pandas_dict_list_columns.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "careful-internet",
   "metadata": {},
   "source": [
    "# handle dict and list columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "alive-program",
   "metadata": {},
   "source": [
    "---\n",
    "* author:  [Prasert Kanawattanachai](prasert.k@chula.ac.th)\n",
    "* YouTube: https://www.youtube.com/prasertcbs\n",
    "* github: https://github.com/prasertcbs/pandas/blob/main/pandas_dict_list_columns.ipynb\n",
    "* [Chulalongkorn Business School](https://www.cbs.chula.ac.th/en/)\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "immune-gardening",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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\n",
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"720\"\n",
       "            height=\"405\"\n",
       "            src=\"https://www.youtube.com/embed/NfM8VXB2d0U\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.YouTubeVideo at 0x1920b868848>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import YouTubeVideo\n",
    "YouTubeVideo('NfM8VXB2d0U', width=720, height=405)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "difficult-brazil",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import json\n",
    "import ast # handle single/double quote in dict/json string\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "governing-forestry",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Python  version: 3.7.9 (default, Aug 31 2020, 17:10:11) [MSC v.1916 64 bit (AMD64)]\n",
      "pandas version: 1.2.1\n",
      "2021-02-21 14:00:08.214632\n"
     ]
    }
   ],
   "source": [
    "print(f'Python  version: {sys.version}')\n",
    "print(f'pandas version: {pd.__version__}')\n",
    "print(pd.Timestamp.now())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "handmade-basics",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']  \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']  \n",
       "2   [120,5,50]              hot                    ['hot']  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_csv('https://github.com/prasertcbs/basic-dataset/raw/master/dict_list_column.tsv', sep='\\t')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "based-distributor",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "str"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(df['price'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "shared-barcelona",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "menu      : <class 'str'> mocha\n",
      "price     : <class 'str'> {\"S\":40, \"M\":50, \"L\":60}\n",
      "price2    : <class 'str'> {'S':40, 'M':50, 'L':60}\n",
      "orders    : <class 'str'> 70,10,100\n",
      "orders2   : <class 'str'> [70,10,100]\n",
      "serve     : <class 'str'> hot,cold,frappe\n",
      "serve2    : <class 'str'> ['hot', 'cold', 'frappe']\n"
     ]
    }
   ],
   "source": [
    "for c in df.columns:\n",
    "    print(f'{c:10}: {type(df[c][0])} {df[c][0]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "logical-hartford",
   "metadata": {},
   "source": [
    "## json.loads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "latest-illinois",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'dict'>\n",
      "{'S': 40, 'M': 50, 'L': 60}\n"
     ]
    }
   ],
   "source": [
    "j=json.loads('{\"S\":40, \"M\":50, \"L\":60}')\n",
    "print(type(j))\n",
    "print(j)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "periodic-burns",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "j['M']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "collected-mattress",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    {'S': 40, 'M': 50, 'L': 60}\n",
       "1    {'S': 45, 'M': 60, 'L': 70}\n",
       "2    {'S': 39, 'M': 49, 'L': 59}\n",
       "Name: price, dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price'].map(json.loads)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "powered-mechanics",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['price_j']=df['price'].map(json.loads)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "latin-anime",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "surface-boost",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_k']=df['price'].apply(lambda s: json.loads(s))\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "literary-behalf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# json.loads(\"{'S':40, 'M':50, 'L':60}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "political-syntax",
   "metadata": {},
   "source": [
    "## ast.literal_eval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "unusual-clearance",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'dict'>\n",
      "{'S': 40, 'M': 50, 'L': 60}\n"
     ]
    }
   ],
   "source": [
    "x=ast.literal_eval('{\"S\":40, \"M\":50, \"L\":60}')\n",
    "print(type(x))\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ordered-design",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'dict'>\n",
      "{'S': 40, 'M': 50, 'L': 60}\n"
     ]
    }
   ],
   "source": [
    "k=ast.literal_eval(\"{'S':40, 'M':50, 'L':60}\")\n",
    "print(type(k))\n",
    "print(k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "removed-production",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_j2']=df['price2'].map(ast.literal_eval)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "joined-alfred",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_k']=df['price'].apply(lambda s: ast.literal_eval(s))\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "imