bKZ07ajI+V4hhsNDYmIZJwSgYhIxqU2NGRmo4ArgfFAAfga0AvMAYrAIuA4d+9LKyYREUm3R7A30O7uOwFnA98HZgBnuPtEIAfsl2I8ItnW0U5Pvq3sP3K5ZkcnKUrzZPHTQLuZtQFrA+8AOwJ3xfJbgL2A+dUqyedzdHWNaVhQ+XxbQ+tLkmJNxnCJtdFxLlnew0nXP1q27Idf2ob29nz5GXNULovlOSp8ptq8gy2rUZ7Pt9HVObrirMNl/UNysaaZCN4kDAs9BawH7APs6u7FWL4MGFurkkKhSHf38oYF1dU1pqH1JUmxJmO4xNroOAv5Nnp7C+ULiwyuLJYXqfCZwdZbxzIrlRcKfVXbbbisfxharOPGdVYsS3No6ETgVnffAtiGcL6g9KquTqA7xXhERIR0E8HrwNL4+jVgFLDQzCbHaVOBe1KMR0RESHdo6GLgcjO7h9ATOA14EJhlZh3Ak8C8FOMRERFSTATu/ibwpTJFk9KKQUREVqUflImIZJwSgYhIxikRiIhknBKBiEjG1ZUIzOyMAe/PSyYcERFJW9WrhszsCOBIYCsz2ztOzhN+A3BqwrGJiEgKal0+eg2wgHDN//fjtD7glSSDEhGR9FQdGnL3le7+PHAMsAGwKfARYIfkQxMRkTTU+4OyecD6wN/i+yJwdyIRiYhIqupNBBvG5wiIiMgIU+/lo0+Z2UaJRiIiIk1Rb49gIvCimS2O74vursQgIjIC1JUI3H3zpAMREZHmqCsRmNkVhBPE73L3ryYSkYiIpKreoaHr4v85YHtAw0IiIiNEvUNDt5a8/Z2Z3ZZQPCIikrJ6h4b2Knn7IcKPy0REZASod2hoWsnrFYDOD4jIsDCqvY2eCmUded2AGeofGjrczLYGPgo87e4PJxqViEiD9PQWmT53YdmymdO2Szma1lTvbaj/HZgF7AT8h5mdnGhUIiKSmnr7RV8GJrr7N4CdgQMSi0hERFJVbyLIuXsvgLu/A7yTXEgiIpKmek8W32tm84B7gF2A+5ILSURE0lSzR2BmRxGeRnYFMBa4y92/lXRgIiKSjlqPqjwL2Bq4xt1/Y2aPAzPMbB13P2d1F2ZmpwL7Ah3AT4G7gDmE21csAo5z977VrVdERAavVo9gKvBFd18OEJ9WdgBhZ75azGwy4aqjnYFJwMbADOAMd59IuH3Ffqtbr4iIDE2tRPCmuw+82dw7wLJBLGsK8BgwH7gJuBmYQOgVANwC7DGIekVEZAhqnSx+28w2c/e/9k8ws80YcCfSOq1HeObxPoTnHv8aaCtJNMsI5yCqyudzdHWNGcTiK9XX1tD6kqRYkzFcYm10nEuW99Deni9fmGNwZbE8R4XPDLbeOpY5mHnz+bZhs/4huW21ViL4NvBLM1sA/BXYhHBk/5VBLGsJ8JS79wBuZisIw0P9OoHuWpUUCkW6u5cPYvHldXWNaWh9SVKsyRgusTY6zkK+jd7eQvnCIoMri+VFKnxmsPXWsczBzFso9FEo9A2L9Q9D2wbGjeusWFZ1aMjdHyc8nWwhsCbwZ2Bndy//e+3q7gX+zcxy8bGXawIL4rkDCOcj7hlEvSIiMgQ1f0fg7kuBq4a6IHe/2cx2BR4gJKDjgOeAWWbWATwJzBvqckREZPXU+4OyhnD3U8pMnpRmDCIi8n66B6uISMYpEYiIZJwSgYhIxikRiIhknBKBiEjGKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhknBKBiEjGKRGIiGScEoGISMYpEYiIZJwSgYhIxikRiIhknBKBiEjGpfqEMhFJWUc7PYW+8mW5XLqxtKBR7W0sWd5DIb/qMXFHvg16epsQVfqUCERGsJ5CHyfMXVi27JJp26ccTevp6S3yzesfobe3sErZzGnb0dGEmJpBQ0MiIhmnRCAiknFKBCIiGadEICKScUoEIiIZl/pVQ2a2PvAQsCfQC8wBisAi4Dh3r3Ctm4iIJCHVHoGZjQJ+DrwdJ80AznD3iUAO2C/NeEREJP2hoYuAy4CX4vsJwF3x9S3AHinHIyKSeakNDZnZYcBid7/VzE6Nk3PuXoyvlwFja9WTz+fo6hrTsLjy+baG1pckxZqM4RJruTjf6imwosyPofrlitDenq9QmEBZLM9R4TMJLnOw9VaKNZ9vo6tzdOVlNkFS22qa5wi+ChTNbA9gW+AqYP2S8k6gu1YlhUKR7u7lDQuqq2tMQ+tLkmJNxnCJtVycPfm2ir8chvDr4XK/mgWgSOPLYnmRCp9JcJmDrbdSrIVCX8ttF0PZVseN66xYltrQkLvv6u6T3H0y8DBwKHCLmU2OH5kK3JNWPCIiEjT7XkPfBGaZWQfwJDCvyfGIiGROUxJB7BX0m9SMGEREJNAPykREMk6JQEQk45QIREQyTolARCTjlAhERDJOiUBEJOOUCEREMk6JQEQk45QIREQyTolARCTjlAhERDKu2TedExGAjnZ6CuWf0tqRb4Oe3pQDkixRIhBpAT2FvorPFZg5bTs6Uo5HskWJQESkjFHtbfRUKR9JPTUlAhGRMnp6i0yv8vS3kdRT08liEZGMUyIQEck4JQIRkYxTIhARyTglAhGRjFMiEBHJOF0+KtLi+q9nX7K8h0J+wLFbLteUmGRkUSIQaXH917O3t+fp7S28r+ySads3KSoZSTQ0JCKSceoRiIgMQrVbUAy320+klgjMbBRwOTAeWAM4F3gCmAMUgUXAce5e/haMIiItpNotKIbb7SfSHBo6GFji7hOBqcCPgRnAGXFaDtgvxXhERIR0h4auB+aVvO8FJgB3xfe3AHsB86tVks/n6Ooa07Cg8vm2htaXJMWajFaIdcnyHtrb8+ULc9DenicHq34mV2ZaveVJlMXysrEmvMzB1juoWGuU5/NtdHWOrjzvICW1raaWCNz9TQAz6yQkhDOAi9y9GD+yDBhbq55CoUh39/KGxdXVNaah9SVJsSajFWIt5NtWuSLoXUXo7S2UvWqov6yiauVJlMXyIhU+k+AyB1vvoGKtUV4o9CWyTQ1lWx03rrNiWapXDZnZxsAdwNXufi1Qej6gE+hOMx4Z5jra6cm3lf1Hh66DEKlXmieLNwBuA4539wVx8kIzm+zudxLOG9yRVjwy/OmpXiKNkeZh02nAOsCZZnZmnDYdmGlmHcCTvP8cgsjwUuW5wzD8LimU7EjzHMF0wo5/oElpxSDyrpKd9sBbNwx2h12thwLqpUjr0kCqZFLpTnvgSVjtsCVrlAhkRKr14HHdrE3kPUoEMiLVevC4btYm8h7ddE5EJOOUCEREMk6JQEQk45QIREQyTieLpbVV+5GWrvwRaQglAmlp1X6kpSt/RBpDiUCar8WO+kfSk6dE6qFEIE3Xakf9I+nJUyL1UCKQxhhwVN+o+/eIDEfDrVepRCANMfCofqTev6fqrStqDGMNZV4ZXoZbr1KJQGQ1VPsDrzWMNZR5RZKk3xGIiGScegQiIq2ixsON3uqp8gzlIVAiEBFpEbUebvSTgyeQT2C5GhoSEck4JQIRkYzT0JCISIpa8TJiJQIRkRS14mXEGhoSEck4JQIRkYxTIhARybimnyMwszbgp8A2wErgSHd/prlRjXBVfrRS9YZYQ7hddCueIBORoOmJAPg8MNrdP2NmOwI/BPZLZElldmSld8lsxbsCVjTYnTnVf7Tys0Mm8E6+Qkexrzjo20W34gkyEQlaIRHsAvwOwN3vN7NPJrWgcjvA0rtktuJdASuptjMfyvfQDlske3LFYrGpAZjZbOAGd78lvn8R2MzdKx3SLgZeSCs+EZERYlNgXLmCVugRvAF0lrxvq5IEoMIXERGRwWmFq4buA/YGiOcIHmtuOCIi2dIKPYL5wJ5m9kcgBxze5HhERDKl6ecIRESkuVphaEhERJpIiUBEJOOUCEREMq4VThanxsx2AC5w98kDpn8O+C7QC1zu7rOaEN77VIn1JOAIwu8pAI52d085vP5YRgGXA+OBNYBz3f3XJeUt0651xNpK7ZoHZgEGFIDD3f3ZkvJWatdasbZMu8Z41gceAvZ096dKprdMm/arEmvD2zQzicDMTgEOAd4aMH0UcDHwqVh2n5nd5O7/TD/Kd2MqG2u0PXCouz+UblRlHQwscfdDzOyDwELg19CS7Vox1qiV2vVzAO6+s5lNBmYQb7vSgu1aMdaoZdo1tt3PgbfLTG+lNq0Ya9TwNs3S0NCzwP5lpm8FPOPur7t7D3AvMDHVyFZVKVaACcCpZnavmZ2aYkzlXA+cWfK+9IeArdau1WKFFmpXd/8lcFR8uynwcklxS7VrjVihhdoVuAi4DHhpwPSWatOoUqyQQJtmJhG4+w3AO2WK1gaWlrxfBoxNJagKqsQKcB1wDLA7sIuZ7ZNaYAO4+5vuvszMOoF5wBklxS3VrjVihRZqVwB37zWzK4FLCfH2a6l2haqxQou0q5kdBix291vLFLdUm9aIFRJo08wkgioG3uKiE+huTijVmVkO+JG7vxqPXH4DbNfkmDYG7gCudvdrS4parl0rxdqK7Qrg7l8BtgBmmdmacXLLtSuUj7XF2vWrhB+u3glsC1xlZhvGslZr04qxJtWmmTlHUMWTwOZmti7wJrAroVvWitYGFpnZVoSxzN0JJ0Cbwsw2AG4Djnf3BQOKW6pda8Taau16CPAv7n4esBzoI5yIhdZr12qxtky7uvuu/a/jDvaYknMALdWmNWJNpE0zmwjM7MvAWu7+H/Es/K2EHtLl7v735kb3fgNiPY1wVLsSWODuv21iaKcB6wBnmln/+PssYM0WbNdasbZSu94IXGFmdwOjgG8A+5tZK26vtWJtpXZ9H+0D3qNbTIiIZJzOEYiIZJwSgYhIxikRiIhknBKBiEjGKRGIiGRcZi8fFakm3jfn/wJPAEXC9dt/BQ6KP+QZ+PlNgG3c/SYz+xEww91fTC9ikcFTIhCp7A/ufmD/GzO7FtiXVW+jAOGHPVsCN7n7N9IJT6QxlAhE6mBmHcCHgNfNbDawMfBB4BbgLOA7wJj47O2TCPeCORD4CLA+4YZsJ7r7rfHeMGcT7m/zOvCou5+V6hcSKaFzBCKV7W5md5rZE8CfgfmEO8Pe7+5TgF2AY929AJwPXFv6jINopbtPBaYDJ8b7988Eprr7bpS/zbBIqpQIRCr7Q3ww0ESgB3gOeA34lJn9gnAP+zVq1LEw/v83YDQwDnjD3ftv13xPo4MWWV1KBCI1uPsSwoNtZgMnAt3ufhDwQ8JwUI5ws7Vyf08D7+HyCtBpZuPi+x2TiVqkfkoEInVw9ycIQzpbA3vHcwE/A/4CbAQ8BuxnZgdWrgXcvQ84Hvitmd0ObELlZ0+IpEI3nRNJWXyq1Ax3X2lm1wC3uftVzY5LsktXDYmkbxlwv5ktB54H/rO54UjWqUcgIpJxOkcgIpJxSgQiIhmnRCAiknFKBCIiGadEICKScf8f1IQKzeJMngkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# explore ratings on each brewery\n",
    "rating_brewery = pd.merge(df_rating_ba, df_brewery_ba, how='left', left_on='brewery_id', right_on='id')\n",
    "rating_brewery = rating_brewery[['overall', 'brewery_name']]\n",
    "mean_rating_brewery = rating_brewery.groupby('brewery_name').mean()\n",
    "mean_rating_brewery = pd.DataFrame(mean_rating_brewery.to_records())\n",
    "\n",
    "# draw rating brewery distribution\n",
    "sns.histplot(data=mean_rating_brewery, x='overall')\n",
    "plt.xlabel('Rating')\n",
    "plt.title('Rating Brewery Distribution on BeerAdvocate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### RateBeer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:26.251989Z",
     "start_time": "2022-11-18T21:04:24.755781Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAESCAYAAADwnNLKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAedklEQVR4nO3de5gcVZnH8e+kJyEEhgywI14h6uIrrgomCAgBIgg84CUu7K6IxgsqF0ERZFG5rPcLCFFBFA0CgiAoIS6oCIpcDIqsIQgI/rgo6AqsoE4IBBJmMvtHnQnN0NNdk0xXd0/9Ps+TJ11Vp069fbqm3qpTt66hoSHMzKy8JrU6ADMzay0nAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5LrbnUAVl9EDAG3AYPAEDANeAQ4VNJvGsz7XmCKpK9FxCFAr6QvjENMM4B7gFvTqAqwAjhK0vXrWn+zRcQ5wB7AQ2nUFOA64BOSHoyI5wIXS9qxTh0vBE6WtF+NaWvmj4hPAP8k6fAxxrgAOEPSkog4E7hQ0s/GUkez1Gi/ScCGZPGe1GDeUdttRLk5wOWA0qhu4GGy9f6OtQ7eanIi6AyvlfTw8EBEHA2cBrymwXyzyZIIks4Y55gel7RNVUz/AZwDbDnOy2mWL0k6GSAiuoCPAT+JiFmS7gdGTQLJFkDUmpBz/kb2AL6R6nvvOtbVDGvaDyAiNgfuiIhLJf2+znyjtlsN94xYx44BTiVrGxtHTgQdJiK6gc2Bv6fhzcg2GJsBzwbuA/4D2Al4E7BHRDwO9JH2TCPiXrKN9u6prnMlnZDq+yjwHmA52V7ymyXNyBHapsADqY45wFeAx8j2FF8N7AkcT7b3vQI4GrgzxbuZpMci4hvASyXtmuq5K32HR4Cvplgnk+0dfy4dmfwCuAOYAXwbeJmkt6X5ZwOnSXpVvcAlDQGfi4h3pfb6PXCbpA0j4qXAt4CpQBdwZmrvM4HnRcQVwMEj4ngn8FNJG6ZFbBUR1wGbAEuB90tann6Hfxs+shseBv4VeC5wfkS8AzgR+KqkiyPizcDHyfbCl5Mdhd2YjjxmAM8h29j+BXi7pAeqv2tETAbmk/32g8CvgSOr4jmHGutFDs9P7bM8LedYYC6wPrAB2e99aXW7SdorInZM32+DFM8nJf1wZOUpWW9CWsfSuPcA709t8TfgcEm/j4gpqc5dyY5WlwIflPRI+o6/Bl4JHCtpUc7vN6H5HEFnuDoibomI+8k2ngDvTv/vD/xK0muAF5FtZOelFfxSsj2302vUuaGkncn2XI+OiBdGxF7Au8g23LOAnjoxrR8RN6d/95Ft+D9fNf3lwFslvZJso/I5YJ+0UT4IuAR4ArgReG2aZw4QEbFhRLwMeDJ1A5wHnCVpFrAd8Lp0BALZBujTkl4CLADeEBGbpGkHAWM5Evot8IoR4/4TuCwtex9gF7IuuveS7bHuVSOOB0bU8c/AfqnuLrKEOCpJxwH3A2+T9Ovh8SkpnQHsJ2lr4L+A/46IjVKRnYF/l/RSsiR8SI3qjydLMlunf5OAL1ZNf8Z6MUqYR6bf/g8R8TBwDPB6SX+JiC2A1wFz0u9/HPApSYNUtVtEbAycTba+ziRLHF9PRxcALx5ex1J7HAp8ObXFrmQJd+e0Tp0EDG/UPwoMALNSO90PVHeJ3iZpKyeBpzgRdIbXpj+oN5CdI7ha0l8BJH0F+GVEHAV8jWwDvOGoNT3lv9P8fwH+Sra3tQ/wfUn9aS+5VgIZ9rikbdK/LYbnrdpw/FnSfenzHmR7qlelP+rzgdVkG8hFwN4RsRXZXuz1ZHtyc4GFEbFBGv50mvcGssSyTap7APhV+i5/BX4IzEsbmb3SsvIaIkuk1RYBx0TEJcC+ZHuWq2vMuyaOGi6R9FBq07NZ+66N3YCrJP0BQNLPyX67WWn6NZIeSZ+Xkv2mI+1N1pf/ZPoep6Vxw2qtF7V8KXXbvILsN1kJXJPmvQ94B/C2iPgCWUKqtU6+hmy9+EH6bX9M9hu8Mk2/p2odew7wAeDKiOgBXk+2/vwyzXsSsHHaCXgD2fqzNE17M/CyquX+YpTvVFpOBB1E0k3AkcA5qVuEiDgR+BTZibtvAleS7XU28njV56E0z8CIeQfHENvPgLvJ9tgBHq2aXCHbgA3/UW8D7EB2/mIR2YZoT+Cn6d+eZF1CF6d5u4AdR8z7uVT3SkkDVcs6HTgQOABYKKk6jlGlrodZPHUCfPh7/ZDsvMf3gFcBt0bE82tUMTKOatXtOAl4Mn0ebvdhUxqEWUnzVJtE1l0GtX/TRnVUz5+3jjUkPQbMI0vWRwJExEyypLgR2fp4Yp1Y7qixXlwxyrLOJevOflma97yq+WYC2wL/SNOOqJq2HVmX27Bc60SZOBF0GEnfJetO+VIatRfwZUnnke3B7UH2hwDZhn3yMyoZ3Y+A/SJiehp+D8/c8NQUES8h66NeWmPyVcCeqWuDiNgHuAVYX9L/kl0NcgjZRuMKsm6UTSX9Nu3h3gAclebtJTtqmFsrDkm/JDvaOJqc3UIRUSHrZnlY0nUjpl0AvEXShWT90Y8AL2ZsbfumiNg4Led9ZFfDQJa8t03LmUO2dzysVv1XAXtFxIvSPLsBLyDr887rJ8ChETE5IiYBh5El37Um6R/Ah4FPRsTzyLrPfiNpPnAt2R55rXXyBmDLiNglfZ9tgLuA59VaTkTsNLxIsvXkrREx3GaHkLUPadrhETElfccFPL3b0kZwIuhMhwP7pD79TwEnR8QtZOcEFpMdMkO2wTkkIj6Wp9LU1bAA+FVE/AaYzjO7SoZVnyO4mWzv/SBJd44sKOl2sv76CyPit8CngTdV7a0vIjuZvVTSH8n2Sqv7bw8AdoiIW8k2et+VVK/L52zgfkm31Ckz3Me9lCx5bU7WvTXSp8m6OH6blr2I7CT67cATEXEjjY/AbifrsroV6Oep/uqPAEek9psHLKma5xLgOxGx5/CI1I7vBy6JiNtSPW+UtKzB8qt9BngQuJns5PZk4IgxzF9T+j2WACcD3wX+KSLuIPvujwKbpC6d6nZ7mCzpfzG173lk5wvuTdW+uGodu4Xs/MC+qety+Ejjp2naAWnaENlvdi/Z73o72e/z4XX9jhNZlx9DbcMiYluyLphT0/BRwPaS3tLayPJLV1UtAr4j6aJWx2PWCXxEYNXuBHaOiNvS3vfupC6ZTpCuNHoo/ft+i8Mx6xg+IjAzKzkfEZiZlZwTgZlZyXXcIyZWr149NDhYvzurUumiUZl25dhbw7G3hmMvzuTJlYfJrs57ho5LBIODQ/T3j3ZFY6a3d1rDMu3KsbeGY28Nx16cvr6e+0ab5q4hM7OScyIwMys5JwIzs5JzIjAzKzknAjOzknMiMDMrOScCM7OScyIwMys5JwIzs5LruDuLzWzi6Zm+PlOn5NscPbFqgOXLHm9c0HJzIjCzlps6pZv9Tl+cq+zCw2azvMnxlI0TgZk1TZ49/b6+noKisdE4EZhZ0zTa0+/urjAwMMjCw2YXGJWN5JPFZmYl50RgZlZyTesaiojtgRMlzYmIC4Fnp0kzgBsk7R8RpwI7wZpzP3MlLWtWTGZm9kxNSQQRcQwwD3gMQNL+afzGwNXAkanoTGAvSQ83Iw4zM2usWV1D9wD71hj/SeA0SQ9ExCRgS+CbEXF9RBzYpFjMzKyOphwRSFoYETOqx0XEs4DdeepoYAPgNGA+UAGujojfSLqlXt2VShe9vdPqLr9SmdSwTLty7K3h2Junu7sy6rSuqun1yo3UDt+33dt9LIq8fPTfgAskDabhFcBXJK0AiIifA1sDdROB31ncvhx7a7Rz7H19PQwMDI46ffjyUaBuuZHa4fu2c7vXUu9+jSKvGnodcHnV8EuAxRFRiYjJwGzgpgLjMTMzik0EAfxheEDSHcD5wA3AtcC5kn5XYDxmZkYTu4Yk3QvsUDX8LzXKnASc1KwYzMysMd9QZmZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl17SX10fE9sCJkuZExEzgMuCuNPnrki6KiPcBBwMDwGck/bBZ8ZhZfT3T12fqlMabhCdWDbB82eMFRGRFaUoiiIhjgHnAY2nUTGC+pFOqyjwb+CCwLTAVWBwRP5W0shkxmVl9U6d0s9/pixuWW3jYbJYXEI8Vp1lHBPcA+wLnpeFZQETEXLKjgg8B2wHXpw3/yoi4G3gl8D9NisnMzGpoSiKQtDAiZlSNuhE4U9KSiDgO+DhwM7CsqsxyYHqjuiuVLnp7pzUoM6lhmXbl2FvDsWe6uyu5yo1lefXq7KqannfZY11+s3TyOjNS084RjLBIUv/wZ+A04Dqgp6pMD9BPA4ODQ/T3r6hbprd3WsMy7cqxt4Zjh76+HgYGBnOVzbu8RnV2d1fWTM+77LEsv5k6bZ3p6+sZdVpRVw1dERHbpc+7A0vIjhJ2joipETEd2Aq4raB4zMwsKeqI4FDgqxGxCngQOEjSIxFxKvALsoR0nKQnCorHzMySpiUCSfcCO6TPNwE71iizAFjQrBjMzKwx31BmZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWckVdPmpmLZD3QXJjsWpgdd2bk6zzOBGYTWB5HyQH2cPk8pjSPWnc67TWcteQmVnJ+YjAzDpK3q4pvzchPycCM+soebum/N6E/Nw1ZGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJNe9ZQRGwPnChpTkRsA5wGDAIrgXdI+r+IOBXYCdY8EmSupGXNisnMzJ6pKYkgIo4B5gGPpVFfAT4g6eaIOBj4CHAUMBPYS9LDzYjDzMwaa9YRwT3AvsB5aXh/SQ9ULfOJiJgEbAl8MyI2A74l6axGFVcqXfT2TmtQZlLDMu3KsbfGRI69u7uSu668Zcerzq6q6c2Is5m/aSevMyM1JRFIWhgRM6qGHwCIiB2Bw4FdgA3IuovmAxXg6oj4jaRb6tU9ODhEf/+Kusvv7Z3WsEy7cuytMVFj7+vrYWBgMHddecuOV53d3ZU105sRZzN/005bZ+q9w6Gwk8UR8RbgDOD1kh4CVgBfkbRC0nLg58DWRcVjZmaZQl5MExFvBw4G5kj6exr9EuDCiJhJlpBmA98uIh4zM3tK0xNBRFSAU4E/AZdEBMC1kj4eEecDNwBPAudK+l2z4zEzs6drWiKQdC+wQxrcZJQyJwEnNSsGMzNrzDeUmZmVnBOBmVnJORGYmZWcE4GZWck5EZiZlZwTgZlZyTkRmJmVnBOBmVnJORGYmZWcE4GZWck5EZiZlZwTgZlZyTkRmJmVnBOBmVnJORGYmZWcE4GZWcnlSgQRcfyI4c83JxwzMyta3TeURcR7gPcCW0XEPml0BZgMfKzJsZmZWQEavaryO8BVwLHAZ9O41cBfmxmUmZkVp27XkKSV6d3DhwCbAVsALwS2b35oZmZWhLwvr78YeBbw5zQ8BFxXb4aI2B44UdKciPhn4Jw0323AYZJWR8T7gIOBAeAzkn449q9gZmbrIm8ieLakHfNWGhHHAPOAx9Ko+cDxkq6JiDOAuRHxK+CDwLbAVGBxRPxU0sr84ZuZ2brKe/no7yPiuWOo9x5g36rhWcC16fPlwOuA7YDrU/fTMuBu4JVjWIaZmY2DvEcEOwN/ioiH0vCQpFETg6SFETGjalSXpKH0eTkwHdgIWFZVZnh8XZVKF7290xqUmdSwTLty7K0xkWPv7q7kritv2fGqs6tqejPibOZv2snrzEi5EoGkLddxOaurPvcA/cAj6fPI8XUNDg7R37+ibpne3mkNy7Qrx94aEzX2vr4eBgYGc9eVt+x41dndXVkzvRlxNvM37bR1pq+vZ9RpuRJBRJxNdqJ3DUkHjiGGpRExR9I1wN7A1cCNwGcjYiqwHrAV2YlkMzMrUN6uoQvT/13ATGAs5wsAPgwsiIgpwB3AxZIGI+JU4Bdk5yqOk/TEGOs1M7N1lLdr6IqqwZ9ExJU55rkX2CF9vhPYtUaZBcCCXJGamVlT5O0a2rNq8DlkN5eZmdkEkLdr6K1Vn58AxnJ+wMzM2ljerqF3R8TLgZcBd0q6ualRmZlZYfJ2DX0AOAD4NXB0RHxP0slNjczMRtUzfX2mTnnqz7fepYFmjeTtGjoA2FnSQERMBn4JOBGYtcjUKd3sd/pi4OnX4o+08LDZRYZlHSrvIya6JA0ASHoSeLJ5IZmZWZHyHhEsjoiLya75nw1c37yQzMysSA2PCCLiILK3kZ1N9iygayX9Z7MDMzOzYtRNBBHxCWBPYLKkHwHnArtFxAkFxGZmZgVodESwN/DvklbAmruF3wK8qclxmZlZQRolgkerHh8NrDlZvLx5IZmZWZEaJYLHI+JF1SPS8NAo5c3MrMM0umroI8APIuIq4A/A5sBewDubHZiZmRWj7hGBpN+RvZ1sKbABcBOwk6SlBcRmZmYFaHgfQXqf8LkFxGJmZi2Q985iMzOboJwIzMxKzonAzKzk8j5ryMyso6waWJ378dxPrBpg+bLHmxxR+3IiMLMJaUr3pDWP6m5k4WGzS32XbGGJICLeBbwrDU4FtgF2BC4D7krjvy7poqJiMjOzAhOBpHOAcwAi4nTgLGAmMF/SKUXFYWZmT1f4yeKI2Bb4F0nfBGYBr4+I6yLiWxHh9+2ZmRWsFecIjgU+mT7fCJwpaUlEHAd8HDi63syVShe9vdPqLqBSmdSwTLty7K3RibF3d1cA6Kr6XK/cWOocr3KNylbH3uo4x/r7d+I6M5pCE0FE9AIvlXR1GrVIUv/wZ+C0RnUMDg7R37+ibpne3mkNy7Qrx94anRZ7X1/PmvcU13tnMVB32tqWHa86q2NvdZxj/f07cZ0ZTdFHBLsAP6saviIiPiDpRmB3YEnB8Zi1jZ7p6zN1ii/ks+IVvdYF2VNMhx0KfDUiVgEPAgcVHI9Z25g6pXtMlzuajZdCE4GkL44YvonsElIzM2sRP2LCzKzknAjMzErOicDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOT8YBOzJvMzhKzdee00a7K8zxDy84OsVdw1ZGZWck4EZmYl50RgZlZyTgRmZiXnRGBmVnJOBGZmJedEYGZWck4EZmYl50RgZlZyTgRmZiVX6CMmImIpsCwN/hH4LHAOMATcBhwmaXWRMZmZlV1hiSAipgJImlM17lLgeEnXRMQZwFxgUVExmZlZsUcEWwPTIuLKtNxjgVnAtWn65cCeNEgElUoXvb3T6i6oUpnUsEy7cuyt0ezYu7sr41quumxXg/nWps7xKteobHXsrY5zrL9/J6/vIxWZCFYAJwNnAluSbfi7JA2l6cuB6Y0qGRwcor9/Rd0yvb3TGpZpV469NZoZe19fDwMDg7nK5i1XXba7u1J3vrWpc7zKNSpbHXur4xzr799p63tfX8+o04pMBHcCd6cN/50R8TeyI4JhPUB/gfGYrTW/Y8AmkiLX5AOBVwDvj4jnAhsBV0bEHEnXAHsDVxcYj9lay/uOAfB7Bqz9FZkIvgWcExGLya4SOhB4GFgQEVOAO4CLC4zHzMwoMBFIWgUcUGPSrkXFYGZmz+QbyszMSs6JwMys5JwIzMxKzonAzKzknAjMzErOicDMrOScCMzMSs73yJtZ6a0aWF33WTzDnlg1wPJljxcQUbGcCMys9KZ0T8r1yJCFh81meQHxFM2JwKyKHyZnZeQ13qxK3ofJ+UFyNpH4ZLGZWck5EZiZlZwTgZlZyTkRmJmVnBOBmVnJORGYmZWcLx+1tVbrmvvR7s6cqHdkmk0ETgS21kZec9/dXWFgYLBm2e8evGOpb+E3a2dOBCUxljtmm7ExznsLf96EAU4aZuOlsEQQEZOBs4AZwHrAZ4D/BS4D7krFvi7poqJimgjGsoHPsyGG1j5PJW/CAB9lmI2XIo8I3g78TdK8iNgUWAp8Cpgv6ZQC45hQmvFIhLxPYmy1sj8ozGy8FJkIvg9cXDU8AMwCIiLmkh0VfEiS/2ZbbCwbWDPrfIUlAkmPAkRED1lCOJ6si+hMSUsi4jjg48DR9eqpVLro7Z1Wd1mVyqSGZdrV2sTe3V0Z13JrW2dXg/laGedY1pl2ac+8Zcer3cdSdrzqrI69neOsNryedPJ2ZqRCTxZHxAuARcDXJF0QEb2S+tPkRcBpjeoYHByiv39F3TK9vdMalmlXY429r69n1Ct1Rspbbixlq8vVu2pobescr7J515l2as+8Zcer3cdSdrzqrI69neOsNrwuddp2pl53b2E3lEXEZsCVwEcknZVGXxER26XPuwNLiorHzMwyRR4RHAtsDJwQESekcUcBX46IVcCDwEEFxmMlkffkdyecIDdrhiLPERwBHFFj0o5FxWDllOfk93AXhU+AWxn5WUNmZiXnRGBmVnJOBGZmJedEYGZWcn7onJlZTiOvQKt3pVknPePKicDMLKfqK9Aa3cjXSc+4ciJoQ2N5oqiZ2bry1qYN5X2iKPjBb2a27nyy2Mys5JwIzMxKzonAzKzknAjMzErOicDMrOScCMzMSs6Xj5qZNUHe92C0wx3ITgQF8o1iZuWR5z0Y0B53IHurVKA8N4p1d1e46ODXFBSRmZnPEZiZlZ6PCNaRu3vMbF3kPZcAzTuf4C3YOvJzgcxsXeQ9lwDNO5/Q8kQQEZOArwFbAyuB90q6u7VRmZmVRzucI3gzMFXSa4CPAqe0Nhwzs3Jp+REBMBv4CYCkGyJi22YtaCz9+SsHBlmvu9KsUMzM2kbX0NBQSwOIiDOBhZIuT8N/Al4kaWCUWR4C7isqPjOzCWILoK/WhHY4IngEqD5lPqlOEoBRvoiZma2ddjhHcD2wD0BE7ADc2tpwzMzKpR2OCBYBe0TEL4Eu4N0tjsfMrFRafo7AzMxaqx26hszMrIWcCMzMSs6JwMys5NrhZPE6iYjtgRMlzRkx/o3AfwEDwFmSFrQgvLrqxH4U8B6yeyYADpakgsOrKSImA2cBM4D1gM9IurRqelu3e47427ntK8ACIIBB4N2S7qma3rZtnyP2tm33YRHxLGAJsIek31eNb9t2z6ujE0FEHAPMAx4bMX4y8CXg1Wna9RFxmaQHi4+yttFiT2YC75C0pNiocnk78DdJ8yJiU2ApcCl0RrtTJ/6kndv+jQCSdoqIOcB8YC50RNuPGnvSzu0+3L7fAB6vMb6d2z2XTu8augfYt8b4rYC7Jf1D0ipgMbBzoZE1NlrsALOAj0XE4oj4WIEx5fF94ISq4eqb/zqh3evFD23c9pJ+AByUBrcA/q9qclu3fYPYoY3bPTkZOAO4f8T4tm73vDo6EUhaCDxZY9JGwLKq4eXA9EKCyqlO7AAXAocAuwGzI+INhQXWgKRHJS2PiB7gYuD4qsmd0O714oc2bnsASQMR8W3gNLL4h3VC248WO7Rxu0fEu4CHJF1RY3Lbt3seHZ0I6hj52IoeoL81oYxNRHQBX5b0cNrD+BHwqhaH9TQR8QLgauA8SRdUTeqIdh8t/k5oewBJ7wReAiyIiA3S6I5o+1qxd0C7H0h20+s1wDbAuRHx7DStI9q9kY4+R1DHHcCWEbEJ8CiwC9mhXSfYCLgtIrYi63PcjezkZluIiM2AK4HDJV01YnLbt3uD+Nu97ecBz5f0eWAFsJrsxCu0eds3iL2t213SLsOfUzI4pOocQFu3e14TKhFExAHAhpK+ma5CuILsqOcsSX9pbXT1jYj9WLI91pXAVZJ+3NronuZYYGPghIgY7mtfAGzQIe3eKP52bvtLgLMj4jpgMvAhYN+I6IR1vlHs7dzuz9DJ25pa/IgJM7OSm6jnCMzMLCcnAjOzknMiMDMrOScCM7OScyIwMyu5CXX5qNl4Sc/D+R5wOzBEdq37H4C3pZueRpbfHNha0mUR8WVgvqQ/FRex2dpzIjAb3c8l7T88EBEXAG/imY9HgOwmqJcCl0n6UDHhmY0PJwKzHCJiCvAc4B8RcSbwAmBT4HLgE8BHgWnp3dtHkT03Z3/ghcCzyB60dqSkK9JzdD5F9oyafwC3SPpEoV/IrIrPEZiNbreIuCYibgduAhaRPTX2Bkl7AbOBQyUNAl8ALqh+t0GyUtLewBHAkem5/KcCe0t6LSMea2zWCk4EZqP7eXpp0M7AKuCPwN+BV0fE+WTPoV+vQR1L0/9/BqYCfcAjkoYfw/yL8Q7abKycCMwakPQ3shfanAkcCfRLehtwCll3UBfZQ9Rq/T2NfIbLX4GeiOhLwzs0J2qz/JwIzHKQdDtZl87LgX3SuYCvA3cBzwVuBeZGxP6j1wKSVgOHAz+OiJ8BmzP6eynMCuGHzpkVLL2Ba76klRHxHeBKSee2Oi4rL181ZFa85cANEbECuBe4qLXhWNn5iMDMrOR8jsDMrOScCMzMSs6JwMys5JwIzMxKzonAzKzk/h/FZ0zHD844BwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# explore ratings on each brewery\n",
    "rating_brewery_rb = pd.merge(df_rating_rb, df_brewery_rb, how='left', left_on='brewery_id', right_on='id')\n",
    "rating_brewery_rb = rating_brewery_rb[['overall', 'brewery_name']]\n",
    "mean_rating_brewery_rb = rating_brewery_rb.groupby('brewery_name').mean()\n",
    "mean_rating_brewery_rb = pd.DataFrame(mean_rating_brewery_rb.to_records())\n",
    "\n",
    "# draw rating brewery distribution\n",
    "sns.histplot(data=mean_rating_brewery_rb, x='overall')\n",
    "plt.xlabel('Rating')\n",
    "plt.title('Rating Brewery Distribution on RateBeer')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Generally, the distributions of brewery ratings on BeerAdvocate and RateBeer are similar. The majority of ratings fall between 3.0 and 4.0. But on RateBeer, there are relatively more ratings between 1.5 and 2.5."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Rating Distribution of the detailed scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:16:46.658694Z",
     "start_time": "2022-11-18T21:16:43.519959Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_title = ['appearance', 'aroma','palate','taste','overall']\n",
    "plt.figure(figsize=(12,8))\n",
    "\n",
    "\n",
    "for index, i in enumerate(data_title):\n",
    "\n",
    "    #creating subplots\n",
    "    ax=plt.subplot(2,3,index+1)\n",
    "\n",
    "    #plotting columns\n",
    "    ax=sns.histplot(data=df_rating_ba, x=i, color = 'purple', binwidth=0.25, label='x')\n",
    "    ax=sns.histplot(data=df_rating_rb, x=i, color = 'pink', binwidth=0.25)\n",
    "\n",
    "    # setting figure legend\n",
    "    plt.legend(title='Website', loc='upper left', labels=['BeerAdvocate', 'RateBeer'])\n",
    "\n",
    "    #renaming the axes\n",
    "    ax.set(xlabel=\"Rating\", ylabel=\"Count\")\n",
    "\n",
    "    # setting figute title\n",
    "    ax.set_title(f\"{i} Distribution on two website\")\n",
    "\n",
    "    # visualizing illustration\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### The abv rankings on the two datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:30.726699Z",
     "start_time": "2022-11-18T21:04:29.714151Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average abv% of beers from BeerAdvocate is 7.5%.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZwAAAESCAYAAADJ+2ORAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAlpklEQVR4nO3df5xcdX3v8dfMDpuwusmiN4i2/Ciob6ktYFASgUgUKzcgxWKpiGAI5WdRoLbViwQJt3itVGlBaXiYAAkClQriLSAkVhGSCIQiXKGmHwyI2N6iIbdLgttk3d25f3zP6rjMzkyyM2dml/fz8cgjM9/znXM+58zZ85nv93zPOYVyuYyZmVmrFdsdgJmZvTw44ZiZWS6ccMzMLBdOOGZmlgsnHDMzy4UTjpmZ5aLU7gBs6pG0C/As8FhELMjK5gNfjIjfaXNsZWBWRDw/pvwxYH5E9LcjrkZl8T8BDJN+MJaBSyLi6y1Y1rHAPwInRsQtNerdCdwaESuaHcM4yzsGmBMRn8pjedY8buFYKxwPPAa8VdL+bY6lIRFxUKcnmwrvzOI9ADgTuElSdwuW8yfATcCftmDeE/E24FXtDsJ2nFs41grnAF8BngLOB87Oyl8p6Vbg9UA/6WD5U+AnwBsj4jkASQ8BSyLi7tEZZi2kK4GfA68kHXQuB+YCvUABOD0i1klaAWwBfhfYE/g+8OGIeLFifnsA/wQsjYirR1s+wHuBPwBGgDcAA8DCiNgg6fXAdaSD3X9ky7wx+/cF4DDgF8DTwKLK5WXL/E1gKbBP9tmVEfHXkvYBvgV8A5gD7AZ8PCJub2BbvwrYBAxlyzgU+CzwClIr6NKIuDOb9sekJFIENgMfiYh/zbbXq4D9gDsj4hOS9gXmA3sDGyTNjYgHs/m8DlgJvA74MbB7Vn4mcGxEHJu9f1O2XnsBhwJ/DfQAg8DiiLgnq3chsDBbhx8Cp2avl2bfwauBrcBJQB9pf+qS9EJEXDTeejWw7SxnbuFYU0n6beDtwFdJB6UPS3p1NnlP4IqIOAi4GfhyRLwA3A6cnH1+f2APYFWV2f8O8MHsl/1s0gHv7RHx29my/kdF3YOB/w7sTzrAn1Ax7TdJB8LPRMTVVZZzBPDRrPvvoYr5fhn4+6z8vGw9yf6fDxwYEQeTEs4BVeZ7E3BvRPwuKTmdLOnEbNq+wKqIOCRb3t9W+fyoeyU9JmkjcBfw6YgYkbQbcD1wSkTMBo4DlkraS9IRpIP6vIh4CylZVya0noh4c0R8Int/NnBXRPyM9OOhspVzNfBgRLw52w5vysr/Hjg8S+YAi7J4+oBbgfOz724hcKOk35L0+6QE8/Zsu/4I+AiwAOiPiLdHxBuBh0mJ5CHgGuCWLNnUWy/rIG7hWLOdQ/qVvBnYLOlHpJbMA8D3I+K7Wb0VpIPhTGA58HfA50gHqesiYqTKvH8SET8GiIgHJC0GzpK0H+mAv7Wi7j0RsR1A0uP8ehfMN4B/IyW9ah6JiH/LXn8POD47mB8CvCNb/gZJ38rqPE5qTTwkaRVwW0Ssr5yhpFeQksx7ss+/kLUsFgAPklpG36hYZq0uo3eOnoOS9Bbgm5J+AMwEXgt8XdJo3TIp+b2D1LL8bsW03SSNLmdtRazTSN/DaVnRSmCdpD0j4ifAu4E/z9Zjo6RvZ6+3SvoaKZH+DfAhYB6p1bYxSxZExL9IWkf6zt4CfDUi/jOb9rGKOJ6W9NEs7vmkfWisY8Zbr4j4fzW2obWBWzjWNNlB9RTSr9xnJD1DOgB+BNiFdFCuVAZ+ERFrgJKkQ0jdJteNs4jKLrFjSL/uAf436VdvoaLuf41ZTuW0s0hdZh+jumqfHcreV85nGCA793Mg6SA8DNwi6U/GzLM45rOjZbtkrwcrkuzYeMcVEY+SksXhQBewITu/c1DWkpxLai12kVqUo+WzgbcC/5nNqrL7749I3XpfzL7Df8hi+ug48Q1VvF4GfJjUutwQET/Klj32po2j6z5UOU1Sn6R9JJ0DXEvq0ryZ1Hqqtk3qrZd1ECcca6YPkfrQXxcR+0TEPqSuoleSzo8cKOmgrO5ZwNqIGMjeLyedB/l+9iu6nt8D7oiIpcA/A+8jHXwa8QCpG2axpIZGzUXEVmAd6Zc/kn4LOBIoS3ovqYvuuxGxBLiBdI5p7OcfBM7NPj+TdGD+ZoMxVyVpd9IB9uFs/m+Q9I5s2kGkcyK/QUo6H5T02uyjZ2cxV3MOqZtu74rv8WzgjOxHxT2kViuS9gLeWbGeD5ISw6dIyQfS9n5T9oMCSW8mtbi+QzqPdrykGVndJaQfAkcBKyLiWiCAY/nV9zvErxL1jqyXtZm71KyZziGdo/llSyYi+iVdRToHsAG4JDsh/TPSQX/USuB/AR9scFnXAH+fdZeVgNXA+yU19CMqIkLSX5LOJRzS4DI/DFybtV7+nXS+YQC4m9Q19oSkF0m/rs+o8vkPAVdLWgR0k365ryCdmN8R90oa3cbTgL+KiG8DSHo/8NeSppN+UJ4SEc8Az0j6LKn7bYQ0qOL4iChXdEUh6UDgIOD3xyzzBmAx6XzLucD1kjaQuiYfG1N3GXAx8HWAiHhe0gnAFyT1kFqXiyLiSeDJ7LzfuiyOfyFtuwOBL2UDAgqkpPW72fy/Ddws6QsR8dHx1msHtqflpODHE5g1RtJFpPMz/5q1UL4PLIiIH7Q5NLNJwS0cs8Y9STo/M0L62/krJxuzxrmFY2ZmufCgATMzy4UTjpmZ5eJleQ6nXC6Xh4aqXVfYWbq6CgwPd36Xp+NsnskQIzjOZpssce6yS9fzpEscdsrLNOFAf/9A/Ypt1tfX4zibaDLEORliBMfZbJMlzlmzen88kc+7S83MzHLhhGNmZrlwwjEzs1w44ZiZWS6ccMzMLBdOOGZmlgsnHDMzy4UTjpmZ5cIJx8zMcvGyvNOANUd3TzcjwMDQMKWe7pdMLwKDA4O5x2VmnckJx3baCHD6iocplboYGhp+yfTlp77tpR8ys5ctd6mZmVkunHDMzCwXTjhmZpYLJxwzM8uFE46ZmeXCCcfMzHLhhGNmZrloyXU4krqAZYCAYWARMBO4A/hhVm1pRNwi6QzgLGAIuCwi7pS0K3AjsDuwFVgYEZskzQWuzOqujohLs+VdAhyTlV8QEetbsV5mZrbzWnXh57EAEXGYpPnAFaRkc0VEfH60kqQ9gPOAtwLTgbWSvgmcAzweEUsknQgsBs4HrgHeDzwN3CVpdjarI4A5wJ7AbYCvODQz6zAtSTgR8XVJd2Zv9wZ+ChwMSNJxpFbOBcAhwLqI2A5sl7QROAA4HLg8+/zdwMWSZgDTIuIp0oxWAUcC20mtnTLwrKSSpFkRsWm8+AoF6Ovrae5Kt0BXV7Gj4xwYGqZU6qIAlEpdL5leLBY6Kv5O354wOWIEx9lskyXOiWrZrW0iYkjSSuAPgD8EfgNYHhGPSLoIuAR4DHih4mNbSV1vMyrKK8u2jKm7L7AN2FxlHuMmnHIZ+vsHdnrd8tLX19PRcZZ6uhnKkk61W9uMjJQ7Kv5O354wOWIEx9lskyXOWbN6J/T5lg4aiIiFwBtJ53NWR8Qj2aTbgbeQEkjlGvQC/WPKq5U1Um5mZh2kJQlH0imSLszeDpDu8/g1SYdkZUcCjwDrgXmSpkuaCewPPAGsA47O6i4A1kTEFmBQ0n6SCsBRwJqs7lGSipL2AooR8Xwr1svMzHZeq7rUvgZcL+l+YBfS+ZqfAF+UNAg8B5wZEVskXUVKHEXgoojYJmkpsFLSWmAQOCmb79nATUAXqcX0EICkNcAD2TzObdE6mZnZBBTK5XK7Y8jdyEi5vHnzi+0Oo65O79ct9XTXfTzBUAc9D6fTtydMjhjBcTbbZIlz1qzeR0ijineKL/w0M7NcOOGYmVkunHDMzCwXTjhmZpYLJxwzM8uFE46ZmeXCCcfMzHLhhGNmZrlo2c07zbqKRejprlmnCAx20MWhZtY6TjjWMiPlMqeveLhmneWn+tFFZi8X7lIzM7NcOOGYmVkunHDMzCwXTjhmZpYLJxwzM8uFE46ZmeXCCcfMzHLhhGNmZrlwwjEzs1w44ZiZWS5acmsbSV3AMkDAMLAIKAArgDLwBHBuRIxIOgM4CxgCLouIOyXtCtwI7A5sBRZGxCZJc4Ers7qrI+LSbHmXAMdk5RdExPpWrJeZme28VrVwjgWIiMOATwFXZP8WR8Q8UvI5TtIewHnAYcBRwGckTQPOAR7P6t4ALM7mew1wEnA4MEfSbEmzgSOAOcCJwNUtWiczM5uAlrRwIuLrku7M3u4N/JTUArkvK7sbeA+p9bMuIrYD2yVtBA4gJZTLK+peLGkGMC0ingKQtAo4EthOau2UgWcllSTNiohN48VXKEBfX08T17g1urqKHR3nwNAwpVIXBaBU6qpaZ7zyUcViIbd17PTtCZMjRnCczTZZ4pyolt0tOiKGJK0E/gD4Q+C9WVKA1E02E5gBvFDxsWrllWVbxtTdF9gGbK4yj3ETTrkM/f0DO7diOerr6+noOEs93QxlSWdoaLhqnfHKR42MlHNbx07fnjA5YgTH2WyTJc5Zs3on9PmWDhqIiIXAG0nnc3atmNQL9JMSSG+d8h2pW1luZmYdpCUJR9Ipki7M3g4AI8A/S5qflS0A1gDrgXmSpkuaCexPGlCwDji6sm5EbAEGJe0nqUA657Mmq3uUpKKkvYBiRDzfivUyM7Od16outa8B10u6H9gFuADYACyT1J29vjUihiVdRUocReCiiNgmaSmwUtJaYJA0UADgbOAmoIt03uYhAElrgAeyeZzbonWyDtTd081InTp+qqhZZ2jVoIGfA39UZdIRVeouI3W5VZYNACdUqfsgMLdK+RJgyc5Fa5PZCPipomaThC/8NDOzXDjhmJlZLpxwzMwsF044ZmaWCyccMzPLRcvuNGDWiK5iEXq6x53uIc1mU4cTjrXVSLlcc1izhzSbTR3uUjMzs1w44ZiZWS6ccMzMLBdOOGZmlgsnHDMzy4UTjpmZ5cLDoq2j1btOp1Ao5BiNmU2EE451tHrX6Vy76JAcozGziXCXmpmZ5cIJx8zMcuGEY2ZmuXDCMTOzXDjhmJlZLpo+Sk3SLsB1wD7ANOAy4N+AO4AfZtWWRsQtks4AzgKGgMsi4k5JuwI3ArsDW4GFEbFJ0lzgyqzu6oi4NFveJcAxWfkFEbG+2etkZmYT14ph0ScDmyPiFEmvBh4F/idwRUR8frSSpD2A84C3AtOBtZK+CZwDPB4RSySdCCwGzgeuAd4PPA3cJWl2NqsjgDnAnsBtgO9nb2bWgVqRcL4K3Frxfgg4GJCk40itnAuAQ4B1EbEd2C5pI3AAcDhwefbZu4GLJc0ApkXEU6QZrQKOBLaTWjtl4FlJJUmzImJTrQALBejr62nO2rZQV1exo+McGBqmVOqiAJRKXVXrjFe+I3UmOr1YLNDX19Px2xM6/zsf5Tiba7LEOVFNTzgR8SKApF5S4llM6lpbHhGPSLoIuAR4DHih4qNbgZnAjIryyrItY+ruC2wDNleZR82EUy5Df//ATqxdvvr6ejo6zlJPN0NZ0hkaGq5aZ7zyHakz0ekjI2X6+wc6fntC53/noxxnc02WOGfN6p3Q51syaEDSnsC9wJcj4mbg9oh4JJt8O/AWUgKpjL4X6B9TXq2skXIzM+swTU84kl4DrAY+ERHXZcWrJI3eg+RI4BFgPTBP0nRJM4H9gSeAdcDRWd0FwJqI2AIMStpPUgE4CliT1T1KUlHSXkAxIp5v9jrZ1Nbd002pzr/uGvdzM7PGtOIczieB3UjnXi7Oyj4G/K2kQeA54MyI2CLpKlLiKAIXRcQ2SUuBlZLWAoPASdk8zgZuArpI520eApC0Bnggm8e5LVgfm+JGoOb92gCWn+qxKGYT1YpzOOeTRpWNdWiVusuAZWPKBoATqtR9EJhbpXwJsGTnojUzs7z4wk8zM8uFE46ZmeXCCcfMzHLhhGNmZrlwwjEzs1w44ZiZWS6ccMzMLBdOOGZmlgsnHDMzy4UTjpmZ5cIJx8zMcuGEY2ZmuXDCMTOzXDjhmJlZLhpKOJIWj3n/mdaEY2ZmU1XN5+FI+mPgdGB/SaNP4ewCdgEubHFsZmY2hdR7ANuNwLdIT/H8dFY2AvyslUGZmdnUU7NLLSK2R8QzpMc7vwbYG/gtYE7rQzMzs6mk0UdM3wrsDvwke18G7m9JRGZmNiU1mnD2iIhDWxqJmZlNaY0mnH+V9LqI+L/1KkraBbgO2AeYBlwG/ABYQWoZPQGcGxEjks4AzgKGgMsi4k5Ju5LOHe0ObAUWRsQmSXOBK7O6qyPi0mx5lwDHZOUXRMT6BtfJzMxy1Oh1OPOAZyX9R/avVuI5GdgcEfOABcAXgSuAxVlZAThO0h7AecBhwFHAZyRNA84BHs/q3gCMDsm+BjgJOByYI2m2pNnAEaRzSicCVze64mZmlq+GWjgR8YYdmOdXSed8Rg0BBwP3Ze/vBt4DDAPrImI7sF3SRuAAUkK5vKLuxZJmANMi4ikASauAI4HtpNZOmZQQS5JmRcSmWgEWCtDX17MDq9QeXV3Fjo5zYGiYUqmLAlAqdVWtM175jtSZ6PRisUBfX8+423N0PRqZR6t1+nc+ynE212SJc6IaSjiSrid1h/1SRJxWrW5EvJh9ppeUeBYDn8uSAqRuspnADOCFio9WK68s2zKm7r7ANmBzlXnUTDjlMvT3D9Sq0hH6+no6Os5STzdD2cF6aGi4ap3xynekzkSnj4yU6e8fGHd7jq5HI/NotU7/zkc5zuaaLHHOmtU7oc832qX2FeAW4B+AjUDNLSNpT+Be4MsRcTPp2p1RvUA/KYH01infkbqV5WZm1mEa7VJbVfH2Hkmrx6sr6TXAauAjEfGtrPhRSfMj4juk8zr3AuuBT0uaThpcsD9pQME64Ohs+gJgTURskTQoaT/gadI5n0tJ3XWXS/oc8JtAMSKeb2zVzcwsT412qb2n4u1rSReBjueTwG6kcy8XZ2XnA1dJ6gY2ALdGxLCkq4A1pJbWRRGxTdJSYKWktcAgaaAApItPbyLdWmd1RDyUxbYGeCCbx7mNrI+ZmeWv0WHRH6x4vQ2oev4GICLOJyWYsY6oUncZsGxM2QBwQpW6DwJzq5QvAZaMF4+ZmXWGRrvUFkn6HeC3gScj4rGWRmVmZlNOo48n+CipJXIo8CVJf97SqMzMbMppdJTaScC8iLiAdKHmB1oWkZmZTUmNJpxCRAwBRMQvgF+0LiQzM5uKGh00sFbSraQRZYeThi6bmZk1rG4LR9KZpKd7Xk+6iv++iPiLVgdmZmZTS82EI2kJ6b5nu0TEXaSbab6r4voam8K6e7op1fhXKBTaHWLHqLetunu62x2iWdvV61JbAMwdvQ9aRDwj6QPAd4G/bHVw1l4jwOkrHh53+rWLDskvmA5Xb1stP/Vt+QVj1qHqdam9WHHTTeCXgwa2ti4kMzObiuolnP+StG9lQfa+PE59MzOzqup1qX0C+Lqkb5FumrkX6caZC1sdmJmZTS01WzgR8S+kp30+CrwC+B5wWEQ8mkNsZmY2hdS9DiciXiCNTjMzM9tpjV74aTZpdRWL0NOdHiVdZXhypwzv7q4RI6TuiMGBwXyDMmsiJxyb8kbKZU5f8fC4j8LulOHdI8DZNzwy7uOuPbTaJrtG76VmZmY2IU44ZmaWCyccMzPLhROOmZnlomWDBiTNAT4bEfMlzQbuAH6YTV4aEbdIOgM4CxgCLouIOyXtCtwI7E66hc7CiNgkaS5wZVZ3dURcmi3nEuCYrPyCiFjfqnWyl6/RkW7j6ZSRbmadrCUJR9LHgVOAn2dFs4ErIuLzFXX2AM4D3gpMJz1z55vAOcDjEbFE0onAYuB84Brg/aQ7HtyVJTGAI4A5wJ7AbYCH8ljTjY50G0+njHQz62St6lJ7Cji+4v3BwDGS7pd0raRe4BBgXURszy4u3QgcQHrA2z3Z5+4G3i1pBjAtIp7Kbia6Cjgyq7s6IsoR8SxQkjSrRetkZmYT0JIWTkTcJmmfiqL1wPKIeETSRcAlwGPACxV1tpIe8DajoryybMuYuvsC24DNVeaxqVZ8hQL09fXs2Eq1QVdXsa1xDgwNUyp11axTKnVRyP4fb3o9jSxjItNH67QzzmKxUPe7HBgarhljI/PIS7v3zUY5zs6S14Wft0dE/+hr4AvA/UBvRZ1eoJ+UWHprlFWWD45TXlO5DP39Azu0Au3Q19fT1jhLPd3jXoQ4aihLSuPVq/f5RupMdPponXbGOTJSrvtdlnq6KdeYTyPzyEu7981GOc7mmjWrt36lGvIapbZK0mgn95HAI6RWzzxJ0yXNBPYHngDWAUdndRcAayJiCzAoaT9JBdIdq9dkdY+SVJS0F1CMiOdzWiczM9sBebVwzgG+KGkQeA44MyK2SLqKlDiKwEURsU3SUmClpLWkFsxJ2TzOBm4CukjnbR4CkLQGeCCbx7k5rY+Zme2gliWciHgGmJu9/h5waJU6y4BlY8oGgBOq1H1wdH5jypcAS5oQspmZtZAv/DQzs1w44ZiZWS6ccMzMLBdOOGZmlgsnHDMzy4UTjpmZ5cIJx8zMcuGEY2ZmuXDCMTOzXDjhmJlZLvK6l5rZy1q9J4aCnxpqU58TjlkO6j0xFPzUUJv63KVmZma5cMIxM7NcOOGYmVkunHDMzCwXTjhmZpYLJxwzM8uFE46ZmeXCCcfMzHLRsgs/Jc0BPhsR8yW9HlgBlIEngHMjYkTSGcBZwBBwWUTcKWlX4EZgd2ArsDAiNkmaC1yZ1V0dEZdmy7kEOCYrvyAi1rdqnczMbOe1pIUj6ePAcmB6VnQFsDgi5gEF4DhJewDnAYcBRwGfkTQNOAd4PKt7A7A4m8c1wEnA4cAcSbMlzQaOAOYAJwJXt2J9zMxs4lrVpfYUcHzF+4OB+7LXdwPvBg4B1kXE9oh4AdgIHEBKKPdU1pU0A5gWEU9FRBlYBRyZ1V0dEeWIeBYoSZrVonUyM7MJaEmXWkTcJmmfiqJCliggdZPNBGYAL1TUqVZeWbZlTN19gW3A5irz2FQrvkIB+vp6dmCN2qOrq9jWOAeGhimVumrWKZW6KGT/jze9nkaWMZHpo3UmQ5y1YiwWCx2z37Z732yU4+wsed28c6TidS/QT0ogvXXK69UdHKe8pnIZ+vsHGg6+Xfr6etoaZ6mnm6Gh4Zp1hrKkNF69ep9vpM5Ep4/WmQxxlmvUGxkpd8x+2+59s1GOs7lmzeqtX6mGvEapPSppfvZ6AbAGWA/MkzRd0kxgf9KAgnXA0ZV1I2ILMChpP0kF0jmfNVndoyQVJe0FFCPi+ZzWyczMdkBeLZw/A5ZJ6gY2ALdGxLCkq0iJowhcFBHbJC0FVkpaS2rBnJTN42zgJqCLdN7mIQBJa4AHsnmcm9P6mJnZDmpZwomIZ4C52esnSaPJxtZZBiwbUzYAnFCl7oOj8xtTvgRY0oSQzcyshXzhp5mZ5cIJx8zMcuFHTL9Mdfd0/9rQwWoKhUIusZjZy4MTzsvUCHD6iodr1rl20SH5BGNN0ciPiCIwODCYRzhmL+GEYzZFNPIjYvmpb8snGLMqfA7HzMxy4YRjZma5cJea2STRVSxCT/e40z3IwzqdE47ZJDFSLtc8R+NBHtbp3KVmZma5cMIxM7NcOOGYmVkunHDMzCwXTjhmZpYLj1Izs19q9PY4ZjvDCcfMfsm3x7FW8o8VMzPLhROOmZnlwgnHzMxy4YRjZma5yHXQgKRHgReytz8CPg2sAMrAE8C5ETEi6QzgLGAIuCwi7pS0K3AjsDuwFVgYEZskzQWuzOqujohL81wnMzNrTG4tHEnTASJifvZvEXAFsDgi5gEF4DhJewDnAYcBRwGfkTQNOAd4PKt7A7A4m/U1wEnA4cAcSbPzWiczM2tcni2cA4EeSauz5X4SOBi4L5t+N/AeYBhYFxHbge2SNgIHkBLK5RV1L5Y0A5gWEU8BSFoFHAl8L59VMjOzRuWZcAaAzwHLgTeQkkYhIsrZ9K3ATGAGv+p2G6+8smzLmLr71gukUIC+vp6dXpG8dHUVWxbnwNAwpVJX3Xr16pRKXRRq1GvWMiYyfbTOZIizVozNWEaxWKi5TzWyXxSLhZbum83kODtLngnnSWBjlmCelLSZ1MIZ1Qv0kxJIb53yenVrKpehv39gJ1YhX319PS2Ls9TTzdDQcN169eoMZQeo8eo1axkTmT5aZzLEWa5Tb6LLGBkp19ynGtkvRkbKDA+PvOz/hpppssQ5a1Zv/Uo15DlK7TTg8wCSXkdqnayWND+bvgBYA6wH5kmaLmkmsD9pQME64OjKuhGxBRiUtJ+kAumcz5qc1sfMzHZAni2ca4EVktaSfsidBjwPLJPUDWwAbo2IYUlXkRJHEbgoIrZJWgqszD4/SBooAHA2cBPQRRql9lCO62Q2qfgx1dZOuSWciKhMEpWOqFJ3GbBsTNkAcEKVug8Cc5sUptmU5sdUWzv5wk8zM8uFE46ZmeXCCcfMzHLhhGNmZrlwwjEzs1w44ZiZWS6ccMzMLBe5Pp7AzKa+7p5uRmpMLwKDA4N5hWMdxAnHzJpqBGpeXLr81LflF4x1FHepmZlZLtzCMbNc1bufG7jbbapywjGzXNW7nxu4222qcsIxsx3SVSymB7WN00rxHadtPE44ZrZDRsplzvryI+M+qM13nLbxeNCAmZnlwgnHzMxy4S41M+s49Uay1RvFNnrxaa1zTR4Jlz8nnCmq3tXePrFrnazeSLZ6o9hGLz4tlbrGPdfkkXD5c8KZoupd7e0Tu2aTw1S6VZATjplNOvW63KZSC34q3SpoSiQcSUXg74ADge3A6RGxsb1RmVmr1Otycwu+M02JhAO8D5geEW+XNBf4PHBce0PaefVOeHYViwyP1GpkT61feGbt0IyurE7oDqsXQ15xwNRJOIcD9wBExIOS3rqzM+qEHaTeCc9rFx1S99Yg/oVnVlu9brlyocDp168fd/r1p80ZdwTcjs5jIndumOh6VMYxnmYd9wrlcnnCM2k3ScuB2yLi7uz9s8C+ETE0zkc2AT/OKz4zsylib2DWzn54qrRwtgC9Fe+LNZINTGCDmZnZzpkqdxpYBxwNkJ3Deby94ZiZ2VhTpYVzO/B7kr4LFIBFbY7HzMzGmBLncMzMrPNNlS41MzPrcE44ZmaWCyccMzPLxVQZNPAS9W53I+lY4FPAEHBdRCxrU5y7ANcB+wDTgMsi4h8rpn8M+GPStUMAZ0VE5B1nFsujwAvZ2x9FxKKKaZ2yPU8FTs3eTgcOAvaIiP5setu3p6Q5wGcjYr6k1wMrgDLwBHBuRIxU1G3bbZvGxHkQ8AVgOIvjwxHx0zH1x90/copxNnAH8MNs8tKIuKWibqdsy68Ae2ST9gEejIgTx9TPdVtWOw4BP6DJ++aUTTjUuN1NtnH/Bngb8HNgnaQ7IuK5NsR5MrA5Ik6R9GrgUeAfK6bPJv1xP9KG2H5J0nSAiJhfZVrHbM+IWEH6I0HS1aTk119Rpa3bU9LHgVNI2wngCmBxRHxH0jWkffT2io+8jzbctqlKnFcCH42IxySdBXwC+FhF/XH3jxxjnA1cERGfH+cj76MDtuVocpG0G3Av8Kdj6ue+Lal+HHqMJu+bU7lL7ddudwNU3u5mf2BjRPxnRAwCa4F5+YcIwFeBiyvej71g9WDgQklrJV2YX1gvcSDQI2m1pG9nO9ioTtqeAGS3N3pzRHxpzKR2b8+ngOPHxHNf9vpu4N1j6tfaj1tpbJwnRsRj2esSsG1M/Vr7R14xHgwcI+l+SddK6h1Tv1O25ahLgS9ExH+MKW/Htqx2HGr6vjmVE84MftUkBRiWVBpn2lZgZl6BVYqIFyNia/bHcSuweEyVrwBnA+8CDpf03rxjzAwAnwOOyuK5qRO3Z4VPkv6gx2rr9oyI24BfVBQVImL02oRq263WftwyY+McPShKOhT4CKlFW6nW/pFLjMB64C8i4h3A08AlYz7SEdsSQNLuwJFkrfEx2rEtqx2Hmr5vTuWEU+t2N2On9QL9OcX1EpL2JDWtvxwRN1eUF4C/jYjns5bDXcBb2hTmk8CNEVGOiCeBzcBrs2mdtj37gDdFxL1jyjtpe46qvFdste22o7dtahlJHwCuAY6JiE1jJtfaP/Jye0VX6e289LvtmG0J/CFwc0RUexxpW7ZlleNQ0/fNqZxwat3uZgPwBkmvktQNvAN4IP8QQdJrgNXAJyLiujGTZwBPSHpldrB8F9CuczmnkfpokfS6LLbRroCO2Z6ZdwD/VKW8k7bnqEclzc9eLwDWjJneEbdtknQyqWUzPyKerlKl1v6Rl1WSRm+TfiQv/W47Yltm3k3qpqom9205znGo6fvmVB408JLb3Ug6CXhlRHwpG620ipR0r4uIf29TnJ8EdgMuljTah7oMeEUW5ydJvzq2A9+KiG+0Kc5rgRWS1pJGrZwG/JGkTtueACJ1qaQ3v/69d8r2HPVnwLIsUW8gdWcg6QZSt0bbb9skqQu4CngW+JokgPsi4pKKOF+yf7Sh9XAO8EVJg8BzwJlZ/B2zLSv82j4KvxZnO7ZltePQ+cBVzdw3fWsbMzPLxVTuUjMzsw7ihGNmZrlwwjEzs1w44ZiZWS6ccMzMLBdOOGYdQNJ3JL2p3XGYtZITjpmZ5cLX4ZjlTNIMYDnQB/w30oW+HwB+lr3fDnwYuBD4PxGxUtIewF0RcXBbgjZrArdwzPL3euArEfEe4L386jb/X4uId5Ge6XIhKREtzKadAlyfd6BmzeSEY5a/54D3SbqRdIuQXbLy+7P/vwsoIjYAJUl7k1pAN+YeqVkTOeGY5e/PgQci4mTSc0gKWfnojSfnkZ6wCOm+WpcDPxjzIDmzSWcq37zTrFPdASyV9CHSreeHSI/1fZ+kC0i3fR/tSvsq6Wmbv9+GOM2ayoMGzMwsF+5SMzOzXDjhmJlZLpxwzMwsF044ZmaWCyccMzPLhROOmZnlwgnHzMxy8f8B6gdIcdybFQoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_ba_abv = region_ba.abv.mean()\n",
    "avg_ba_abv = round(avg_ba_abv, 2)\n",
    "print(f'The average abv% of beers from BeerAdvocate is {avg_ba_abv}%.')\n",
    "fig, ax = plt.subplots()\n",
    "ax.set_xlim(0, 20)\n",
    "sns.histplot(region_ba.abv, binwidth=0.5, ax=ax)\n",
    "plt.title(\"Abv rankings on BeerAdvocate\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:31.653529Z",
     "start_time": "2022-11-18T21:04:30.727461Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average abv% of beers from RateBeer is 6.84%.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_rb_abv = region_rb.abv.mean()\n",
    "avg_rb_abv = round(avg_rb_abv, 2)\n",
    "print(f'The average abv% of beers from RateBeer is {avg_rb_abv}%.')\n",
    "fig, ax = plt.subplots()\n",
    "ax.set_xlim(0, 20)\n",
    "sns.histplot(region_rb.abv, binwidth=0.5, ax=ax)\n",
    "plt.title(\"Abv rankings on RateBeer\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "According to the pictures above we found that the most popular beers has abv range from 4 to 10."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### (Interactive) Find the popular beers in top 10 regions with the largest number of users in each year. \n",
    "**(The interactive picture couldn't load in github page, so we put these image in read.md)**\n",
    "##### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:32.167361Z",
     "start_time": "2022-11-18T21:04:31.654392Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_ba_style_change = region_ba[['date', 'location', 'beer_id', 'beer_name', 'brewery_id', 'brewery_name', 'overall', 'nbr_ratings', 'nbr_reviews', 'text', 'style']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:34.108984Z",
     "start_time": "2022-11-18T21:04:32.168290Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_ba_style_change['formal_date'] = region_ba_style_change.date.apply(lambda x: datetime.datetime.fromtimestamp(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:34.321017Z",
     "start_time": "2022-11-18T21:04:34.109583Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_ba_style_change = region_ba_style_change[['formal_date', 'date', 'location', 'beer_id', 'beer_name', 'brewery_id', 'brewery_name', 'overall', 'nbr_ratings', 'nbr_reviews', 'text', 'style']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:34.372274Z",
     "start_time": "2022-11-18T21:04:34.321857Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                          3267567\n",
       "mean     2013-02-26 08:27:11.016104192\n",
       "min                1996-08-22 12:00:00\n",
       "25%                2010-09-23 12:00:00\n",
       "50%                2014-12-30 12:00:00\n",
       "75%                2015-11-24 12:00:00\n",
       "max                2017-08-01 12:00:00\n",
       "Name: formal_date, dtype: object"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_ba_style_change.formal_date.describe(datetime_is_numeric=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:37.765809Z",
     "start_time": "2022-11-18T21:04:34.373078Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_ba_style_change.sort_values('formal_date', inplace=True)\n",
    "region_ba_style_change['year'] = region_ba_style_change.formal_date.apply(lambda x: x.year)\n",
    "region_list = region_agg.location.to_list()\n",
    "year_list = region_ba_style_change.year.unique()\n",
    "region_list = region_list[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:37.769379Z",
     "start_time": "2022-11-18T21:04:37.766996Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# since there may have insufficient in early years, we want to make an alert in the interactive graph\n",
    "def popup(text):\n",
    "    display(HTML(\"<script>alert('{}');</script>\".format(text)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:38.006704Z",
     "start_time": "2022-11-18T21:04:37.772701Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8672596a82524627bf45ab9e571d654e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='region', options=('United States, Pennsylvania', 'United States, C…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.plot_interactive(region, year)>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def plot_interactive(region, year):\n",
    "    data_to_plot = region_ba_style_change[(region_ba_style_change.location==region) & (region_ba_style_change.year==year)][['brewery_name', 'style', 'beer_name','overall']]\n",
    "    data_to_plot = data_to_plot.groupby('style').mean().reset_index()\n",
    "    data_to_plot = data_to_plot.sort_values('overall', ascending=False).head(10)\n",
    "    if len(data_to_plot) > 0:\n",
    "        sns.barplot(y='style', x='overall', data=data_to_plot, palette=\"magma\")\n",
    "        plt.title(f'10 beer styles in {region} that are the most popular in year {year}.')\n",
    "        plt.xlabel('Overall score')\n",
    "        plt.ylabel('Beer style')\n",
    "        plt.show()\n",
    "    else:\n",
    "        popup(f\"Insufficient data for {region} in {year}.\")\n",
    "\n",
    "\n",
    "interact(plot_interactive, region=region_list, year=year_list[:])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### RateBeer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:38.923453Z",
     "start_time": "2022-11-18T21:04:38.007469Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_rb_style_change = region_rb[['date', 'location', 'beer_id', 'beer_name', 'brewery_id', 'brewery_name', 'overall', 'nbr_ratings', 'text', 'style']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:40.698653Z",
     "start_time": "2022-11-18T21:04:38.924553Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_rb_style_change['formal_date'] = region_rb_style_change.date.apply(lambda x: datetime.datetime.fromtimestamp(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:40.834104Z",
     "start_time": "2022-11-18T21:04:40.699571Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_rb_style_change = region_rb_style_change[['formal_date', 'date', 'location', 'beer_id', 'beer_name', 'brewery_id', 'brewery_name', 'overall', 'nbr_ratings', 'text', 'style']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:40.885046Z",
     "start_time": "2022-11-18T21:04:40.834858Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                          3414540\n",
       "mean     2011-03-08 04:16:46.752535296\n",
       "min                2000-04-12 12:00:00\n",
       "25%                2008-03-21 12:00:00\n",
       "50%                2011-10-01 12:00:00\n",
       "75%                2014-05-15 12:00:00\n",
       "max                2017-07-31 12:00:00\n",
       "Name: formal_date, dtype: object"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_rb_style_change.formal_date.describe(datetime_is_numeric=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:43.822826Z",
     "start_time": "2022-11-18T21:04:40.885830Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "region_rb_style_change.sort_values('formal_date', inplace=True)\n",
    "region_rb_style_change['year'] = region_rb_style_change.formal_date.apply(lambda x: x.year)\n",
    "region_list = region_agg.location.to_list()\n",
    "year_list = region_rb_style_change.year.unique()\n",
    "region_list = region_list[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:44.057563Z",
     "start_time": "2022-11-18T21:04:43.828967Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65d7c1b477f74bd5abd12eb9ed86408b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='region', options=('United States, Pennsylvania', 'United States, C…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.plot_interactive(region, year)>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def plot_interactive(region, year):\n",
    "    data_to_plot = region_rb_style_change[(region_rb_style_change.location==region) & (region_rb_style_change.year==year)][['brewery_name', 'style', 'beer_name','overall']]\n",
    "    data_to_plot = data_to_plot.groupby('style').mean().reset_index()\n",
    "    data_to_plot = data_to_plot.sort_values('overall', ascending=False).head(10)\n",
    "    if len(data_to_plot) > 0:\n",
    "        sns.barplot(y='style', x='overall', data=data_to_plot, palette=\"magma\")\n",
    "        plt.title(f'10 beer styles in {region} that are the most popular in year {year}.')\n",
    "        plt.xlabel('Overall score')\n",
    "        plt.ylabel('Beer style')\n",
    "        plt.show()\n",
    "    else:\n",
    "        popup(f\"Insufficient data for {region} in {year}.\")\n",
    "\n",
    "\n",
    "interact(plot_interactive, region=region_list, year=year_list[:])\n",
    "plt.read('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Find the change of number of ratings\n",
    "##### (interactive) Of each style of beers\n",
    "###### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:44.274349Z",
     "start_time": "2022-11-18T21:04:44.058263Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "beer_style_list = region_ba_style_change['style'].groupby(region_ba_style_change['style']).agg('count')\n",
    "beer_style_list.sort_values(ascending=False)\n",
    "beer_style_list = np.array(beer_style_list.index.to_list())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:44.495541Z",
     "start_time": "2022-11-18T21:04:44.275116Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "71cce784d6904febba20e259f5a8282d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='style', options=('Altbier', 'American Adjunct Lager', 'American Am…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.plot_year_vs_no_ratings_interactive(style)>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def plot_year_vs_no_ratings_interactive(style):\n",
    "    specific_style = style\n",
    "    specific_style_region_by = region_ba_style_change[region_ba_style_change['style']==specific_style]\n",
    "    specific_style_region_by = specific_style_region_by.dropna(how='any')\n",
    "    specific_style_region_by = specific_style_region_by.drop_duplicates('year')\n",
    "    sns.lineplot(data=specific_style_region_by, x='year', y='nbr_ratings')\n",
    "    plt.title(f'The number of ratings of {style}')\n",
    "    plt.xlabel('Year')\n",
    "    plt.ylabel('Number of ratings')\n",
    "    plt.show()\n",
    "\n",
    "interact(plot_year_vs_no_ratings_interactive, style=beer_style_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "###### RateBeer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:44.708794Z",
     "start_time": "2022-11-18T21:04:44.496895Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "beer_style_list_rb = region_rb_style_change['style'].groupby(region_rb_style_change['style']).agg('count')\n",
    "beer_style_list_rb.sort_values(ascending=False)\n",
    "beer_style_list_rb = np.array(beer_style_list_rb.index.to_list())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:45.127898Z",
     "start_time": "2022-11-18T21:04:44.709624Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0f58f33ccb2d46e3880fade42d6660ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='style', options=('Abbey Dubbel', 'Abbey Tripel', 'Abt/Quadrupel', …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.plot_year_vs_no_ratings_interactive(style)>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def plot_year_vs_no_ratings_interactive(style):\n",
    "    specific_style = style\n",
    "    specific_style_region_by = region_ba_style_change[region_rb_style_change['style']==specific_style]\n",
    "    specific_style_region_by = specific_style_region_by.dropna(how='any')\n",
    "    specific_style_region_by = specific_style_region_by.drop_duplicates('year')\n",
    "    sns.lineplot(data=specific_style_region_by, x='year', y='nbr_ratings')\n",
    "    plt.title(f'The number of ratings of {style}')\n",
    "    plt.xlabel('Year')\n",
    "    plt.ylabel('Number of ratings')\n",
    "    plt.show()\n",
    "\n",
    "interact(plot_year_vs_no_ratings_interactive, style=beer_style_list_rb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### Total number of ratings on the websites and overall scores\n",
    "###### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:48.420065Z",
     "start_time": "2022-11-18T21:04:45.129310Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEgCAYAAACegPWEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAA74ElEQVR4nO3dd5xcVfn48c/MbO/JZjc9IWX3SSgBQq+hdwUVBFGkCCKgFMGvUhREFEFA8ad0pWNDQEApCoqAIL2EhCeFFNI3u9lsma2z8/vj3AnDuGV2Zqfs5nm/Xnll9+4995w7e/c+95xzzzm+cDiMMcYYkyh/pgtgjDFmeLNAYowxJikWSIwxxiTFAokxxpikWCAxxhiTFAskxhhjkmKBxKSdiGwjImER+VrM9ktE5J4hzGe5iOw6VMcbIK8yEXlZRD4Qkc/HmeYuETkk1WVLNxHZTURu877eVUQeznSZTGrlZLoAZqvVA9woIi+pqma6MENgJ2Csqs6MN4Gqnpm64mTUdsAkAFV9Azg+s8UxqWaBxGRKG3Aj8JCI7KWqndE/9Gom81X1htjvRWQ58BBwEDAKuB7YB9gF6AI+q6prvEOdJyI7AvnAjar6W+94nwGuAPKAIHCJqr4iIlcBewETgHdV9Ssx5ToOuBJXm28Gvg1sBn4LTBSRd4C9VLXN278W+A8wQVU7RSQArAQOAW4FfqWqD4vI3sB1QDEQAn4IPAWs8463REQuBb6hqlO9Y//D+wwLvXPp8dJ+R1X/HfuBi8j3gS8B3cAi4JtAWT/lWwPcDOwA5ALPecfuFpEO4C/AjsCXvYCBiEwGrgbKReRu4F7vHLf3fodB73hjgceBeuAzwDjgTFV9XkTyvM9iHhAA3gbOV9Wm2HMy2cGatkwm/RhoAX6SQNoCVd0T+AFwB3Czqu4IfAycFrVfm6rOBQ4FrhWR7USkxsvzKFXdGfg68IiIFHtppgI79xJEZgG3AV/w8voB7ma6FjgTWKqqO0WCCICqLgI+AD7rbToMWKaqC6OOOwq4GzjFK+uxuCAzEXgCOMLb9QggT0RqRaQcdxN/DvgZcK6q7gp8Hzgg9sMSkdOBI4HdVHUOMB+4Z4Dy/Rx4U1V3AXYGxuACJ7gA/ISqSiSIeOf7sfe5vKiqp8eWA5iLewDYH7gYaFHVvXEB63vePt/DBbtdvM95DfDTXo5lsoQFEpMxqtoDfAU4XUQOHWTyP3v/LwXWqeq7Ud+Pjtrvdi+vNcCzwMG4oDIeeM6rQTyIe5qPNEu9qqrdveR5EPCcqn7kHfN5YAOuJtSfu/gkuJ0O3Bnz87288jzmledvQBiYAzwKHCkipbin9oe88h8FPO3V5H4PPCoid/FJDS3WkcDdqtrqfX8zcLD39N9X+Y4BzvbK9CawO642EfHiAOfdmydUtUtV1wGtwNPe9ujf2zG4YPq2l/dxwLYJ5GXSxJq2TEap6scicjauCeS+qB+FAV/U93kxSTuivu7qJ4tQ1Nd+b98cXEA4MfIDr0lmDfA5XC2pNwGvXNH8uGafzv/dfYs/ATeJyGxcc81pvRx3oaruEVWeCUCd97O7gKOBfwF/B87BNRH9HkBVLxeR3+ICzGm4J/3dByi7H/c5+PopXwA4IVJ7EpGKmGP09Tn1pyPm+95+dwHgAlV9ysu3BChIIC+TJlYjMRmnqg/j+gMujNpcB+wKW26q8xI8/GneMabg2v2f8/4d5jVVISJHAe/h+hr68xxwuIhM99IdBEwG/ttfIlVtx9307wH+rKrBmF1eBWpEZH/vuDsBi4GJXtoXcP0yz3pf7wXsBzwjIjlen1GRqt4GnAvMEZH8mDyeBs6Iar47H/i3qnb0U75ngItExOcd73Fcv8pAunHBNVHPAN8UkTwR8eNqSNcmcTyTYhZITLY4H1gR9f3/A8aLiOL6D55P8LgFIvIWrrnoW6q6SFUX4PpFfi8i7wI/wnXQ9/uE7aU7F9efMh/Xbv8ZVd0cRznuxNUS7urluHXAF4CfeeW5H9dfstzb5VGgFnje6395F3hZVdu9JrgLcS8tvIWrXZyhqrFP/r8B/gG8JiILcX0VXx6gfOfjOv/fxwXa9+m92SzWq8B0EXkkjn178yNgOa6TfQGu1nRxgscyaeCzaeSNMcYkw2okxhhjkmKBxBhjTFIskBhjjEmKBRJjjDFJsUBijDEmKVvlgMSenp5wKDR83lYLBHwkWt5E01qelqflaXnGys0NbASqYrdvlYEkFArT2Bg7Jix7VVQUJVzeRNNanpan5Wl5xqqqKl3R23Zr2jLGGJMUCyTGGGOSYoHEGGNMUiyQGGOMSYoFEmOMMUmxQGKMMSYpFkiMMcYkxQKJMcakWUtHN5c+sZD1Te2ZLsqQsEBijDFptmBdM/9YVMdzH27IdFGGhAUSY4xJs4agW6p+4brmDJdkaFggMcaYNGsIdgKwcG1ThksyNCyQGGNMmtW3uhqJrm8m1DN8JpDtiwUSY4xJs0iNpL2rh5Wb2jJcmuRZIDHGmDSrb+2kMNfdfhdtaMlwaZJngcQYY9KsIdjFnAll5AZ8LKqzQGKMMWaQGoKdVJfkU1NdyqINrZkuTtK2yoWtjDEmU3rCYRqCXVQW55GXn8PzCzcQDofx+XyZLlrCrEZijDFp1NTeTagnzOjiPGaPK2VTWxf1rZ2ZLlZSLJAYY0waRd7YqizKZdvxZQDoMG/eskBijDFp1OCNIRldlMescaUAw77D3QKJMcakUaQZa3RxLqUFuUwsLxj2rwBbIDHGmDSq95q2RhflAVBbXcKiOmvaMsYYE6eGYBc5fh9lBe6l2dqqYj7e1EZrZ3eGS5Y4CyTGGJNGDa2djC7Kxe+97ivVJYSBJcO4VmKBxBhj0qgh2LWlWQtc0xYM7ze3UjIgUURygXuBbYAQcBbQDdwDhIH5wHmq2iMiZwFnez+/RlWfFJFC4AGgGmgGTlXVOhHZE7jZ2/dZVf2hl9+VwNHe9gtV9bVUnJcxxiSrIdjJ6OLcLd9Xl+RRXpAzrN/cSlWN5CggR1X3Bq4GfgzcBFyhqvsBPuBYERkHnA/sAxwOXCsi+cA5wPvevvcBV3jHvQ04GdgX2ENE5orIXGAesAdwEvDrFJ2TMcYkrb6181M1Ep/P5zrch/GbW6kKJIuAHBHxA2VAF7AL8IL386eAQ4DdgZdVtUNVNwNLgDm4QPF09L4iUgbkq+pSVQ0DzwAHe/s+q6phVV3p5VuVovMyxpiEhb3pUaIDCUBtVQlLN7bSPUzXJknVXFstuGatD4ExwDHA/l4AANdcVY4LMpuj0vW2PXpbU8y+04F2oL6XY9T1VbhAwEdFRVECp5UZgYA/4fImmtbytDwtz6FP1xjspLsnzKQxxVRUFG1Ju/M2o3nwzVU0dPVQO7Y0ZWVNNm1fUhVILgKeUdVLRWQy8DwQHYJLgUZcYCgdYPtA+3b2sb1PoVCYxsZg/GeTYRUVRQmXN9G0lqflaXkOfbpl9W6/Ih80Nga3pJ1U4vpM3li6ker8QMrKmmzaqqreg1yqmrY28UmNogHIBd4WkQO8bUcCLwKvAfuJSIGIlAOzcR3xL+P6Wbbsq6pNQKeIzBARH65P5UVv38NFxC8iUwC/qm5M0XkZY0zCIvNsRXe2A0wdXURewDdsp5RPVY3k58BvReRFXE3kMuAN4E4RyQMWAg+rakhEfokLCH7gclVtF5FbgXtF5CVcjeNk77jfAB4EArh+kf8CePm84h3jvBSdkzHGJGXL9CgxfSQ5fh8zxhQP2ze3UhJIVLUF+GIvP5rXy753AnfGbAsCJ/Sy76vAnr1svwq4KrHSGmNMejQE3YSNlTGBBNx4kn8t3jgs1yaxAYnGGJMm9a2dBHxQVvi/z/C1VSVsbu9mfXNHBkqWHAskxhiTJg3BTkYV5W2ZHiWaVBcDDMsJHC2QGGNMmkSW2O3NzKpifDAsByZaIDHGmDSp9yZs7E1xXg6TRxVajcQYY0zfGoJdjO6jRgJuSnmrkRhjjOmVmx6lk8o+aiTg3txavbmdlo7htTaJBRJjjEmD5o5uukLh/xlDEq22yk0pP9zGk1ggMcaYNGhodWNIYke1R9vy5tYwG+FugcQYY9Igdq323lQW5zG6KHfY9ZNYIDHGmDTYMqq9n852n89HbVXJsHtzywKJMcakQYM3z1Z/ne0AtdXFfFTfSleoJx3FGhIWSIwxJg0agm56lPLCAQJJVQldofCWKeeHAwskxhiTBvXBLir6mB4lmlQPvze3LJAYY0wa9DeqPdrkUYXk5/iH1ZtbFkiMMSYNGoJdvU4fHyvg91FTNbzWJrFAYowxadDQ2kllP2NIotVWlbBoQyvhcDjFpRoaFkiMMSbFItOj9DeGJFptdTHNHd2sbRoea5NYIDHGmBRr7QzRGQr3O2FjtC1TpQyTgYkWSIwxJsU+Was9vqatmqpi/L7h8+aWBRJjjEmxyPQo8XS2AxTkBpgyqnDYvLllgcQYY1IsngkbY7mpUqxGYowxBjeqHfqfZytWbXUJa5s62NzWlapiDRkLJMYYk2L1wS78PigvGESNxJtSfvEwmMDRAokxxqRYQ2snFYW5BPz9T48SbTgtcmWBxBhjUqwh2DWoZi1wzWBjivOGxSvAFkiMMSbF3GDE+Ju1Imqri4fF2iQWSIwxJsXchI2Dq5GAa976qD5IZ3d2r01igcQYY1LITY/SlVggqS4h1BPmo/rsrpVYIDHGmBRq7QzR0d0T94SN0Wqr3Jtb2T4w0QKJMcakUDxrtfdl8qhCCnP9Wf/mlgUSY4xJoYZBzrMVze/zUVNVkvVvblkgMcaYFIqMak+kjwRc89aiulZ6snhtEgskxhiTQvXByDxbCQaS6hJaO0Os2dw+lMUaUhZIjDEmhepbO/EBFYWDb9oCF0ggu9cmsUBijDEp1BDsZFRRLjmDmB4l2ozKIgI+0CwemGiBxBhjUqihNbExJBEFuQGmji7K6hpJTqoOLCKXAp8F8oBbgBeAe4AwMB84T1V7ROQs4GygG7hGVZ8UkULgAaAaaAZOVdU6EdkTuNnb91lV/aGX15XA0d72C1X1tVSdlzHGDEai06NEq60u4a2PG4emQCkQV41ERPwiEhCR/URkwNAqIgcAewP7APOAycBNwBWquh/gA44VkXHA+d5+hwPXikg+cA7wvrfvfcAV3qFvA04G9gX2EJG5IjLXy2MP4CTg13GduTHGpEF9sCvhjvaI2qpiNrR00hjMzrVJBgwkInIdcBbwI+By4M44jns48D7wKPAE8CSwC65WAvAUcAiwO/Cyqnao6mZgCTAHFyiejt5XRMqAfFVdqqph4BngYG/fZ1U1rKorgRwRqYqjjMYYk1LhcJiG1qGpkQBolg5MjKdGsq+q3g7spapHAJPiSDMG2BU4AfgG8CDg9wIAuOaqcqAM2ByVrrft0duaBtg3ersxxmRUsCtEe3dP3Gu190WqsvvNrXj6SAIisjuw3GvWiudpvx74UFU7ARWRdlzzVkQp0IgLDKUDbB9o384+tvd9QgEfFRVFcZxGdggE/AmXN9G0lqflaXkmn26zN9nipKqSXveJN8+KChhXVsDyxnYqKooycp79iSeQ3Av8P+AM4HpcZ/dAXgIuEJGbgPFAMfCciBygqv8CjgT+CbwG/FhECoB8YDauI/5l4Cjv50cCL6pqk4h0isgM4CNc89kPcR3s14vIDbjakl9VN/ZXuFAoTGNjMI7TyA4VFUUJlzfRtJan5Wl5Jp9u+TrXiFLo6/2eM5g8Z44pYv7qzTQ2BjNyngBVVaW9bo8nkGwCDlbVFuDCeDLz3rzaHxcI/MB5wDLgTq9WsxB4WFVDIvJL4EVvv8tVtV1EbgXuFZGXcDWOk71DR5rJArh+kf8CiMiLwCtReRljTMZtGdWeZNMWuH6SV5Y10N4VSvpYQy2eQDID+JuINAKPAI+rasNAiVT1/3rZPK+X/e4kpgNfVYO4/pXYfV8F9uxl+1XAVQOVyRhj0ikyYWNlkp3tAFJVTCgMS+uDjOujZpApA3a2q+qPVXV/4GrceI91KS+VMcaMAA1Bb3qUIaqRQHZ2uA9YIxGRn+NqARuBh4BTU10oY4wZCRqCXZQXJj49SrQJ5QUU5wWyMpDE8/pvIdAGrARWAGtSWiJjjBkh6odgDEmE3+fbMqV8tomnaesbqnoQbnqTi3Cv9hpjjBlAfWtXQisj9qW2uoTFdS309GTX2iTxNG1dDBwBFAF/xU1fYowxZgANwU62Hz90HeO1VSW0dfWwoiHIqJzkm8uGSjxvbXUDp6vqqlQXxhhjRpKGYOcQ10iKAVi4tom9J2fPBB7xBJLngN+JSAVuDMd8VX0ypaUyxphhrq0rRFtXz5CMIYmYXllMwO9j4brmrAok8XS23wycjntr6zfYeA1jjBlQfWtkrfah6WwHyMvxM72yiAVrmwbeOY3imkZeVZcAYVWtw02KaIwxph9bAskQNm2Bm1L+w3XZdRuOJ5A0iMjZQLGInMQAEyIaY4xxY0hgaEa1R6upKmFDc8eWQJUN4gkkXwOm4Zq2dvW+N8YY04+GoDc9ylDXSLwO98VZtDZJn4FERCLrjowDfosbQ3IHbq0RY4wx/WhodTWSUYVDXyMBWJxFAxP7e2vr296/22O2h4GDUlYiY4wZAeqDnZQX5JATiKsrOm4VhbmMKytAs2iqlD4Diap+2/vyZtyMvz3pKZIxxgx/DUOwVntfZo8vzaqpUuIJlYcA74rIj0VkeqoLZIwxI0FDa+eQd7RHzB5XxsqGYNasTRLPXFvfBHYB3gF+JSL/SHWhjDFmuKsPdg7pYMRos8eXEgrDR/XZsdJrvI13u+OWth2LG+lujDGmHw1DPGFjtNnj3Pxd2TKlfDyTNi4A3gXuUtUzU18kY4wZ3tq7QgS7QkM6qj3a5FFFFOUGsubNrXjm2tpPVW3qeGOMiVN9MDWj2iP8fh81VcUsypKxJPH0kVgQMcaYQYiMIalMUR8JQE1VMYvrWukJZ35tkv4GJGbP1JLGGDOMNGypkaSmaQvcIletnSHWbG5PWR7x6q9G8gSAiNyaprIYY8yI8MnMv6mrkdRWuxHu2TCepL8+kjYReR2oEZEdvW0+3CzAe6e+aMYYMzzVexM2pqqzHWBGZRF+n3tz66CazM5c1V8gORKYgJsi5RxcEDHGGDOAhlY3PUruEE+PEq0gN8DUUUVZ8Qpwn2epqj3e8rrHAscA/wccB6xNT9GMMWZ4agh2pbRZK6K2ujgrXgGOJ1zeDswE/g5sA9yVygIZY8xw1xDsTGlHe0RtVQnrmjvY3NaV8rz6E884khpV3d/7+jER+U8qC2SMMcNdQ7AL8TrDU6lmy9okrew6pSLl+fUlnhpJgYgUAYhIIRBIbZGMMWZ4q2/tTGlHe0RtVeTNrcz2k8RTI7kZN/vvfGBb4MrUFskYY4av9q4QrZ2hlM2zFa2yOI/K4ryMvwI8YCBR1QdF5ClgOrDMRrobY0zfPlmrPfWBBNwI90y/uRVPjQRVbQAaUlwWY4wZ9tIxqj1abVUJD61cRVeoJ6WvG/cnM7kaY8wIVd8aGYyYnhqJVBfT3RNmWQbXJhkwkIjIJekoiDHGjARbaiRp6GwHqMmCDvd4aiRHiYi9qWWMMXH4JJCkp0YyZVQh+Tl+Fm3IXId7PH0kY4A1IrIMCGNzbRljTJ/qW7sozc8hLyc9PQcBv4+ZY4pZnMEaSTyB5DMpL4UxxowQDcH0jCGJVltdzHOLNhIOh/H50j8tYjwhsxv4MXAHcBgwLqUlMsaYYayhtTMtY0ii1VSV0NTezfrmjrTmGxFPjeQO4Ebg+8C/gXuBPQdKJCLVwJvAobhgdA+uaWw+cJ6q9ojIWcDZ3s+vUdUnvdHzDwDVQDNwqqrWicieuMGR3cCzqvpDL58rgaO97Req6mtxnrsxxgy5+mDXlhHn6VJb5aZKWVTXyriygrTmDXFOkaKqz+P6RhQYcDkuEcnFTfbY5m26CbhCVffDTUd/rIiMA84H9gEOB64VkXzclPXve/veB1zhHeM24GRgX2APEZkrInOBecAewEnAr+M4H2OMSZmGYCeVaRpDEjGzqhgfZGxgYjyBpENEDgcCXq0gnnUdb8Dd+Nd43+8CvOB9/RRwCLA78LKqdqjqZmAJMAcXKJ6O3ldEyoB8VV2qqmHgGeBgb99nVTWsqiuBHBGpiqN8xhgz5Dq6e2jpCKXtja2I4rwcJlUUZGyqlHiatr6OCwxjgEtwNYY+ichpQJ2qPiMil3qbfV4AANdcVQ6UAZujkva2PXpbU8y+03FBrb6XY9T1V8ZAwEdFRVF/u2SVQMCfcHkTTWt5Wp6W5+DTrWl0jTCTxhTHdayhPM/tJpazYG3zgMdL5jPqSzxzba0SkZ8AtcB8VV02QJIzgLCIHALshGueqo76eSnQiAsMpQNsH2jfzj629ysUCtPYmLlRoINVUVGUcHkTTWt5Wp6W5+DTLVvrnncLfMR1rKE8z2kVhTz9wXpWrW+iJL/vW3syn1FVVWmv2+MZ2X4FcAuuL+M3InJhf/ur6v6qOk9VDwDeAb4KPCUiB3i7HAm8CLwG7CciBSJSDszGdcS/DBwVva+qNgGdIjJDRHy4PpUXvX0PFxG/iEwB/Kq6caBzMsaYVIis1Z7ut7bATd4IsCQDzVvxNG0dBezrvWWVA7wE/GKQ+VwM3CkiecBC4GFVDYnIL3EBwQ9crqrtInIrcK+IvISrcZzsHeMbwIO49VCeVdX/AojIi8Ar3jHOG2S5jDFmyDS0ulHtlWkeRwJQW/3JVCk7TSpPa97xBJINQBHQAuQxQP9DNK9WEjGvl5/fCdwZsy0InNDLvq/Sy2vHqnoVcFW8ZTLGmFSJTCE/Ks2d7QDVJXmUF+RkpMO9z0AiIq/gxn1UA4tF5F3cwla2HokxxvSiIdhJSX6A/DRNjxLN5/NRW12SkVeA+6uRnJS2UhhjzAhQ39qV9ld/o9VUFfPwO2vo7gmT40/fVCl9BhJVXQEgIrvjgkr0cMlzU1wuY4wZduqDnRnpH4mQ6hI6Q2FWbgoyvbI4bfnG00dyL3AdsCnFZTHGmGGtobWTmVXpu4HHikzNsmhDa9YFksWqek+qC2KMMcNdQ7ArbWu192ab0YXkBnws2tDCEbOrB04wROIJJH8Wkd8DCyIbVPXq1BXJGGOGn87uHpo7utO2VntvcgJ+plcWszjNb27F82rBucDbwPqof8YYY6Kke2XEvtRWFaMbWgiHwwPvPETiqZE0qOp1KS+JMcYMY5ExJJkOJDXVJTzxwXrqWzsZU5KfljzjCSQbReR24C3cuBJU9Y6UlsoYY4aZSI0k3VPIx4qsTaJ1rWkLJPE0bS3BTQc/Dhjv/TPGGBOlvjVbmrbcm1uL0zgwMZ4ayd0pL4UxxgxznzRtZbZGUlqQw4Sy/LROlRJPIPkDrknLD0wDFuMWlDLGGOOpb+2kOC9AQW4g00Whpiq9U6XEsx7JXpGvRaQCt4SuMcaYKA3BroxMH9+b2upi/r20nrauEIVpCGyDnVlsMzAjFQUxxpjhrCHYmfFmrYjaqhLCwNKN6WneGrBGEjULsA+oAv6R6kIZY8xw09DaxbTK7FjCu6bavbm1aEML248vS3l+8fSRRM8C3K6qNiDRGGNi1Ac72WVyeheU6suEsgKK8wJp63Dvbz2Sr/axHVW9L3VFMsaY4aUr1ENTe3fW9JF8sjZJ5pu2Zsd87wNOB4KABRJjjPFsefU3SwIJuIGJj89fR6gnTCDFa5P0tx7JpZGvRWQmcA/wJHBhSktkjDHDzJZR7VnS2Q6uw72tq4dVjW1MHZ3avpt4OtvPwwWPi1T1yZSWxhhjhqGG1uyYZytardfhvriuNXOBREQm4ka1NwC7q6otbGWMMb2oj8z8m+F5tqJNqywm4PexqK6FQ6QqpXn1VyOZD3QCzwO/FpEtP1DVk1NaKmOMGUYaWiNNW9lTI8nP8bPN6MK0dLj3F0iOS3nuxhgzAtQHuyjKzY7pUaLVVpXw5seNKc+nv872F1KeuzHGjAANrZ0Znz6+N7XVJTy1cAObgp2MSmFtabBTpBhjjInhpkfJnmatiBpvbZJUD0y0QGKMMUmqD3Zl1RiSiMgiV6lew90CiTEmK4TDYf70zhqOuv1Vnpq/LtPFGZSG1uyZsDHaqKI8qkvyUj6lvAUSY0zGrW/u4Ft/fp/rn1tCsDPE/z3yHgvXN2e6WHHpCvWwub07q97YilZTVcKiOgskxpgRKhwO87cF6znp3jd4d3UTlx4ykz+fsRuji/K45LEP2NjSkekiDmjLErtZ2NkObmDi8vogHd09KcvDAokxJiM2BTv57hMLufIpZUZlMb87dRc+v+MEKovzuO3Lc2nu6ObivyygvSuU6aL2q94LdtlaI6mtKiEUhmX1qesnsUBijEm7F5bUc9K9b/LSR/V8a79p3H7ijkyqKNzy89njy7j6yFksWNfMNc8uIhwOZ7C0/dvYEqmRZGkgqS4BSOnAxHjWIzHGmCHR0tHNTf9cyhMfrKemqphfHz+Hmd6bRbEOqBnDuftuwy0vLWd6ZTFn7DklzaWNT6T5LRs72wEmVRRQmOtPaT+JBRJjTFq8sbKRHz6tbGjp4Iw9JnPmXlPJDfTfKHLa7pP5qD7IrS8vZ1plEQfWjElTaeMXqZFky1oksfw+HzPHlKR0LIk1bRljUqq9K8SN/1zKOX96j7wcP3edtBPn7DttwCACboGmKw6rZfvxpfzgbx+iKX6NNRH1rR0U5vopzLLpUaLVVhezaENLypoILZAYY1Lmg3XNnPLAW/z+rdV8cacJPHjKXHaYMLg1xPNz/Pzs2O0oK8jh4sc+YKP3llS22NiSnaPao9VWFdPaGWJNU3tKjm+BxBgz5LpDPfziucV87aG3CXaG+NXxO/Cdg2cmPKnhmOI8bjpueza3dfF/f/kgpa+yDtbGlo7sDyReh/viFHW4D3kfiYjkAr8FtgHygWuABbgVFsO46enPU9UeETkLOBvoBq5R1SdFpBB4AKgGmoFTVbVORPYEbvb2fVZVf+jldyVwtLf9QlV9bajPyRgTn5aObh6fv44/vr2G1ZvbOXrbai4+cCalBcnfamRsCT88UvjuEwv5yd8XcdURgs+X2iVk47GxpZOJZfmZLka/Zo4pxu8jZR3uqehs/wpQr6qniEgl8DbwDnCFqv5LRG4DjhWRV4DzgV2BAuAlEfk7cA7wvqpeJSInAVcAFwC3AV8APgL+KiJzvfzmAXsAk4E/A7ul4JyMMf1YuamNP7y1mic/WE+wK8SOE8q48jPbsvPYkiHN56DaKs7eO8jt/1nB9MpiTt198pAePxEbWzqYM74008XoV0FugMkVqVubJBWB5E/Aw1HfdwO7AJFp6Z8CDgNCwMuq2gF0iMgSYA6wL3B91L7fF5EyIF9VlwKIyDPAwUAHrnYSBlaKSI6IVKlqXQrOyxgTJRwO898Vm/j9W2t4eVkDOX4fh82q4qS5E5k9tpSKiiIaG4NDnu/X9pzCsvogv35xGduMLmLezMohzyNe3T1hGtu6svbV32i11SV8sLYpJcce8kCiqi0AIlKKCyhXADd4N3twzVXlQBmwOSppb9ujtzXF7DsdaAfqezlGv4EkEPBRUZHaNYyHUiDgT7i8iaa1PC3PvtK2dYZ47N3V3PfKSpbUtVBZnMe3DpzBl3abQlVpfp/phrK8N564Eyf/5r9c+dSH/OGsPZFxpSnPszcbmtsJh2HimJK0/l4SSbfjlFH8Xeto7QwN+f0vJeNIRGQy8Chwi6o+JCLXR/24FGjEBYbSAbYPtG9nH9v7FQqFU/KklCrJPNklmtbytDxj065raudP76zhsffX0dTezazqEq46QjhUqsjL8UMo9Kk8Un2e1x0zm1MffJuz7n+De768M6OL8tL+2S7zXkcu9JHW30si6SaXuhcClmxoZmpJYi8HVFX13oQ35G9tichY4Fngu6r6W2/z2yJygPf1kcCLwGvAfiJSICLlwGxcR/zLwFHR+6pqE9ApIjNExAcc7h3jZeBwEfGLyBTAr6obh/qcjNlahcNh3lixie89sYDj7nqNB95YxW5TKrjzxB257ys7c/R2Y10QyYCqknxuOHY7GoJdfPfxBXRm4E2u+i1rtWd/09YeU0dx5RG1zBo3uNev45GKGsllwChc38b3vW0XAL8UkTxgIfCwqoZE5Je4gOAHLlfVdhG5FbhXRF7C1ThO9o7xDeBBIIDrF/kvgIi8CLziHeO8FJyPMVud7p4wzy+q44E3VrFwfQtlBTl8eddJnLDTBMaVFWS6eFtsO66UK48QLntyIT/9x2JuPHGntObfEMzuUe3RAn4fx2w3jvwcP21DfOxU9JFcgAscseb1su+dwJ0x24LACb3s+yqwZy/brwKuSqy0xpho7V0h/rpgPfe/vorVm9uZOqqQqz+7LQdsMyprR24fKlUsq2/lzldWMv7vizhpzvghed04Hg2tXQBZP44k1WyuLWMMze3dPPzuGn7/1moagl1sN66UC+ZNZ97MSkaPKs76PsUz95rKqsZ27nhxGfe+soLDZ1Xx+Tnj2XZcaUrHmtQHOynMDVCUl51BNl0skBizFatr6eChN1fz6Htrae0Msdc2ozh198nMnVSeFYP94uX3+bj6qFl8fd4M7n15Gc98uIHH569Hqkv4/JxxHD67muK8obvdhcNhVjS08eH6FioT7LgeSSyQGLMVWt4Q5IHXV/G3hesJ9YQ5VKo4ZbfJSPXQDiBMt+0nlnP5YbVcMG86Ty/cwCPvreXafyzh5heWccTsaj6/4/iEz3Fjayevr9zEaysaeW3FJjZ4s/4eu+OEoTyFYckCiTFbkQ/WNnHv66v41+KN5OX4OXb7cXx510mfWlRqJCjJz+H4nSbwhR3HM39tM4+8t5a/LljPI++tZfvxpXxuzngOk6p+5/4KdoZ4e9VmXvOCx5KNblR4eUEOu06pYPepo9h9SgXbb1OZ9U1/qWaBxJgRLhwO8+LiOm755xLe+Hgzpfk5nL7nFE7cecKI7yT2+XzsMKGMHSaUcdEB0/nrgg08+u5afvTMIn7+r6Ucve1YPjdnPDPGFNMd6uG9NU28tmITr61s5P01TXT3hMkL+NhpYjnf3G8au0+toLaqhIB/+DT7pYMFEmNGsKb2Li5/8kNeXbGJ6pI8Lpw3nePmjBvS/oLhoqwgly/NnchJO0/g7dWbeeTdtTzy3lr+8PYaaqqKWdvUQUtHNz5g1tgSTt5lErtPrWDHCWUJz1q8tdj6riZjthIrN7Vx0aPzWbO5ne8fPZsjayrjWkxqpPP5fMydVMHcSRVsCnby5Afr+ffSeo6ZM56dxpWwy+QKKgqzf4BhNrFAYswI9PrKTXzviYX4gFtOmMOB24/f6tvxezOqKI9TdpvMKbtNTtkkk1sDCyTGjDCPvLuG659bwpTRRdx03HYjriPdZB8LJMaMEN09YX7xr6X84e017DNtNNccPYuSfPsTN6lnV5kxI0BzezeXPbmQV1ds4uRdJnL+/tPtzSKTNhZIjBnmPt7Uxrcfm8/Hje1ccVgNx+4wPtNFMlsZCyTGDGNvrGzke08sAODXx+/ALpMrMlsgs1WyQGLMMPXoe2u57rklTKko5KbPWae6yRwLJMYMM909YX75wkf87q3V7LnNKK49ZrZ1qpuMsqvPmGGkub2Lbz86n1eWb+KkuRO5YN50cqxT3WSYBRJjholVjW1c8vibLK8PcumhNXx+jnWqm+xggcSYLNTdE2ZFQ5DFda0srmthUV0r769pIifg51df2IFdp1RkuojGbGGBxJgMa2rvYnFdK4vqWlm8oYXFda18VN9KZygMQI7fx7TKIg6sGcOFh9ZSHrCmLJNdLJAYk0aNbV28smod76xo2BI01jV3bPn5qMJcaqqKOWGnidRWF1NTVcw2o4u2TLZo80GZbGSBxJg0WNXYxkNvrubx+evo6O7B74Opo4qYM6GML1QVU1NdQm1VMWOK84bVErfGgAUSY1Jqwbpm7n99Fc8vrsPv83Hk7GpO2Wca4wsCtsaFGTEskBgzxMLhMK8s38T9r3/MGx9vpjgvwFd2ncRJcydSVZJvzVNmxLFAYswQ6Q718KzWcf/rq1iysZWqkjzO338an5sz3gYMmhHNrm5jktTa2c1f3l/HQ2+uZn1zB9Mqi/jB4bUcMbvaViQ0WwULJMYkqK65gztfWsbD76yluaObnSeV871DZrL3tNH4rcPcbEUskBgTp+6eMMvrgyxc38xbqzbzzIcb6A6FObBmDKfsNontx5dluojGZIQFEmN60dndw9L6Vj5c34JuaOHD9S0s2dhKR3cPAIW5fo6fO4njdxjHlFE2667ZulkgMVu99q4Qi+ta+XBDCx+ub+bD9S0srQ8S6nEjy4vzAswaW8LxO05g1tgSZlWXMHlUIZWji+3tK2OwQGK2EqGeMBtaOljV2MbqxnZWbW5ndWMbKxrbWVrXghczKC/IYfbYUvaaNppZ1SXMGlvChPIC6/Mwph8WSMyI0d4VYvXmdlY1trN6c9un/l/b1E6XN3cVQMDvY0JZPjOqS5k3fTSzxpYg1SWMLc23keXGDJIFEjNsLW8I8u8l9byyvIGVje1siJqzClyT1KSKQmqqijlgZiUTKwqZVF7ApIpCxpbmE/D7bHCgMUPAAokZNkI9Yd5f08S/l9bzwtJ6Vm5qA6Cmqph9Z46huiiHSeWFTKwoYFJ5IeWFOVa7MCYNLJCYrNbWFeK/yzfxwtJ6Xvqogca2LnL8PnaZXM6JO09k/xmjGVdWYDULYzLIAonJOhtbO3nJq3W8vrKRju4eSvID7DNtNPvPqGTvaaNtyhFjsoj9NZqM6+juYUldC/PfW8cz89cyf20zYWB8WT7H7TCOeTMr2XliOTk23YgxWWlEBBIR8QO3ADsCHcCZqroks6Uyvens7mHJxlYWrm9m4foWFq5r/tSYjdljS/j63lOZN7OSmWOKrY/DmGFgRAQS4DigQFX3EpE9gRuBYzNbJNMV6mHpxlYWrHcD/Rauc6PDu72gUV6Qw6yxJZwybRKzx5awz6yx5Pf0ZLjUxpjBGimBZF/gaQBVfVVEdk1FJuFwmLqWTnrC4YF3HkJBfDQ1tSeUts3nY/MAaXs7m6YeaGgM0t0TJhQK0x0O0x3qobsn/Mm/UJhQT8+nvm/v7mFVcwfvrNzEko2tW8ZulOa7oHHyLpPYdpw30K+s4FM1joqyAuswN2YYGimBpAzYHPV9SERyVLV7KDP50ztr+NnzS4fykCNSaUEOUl3CSTtPZPa4UmaPLWFieYE1UxkzQo2UQNIElEZ97+8viAQCbiDaYB2/+1RGlaV/gj6/30dPz+BrQeFwGH/ATziOtLH3+JyAHz+QG/AT8PvICfjI9fvICfjJ8f6P/j434CPg95GX42dsWSHhBGptgYA/od9LMmktT8vT8kzeSAkkLwOfAf7o9ZG839/OoVA4oSYUP3DojNEJFTAZyYyRSDRtwnmGegiHE/t8h9V5Wp6W51aWJ0BVVWmv20dKIHkUOFRE/gP4gNMzXB5jjNlqjIhAoqo9wDcyXQ5jjNka2QgvY4wxSbFAYowxJikWSIwxxiTFAokxxpikWCAxxhiTFAskxhhjkuJLZATyCFAHrMh0IYwxZpiZClTFbtxaA4kxxpghYk1bxhhjkmKBxBhjTFIskBhjjEmKBRJjjDFJsUBizDAmIrZamEnKUFxDFkhM3BK94JK5UNN9oxSRQAJpfCIyLsH8EpqBW0T8IrKTqmbktctMXAuJsmuoz3RDdg3Z67/DmIj4Er0IRMTvTb/f7z7AWcASYJ2qfjCYPEXkdFW9ezBl9fK8EmgHXlHVf8WTV1Tas4HlwEeqqnGm8wG3AOeralc8n01Mnr8FnlHV3w2yrDcCi4G/qOrqQaT1AY8DJcBBgC+e8np5nufluSzezycq7ZnAx8AaVX03nt9pstdQIuwairu8g76G+mI1kmHCe3q4W0QuE5GLAAZxQ/eLyG0icomI/MRL2+NdiH2l8QEPALsBOwP3icghqhqO5wlPRLYBzo4u60DpvJ/fCxQDq4BfisiEOM8xUt6dgd2BUyNPhgPl632OXwYeFJH8yB9UHOX1A38AXsCtzlkuIlMHSuv97EFgPW41z5zI02iceT4I9ACrVDUcZxDxAfcDc4G9gW9GnmTj/L08COwATAbuF5EDB/qdDsE1dIeInDnQfr3kadfQwHkO+hrqjwWS4eMBYAPwBrB99B9HfwHB8yCwEXgYGCMif4UtwaSvi25PoERVz1TVG4CrgN+IyAFxBrCdgFxgpoj82MtvoBvIvkCpql6iqvcDHwCF3jkOdOPZCxilql/HrZi5PXCUiMzrL9+oz+7PQB7wiIjsLSLbx3GenwP2AJ7BfbY/Bf4pIkcMkHYW7o/4TuB7wIXAv0Xk8Djy/BmwSFWPBTZ7S0vH8/kcgPt8TgduBSYAe4vIPnH8Xg4AylX1W6p6By4g3SEiewxQ3j1I7hrKAW4UkXPj2DcimWtoTxK7hiLbE7mGjiOxa0hI/zXUJwskw4CI5OKWEP6dqj6LezI8U0Tuhy0rRPaVNt9Le4+qLgd+AIwVkdu9tH1ddAqsEpFtRSRHVZ8ALgIuFpExcRS7CbgL+AUwWkSujuTXzwVbB3woImXe9/l8sorn6AHyWw285T1lHwjMwE3n8JSIbNvXeXrBNAC0qepxQBvwL9yTd79BWlX/DNwNPA38XVXPAc4HbhWRGf2UdRnu8zkL+LOqXgRcAtwuIjV9JfLK8jtVvcrb5MM9OcdTO10MrBSRi4ErgGm4J++/i8icAdIvB+pEZLL3/cvAE8C3RGR0P7/Pj4EVIjJ7sNeQ96DUDJzg7f/NAc4vohX4QEQii4sP5hraBLzuXQ9xX0O4ByaAYALX0CO4v5OnGNw1tB5oxDU3DuYaCgCPJXgN9ckCSZbzqsldwDvAGSLyDNCCeyosj9yg+0nbgfvj2lVEKoGxwEPez/eO2d8vIieKyClemg7gFGCUuDbfR3BzlHX1klck7WkAqvo8cLeqLsY9NVWIyI3ez8K9pPsy7gb7/1S1SUQKcH/EdSLyReA2ESmIvml5ab8oIl8B1gLXqmo3rga2o6r+ChfIPtX5GZXnqd55hYAFXhlKgWeBy6KbKHpJe7p3g3gS+CPwnLi2/ydxT5YdvaT7ooh8FfckuRHXNl0qIrmq+jjwl//9LW5J+yXgJOCtqB/dCnxGRHbrJ92J3uezAXfDysE1be2nqjcDN+M9sfeTdi3ud36ZiFwPXIu7/hYBLTG/T5+IHOZ9uwEIAV/D/f4Huoai064H/uE9OB0LXNJXzSQm3fvArara7D1ETaH/ayg67RLgZ9718AAwp59raEs6Ve2MpBeRkxn4GorOE1xt5DFcUO/vGopO14wLfEcAJXFcQz6vlhNS1RejfvRr+rmG4mWBJEt5f8g3ADeIyPG4m9VtQCfwbVVdhqtKbx4g7WdxTztfAK7BXThPAUtxT2uRND7ck/XuwGeAG3A3yWnAxcBh3h/JbkBBTH7RaQ8XkftjgtR7uBtPj4hU9ZHuWNyNcar3x9eOu1F9GTgd+IGqtkduWlFp9/DKewcwx3vy3QhM9gLiwbg/uN7yPALXbr8PMBu4DrhSVY8B3gU+9RZN7HniAmQBLjAvBXbxPqN9cDfQ2HR7AMcANwH/xN1QBThNRE7FBZa2PvLc1TvPu8U1mxSq6nzcDWcf7yGhr7J+FtcRnAs8j2vumeJ9PofgAkV/aa/zzvFvwL+BnwDlwDygjE/bBnhaRE7yHoAuxT28fJd+rqGYtCd6N/NIE+x83O/qpyJy1gDpulR1lXdD7qCfa6iXtN2q2uzlWR/1GX3qGopJd0LUZzYF1/nd5zUUk/aL3vev44L8SmBub9dQb2UFfgQsBLbD9ef0eg1Fpf1bpLxemQOquoA+rqHBsECSve4DgrgL7HhcFbYbF0gOENdEcQbuj7uvtL8BTsZVsX8CXI67KeyIu6FFz4C8PbBRVS9W1S/ibi4H4/4wVgL7AV8Evqqq62Pyi077JdxT4VeAGgDvon8TuEJV6/rJ88PodEAtcCpwgaou7CfPE4EFuNrTLBE5FPgW8FWvvKsGKOtncR20R6vqq16Zv6mqsTNER6c9CddcdCKuY/do3BtRXwVOV9W1faT7IrAGOAy4HXgNmIm7oZ8QU9beyvuBd56RZo/luJpb7BN+bJ6LcddCGHdTPMM7zldUdeUAadd5n9FHuIcCwT1ofMsL2tEm4Dq5rxWRM1W1DdeEtx5XE+rrGopO+1Nxb/xFOqxzVfVDXDD95wDpToNP1Xpn0vc11GtaL8/PAufS+zUUne56ETnDy+8y4PMDXEPRaa/z0vao6sfA54Fz6P0aii3rmV6gvAgXsKbT9zUUW97TvfJFAtUSer+G4maBJAuJSDWuPfc3qvou8GPcRVaD64wDmAh8PfaPIybtO7hmiK/jmnoacG2iRwDnqupHUUlXA9MiT0mq+hNcE8xFqnqLql4OnKza6+uQsWmvxzVrXBPZwftj6Ygz3U+8n//Vy3NRnHmuB76vqn9X1W/j/qhjyxub7jpcgP4/77Pur9MxNu1PgXrgclX9i7qO7C/EkedPvDwvVtV7VPW7wFnezTLe84y8ffeMd85NcZznZtzrqd9S1e8Ax8f5+7zWS3u1F3RuAw73nmZjdeCCzuHApSLyNa92eZOq/oC+r6HYtFdEBYUur6a5SFWXDJDu+5F0XtPjP+j7GuozLfBX7zPq7RqKTXd5JGiq6itei0B/HdexaSN5/lFVz6T3ayg2XeSz7cA15V0GnNnHNRSb9orooKmqz9H7NRQ3G0eSpUTkcmAU7g2LAtxNuQZ3w3l/kGl/5KU9V1Xf9v4ou72L/ULc0+oS3NPbPsALqvq0d6xHgEtU9SOJev8/zrQPA99R1wzHINI96p3np552B1veQaZ72Eu3PIE8H8UFhS2f0SDK+h1VXZrgZ/tdVV2awOdzcfTvJJm0UekUWKFunMhYVV3vtbvfD9ysqrdG9u/lPPtL+wtVva2PsvaX7peqegsxEk0bZ7qfq+rtyebZyzWUts82URZIsoR8MrioA9dc0QZU42oP4Jq3TgLeVdX/JJD2ROC9SFrvQnsC17bfhaupdAHzcW3Yinsy/T5wmFebIZm0g0x3uLp26nTmmYnzHEl5FgDzo29OIrIXrl/uQFXd3E+e/aYFmqJurvGmOwjYHHNzHXR5Ey1rOvNM5rONTpsoa9rKHo/g3pT6G65J6WBcZ/dxuI7OA3Ft8LHtpvGm/WZM2slAnapeAFyNe9ujGNeW+iCuPfsIXFttA5+WaNrBpKvn09KRZybOcyTl+UfcCwenwZYn5FeAvXu5WQ0qrX7S5zGYdI366Y71RMubaFnTlmcyny1DIKE5WkxKbAJ+rKptItKK+8M9G9fHsQbXR3JCbJPEYNN6tZejcB3ZRSIyXlXXisi7uNcW98MNejwN8OsnrzaSaFrLc6vJ8x0v3f4iUo4bKwNRr7EmmtbyTF2eQ8FqJBkmIhd6X5YBv/K+7sGNqQjjRto+hXtDJrZjfVBpvSrvY7hAcyRusNeTIjLRezL5J64vZYa6VyGjbzoJpbU8t6o8m3CvF88ExkSezPXTzTyDTmt5pi5PhojVSDJIREr4ZM6j44FnRORuPnll8Su4Uewf8r+DkxJJewGuyvtNcSNcf4Z7e+h5ce+g1wAVuAGPsRJNa3lunXkGk8gzNq3lmbo8h4QFkszaDfeO/h7Adap6mBcYPo8bZHQMbnxDb08PiaRdDlSKSCFQCeykqgd5Vd/P4gZUna//+/56MmktT8vT8szePIeENW1l1iLciOOTcaNo71A3eK8D1z79Ne19QFOiaV8C7lA3SKwbKPK2t+LeyjlVVT/oI79E01qelqflmb15Dgl7/TfDRKRIVYPiJrG7EQip6hkS33ohyaQtxY0veR5XLT4/3gst0bSWp+VpeWZvnsmwpq0MU9Wg9/9GEbkEuEZExqnqulSmxc2VdD6uaeyr6iZXjFeiaS1Py9PyzN48E2ZNW1lE3TxU58QZCJJN2wD8nsQutETTWp6Wp+WZvXkmLhwO27+t9F9tbW1eutNanpan5Zm9eSb6z/pIjDHGJMWatowxxiTFAokxxpikWCAxxhiTFAskxhhjkmKBxJg0EZGHReR7Ud+XiIiKyI6ZLJcxybJAYkz6fAM4R0S29b6/ATetxbsZLJMxSbPXf41JIxH5DHCp9+8K4AzgDtwqdu3A11X1YxG5FtgVt47EQlU9XUSuwi0yVYKbS21hL1kYk3ZWIzEmjVT1CdzU/vfgFoy6Abde94He1z8VkTJgk6oeigsce4rIRO8QC1V1bwsiJpvYXFvGpN99QJGqrhaRHYDLROS7uLXSO4E2oFpEfodbD6QEyPXSaiYKbEx/rEZiTGZ9CHxXVQ/ALY/8MG6Fu8mq+iXgMqAQF2TArYBpTFaxGokxmXUJcKuIFOACxgW4pZK/LyKv4taX+QiYkLkiGtM/62w3xhiTFGvaMsYYkxQLJMYYY5JigcQYY0xSLJAYY4xJigUSY4wxSbFAYowxJikWSIwxxiTFAokxxpik/H8wqzwSBAX8eQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(region_ba_style_change.groupby(['year']).count()['formal_date'].index, region_ba_style_change.groupby(['year']).count()['formal_date'])\n",
    "plt.ylabel('Number of views')\n",
    "plt.xlabel('Year')\n",
    "plt.title('Number of views over time')\n",
    "plt.xticks([i for i in range(1998, 2018)], rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "According the image above, the number of views are increase over time. In 2014, it has the largest number of reviews. But after that, the number of views are keep decreasing. It can be noted that the number of evaluations suddenly doubled before 2013 and 2014, but it is not possible to analyse what caused this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:48.426513Z",
     "start_time": "2022-11-18T21:04:48.421262Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# The bootstrapping function to use\n",
    "def bootstrapping(array, num_of_iters: int, sample_rate: float, confidence_level: float, value='mean', replace=True,test=False):\n",
    "    \"\"\"array: takes in an array of your concern(numpy array)\n",
    "        num_of_iters: the number of iterations you want to bootstrap\n",
    "        sample_rate: the percentage of data to be sampled each time\n",
    "        confidence_level: the confidence interval, e.g. 95%(input as 0.95)\n",
    "        value: which value to calculate(be bootstrapped) mean / median\n",
    "        replace: takes a boolean, if we will replace or not\n",
    "\n",
    "        output: e.g. 95% confidence intervals of the mean\n",
    "\n",
    "        Author: ZHANG Yuyao\n",
    "        \"\"\"\n",
    "    if test:\n",
    "        # array = np.array(array)\n",
    "        if value=='mean':\n",
    "            true_value = np.mean(array)\n",
    "        elif value == 'median':\n",
    "            true_value = np.median(array)\n",
    "        print(f'The true value of {value} is {true_value}')\n",
    "\n",
    "    sample_value = []\n",
    "    data_len = len(array)\n",
    "    sample_len = int(data_len*sample_rate)\n",
    "    for i in range(num_of_iters):\n",
    "        y = np.random.choice(array.tolist(), sample_len, replace=replace)\n",
    "        if value=='mean':\n",
    "            value_ = np.mean(y)\n",
    "        elif value == 'median':\n",
    "            value_ = np.median(y)\n",
    "        sample_value.append(value_)\n",
    "\n",
    "    bs_value = np.mean(sample_value)\n",
    "\n",
    "    alpha = 1-confidence_level\n",
    "    sample_value = np.sort(sample_value)\n",
    "\n",
    "    left = np.percentile(sample_value, alpha*50)#(100/2)\n",
    "    right = np.percentile(sample_value, 100-alpha*50)\n",
    "\n",
    "    # print(f'The {value} is {bs_value}.\\nAnd the {confidence_level} confidence interval for that is: ({left}, {right}).')\n",
    "    return bs_value, left, right"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:48.429836Z",
     "start_time": "2022-11-18T21:04:48.427699Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def get_stats_num_by_time(x):\n",
    "    bs = bootstrapping(x['overall'], 1000, 0.1, 0.95)\n",
    "#     print(bs[1])\n",
    "    return pd.Series({'mean': bs[0],\n",
    "                  'left':bs[1],\n",
    "                  'right':bs[2]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:04:48.432687Z",
     "start_time": "2022-11-18T21:04:48.430544Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def get_stats_num_by_time(x):\n",
    "    bs = bootstrapping(x['overall'], 1000, 0.1, 0.95)\n",
    "#     print(bs[1])\n",
    "    return pd.Series({'mean': bs[0],\n",
    "                  'left':bs[1],\n",
    "                  'right':bs[2]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:07:16.260817Z",
     "start_time": "2022-11-18T21:04:48.433434Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "stats_to_plot_ba = region_ba_style_change.groupby(['year']).apply(lambda x: get_stats_num_by_time(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:07:16.352689Z",
     "start_time": "2022-11-18T21:07:16.262162Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.errorbar(stats_to_plot_ba.index, stats_to_plot_ba['mean'],\n",
    "             yerr = [stats_to_plot_ba['mean']-stats_to_plot_ba.left,\n",
    "                     stats_to_plot_ba.right-stats_to_plot_ba['mean']],\n",
    "             capsize=3)\n",
    "# plt.plot(region_ba_style_change.groupby(['year']).mean()['overall'].index, region_ba_style_change.groupby(['year']).mean()['overall'])\n",
    "plt.ylabel('Overal Score')\n",
    "plt.xlabel('Year')\n",
    "plt.title('Overall Score over time')\n",
    "plt.xticks([i for i in range(1998, 2018)], rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "It is easy to notice it has very high overall score in 1999. But after analysis the data, we notice it only have few reviews in 1999, so it make a illusion that sudden rise in ratings."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "###### RateBeer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:07:20.121973Z",
     "start_time": "2022-11-18T21:07:16.353709Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(region_rb_style_change.groupby(['year']).count()['formal_date'].index, region_rb_style_change.groupby(['year']).count()['formal_date'])\n",
    "plt.ylabel('Number of views')\n",
    "plt.xlabel('Year')\n",
    "plt.title('Number of views over time')\n",
    "plt.xticks([i for i in range(1998, 2018)], rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "The number of views are keep increase until 2016."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.137998Z",
     "start_time": "2022-11-18T21:07:20.122671Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "stats_to_plot_rb = region_rb_style_change.groupby(['year']).apply(lambda x: get_stats_num_by_time(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.225311Z",
     "start_time": "2022-11-18T21:09:52.139401Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.errorbar(stats_to_plot_rb.index, stats_to_plot_rb['mean'],\n",
    "             yerr = [stats_to_plot_rb['mean']-stats_to_plot_rb.left,\n",
    "                     stats_to_plot_rb.right-stats_to_plot_rb['mean']],\n",
    "             capsize=3)\n",
    "# plt.plot(region_rb_style_change.groupby(['year']).mean()['overall'].index, region_rb_style_change.groupby(['year']).mean()['overall'])\n",
    "plt.ylabel('Overal Score')\n",
    "plt.xlabel('Year')\n",
    "plt.title('Overall Score over time')\n",
    "plt.xticks([i for i in range(1998, 2018)], rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "The overall score are keep increasing. In 2017 it was almost 2 points higher than the year of 2000."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Relation between the overall and 4 scores\n",
    "##### BeerAdvocate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### RateBeer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Hypothesis test about change in overall score\n",
    "##### RateBeer\n",
    "We observed that in the RateBeer dataset, the overall score has increased over the years.\n",
    "Therefore, we want to conduct a hypothesis test about whether it really changed.\n",
    "\n",
    "First, we split the data by the year 2011. (by experiment we find that the number of reviews are relatively close before and after this time point."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### Hypothesis test about change in overall score\n",
    "##### RateBeer\n",
    "We observed that in the RateBeer dataset, the overall score has increased over the years.\n",
    "Therefore, we want to conduct a hypothesis test about whether it really changed.\n",
    "$\\\\$\n",
    "First, we split the data by the year 2011. (by experiment we find that the number of reviews are relatively close before and after this time point."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.419244Z",
     "start_time": "2022-11-18T21:09:52.226129Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of reviews before 2011 is 1770869.\n"
     ]
    }
   ],
   "source": [
    "region_rb_style_change_before2011 = region_rb_style_change[region_rb_style_change.year<=2011]['overall']\n",
    "print(f'The number of reviews before 2011 is {len(region_rb_style_change_before2011)}.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.580459Z",
     "start_time": "2022-11-18T21:09:52.420217Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of reviews after 2011 is 1643671.\n"
     ]
    }
   ],
   "source": [
    "region_rb_style_change_after2011 = region_rb_style_change[region_rb_style_change.year>2011]['overall']\n",
    "print(f'The number of reviews after 2011 is {len(region_rb_style_change_after2011)}.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.583566Z",
     "start_time": "2022-11-18T21:09:52.581527Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# we sample 1643671 data from before 2011 to make the dataset with the same\n",
    "sample_before2011 = region_rb_style_change_before2011[:len(region_rb_style_change_after2011)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.595095Z",
     "start_time": "2022-11-18T21:09:52.584735Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "hypo_test = stats.ttest_rel(sample_before2011, region_rb_style_change_after2011)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:52.597584Z",
     "start_time": "2022-11-18T21:09:52.595778Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p value is 0.0 here so we can reject the null hypothesis that the mean of the splited data are the same, so the overall score has really increased over the time.\n"
     ]
    }
   ],
   "source": [
    "print(f'p value is {hypo_test.pvalue} here so we can reject the null hypothesis that the mean of the splited data are the same, so the overall score has really increased over the time.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "##### BeerAdvocate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-11-18T21:09:53.288276Z",
     "start_time": "2022-11-18T21:09:52.598399Z"
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of reviews before 2011 is 1637859.\n",
      "The number of reviews after 2011 is 1629708.\n",
      "p value is 0.0 here so we can reject the null hypothesis that the mean of the splited data are the same, so the overall score has really increased over the time.\n"
     ]
    }
   ],
   "source": [
    "region_ba_style_change_before2011 = region_ba_style_change[region_ba_style_change.year<=2014]['overall']\n",
    "print(f'The number of reviews before 2011 is {len(region_ba_style_change_before2011)}.')\n",
    "region_ba_style_change_after2011 = region_ba_style_change[region_ba_style_change.year>2014]['overall']\n",
    "print(f'The number of reviews after 2011 is {len(region_ba_style_change_after2011)}.')\n",
    "sample_before2011_ba = region_ba_style_change_before2011[:len(region_ba_style_change_after2011)]\n",
    "hypo_test_ba = stats.ttest_rel(sample_before2011_ba, region_ba_style_change_after2011)\n",
    "print(f'p value is {hypo_test_ba.pvalue} here so we can reject the null hypothesis that the mean of the splited data are the same, so the overall score has really increased over the time.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Some reasons:\n",
    "\n",
    "\n",
    "In the past, fewer people have access to the Internet, so as the rating websites. Therefore, the users may be some experts in beer or have great passion for beers. In this case, they might be more strict while rating the beers.\n",
    "However, now we all can get online to rate the beers, so there exists user groups with higher diversity, which may leads to more relaxed standard, thus the score is higher."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python [conda env:py39]",
   "language": "python",
   "name": "conda-env-py39-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}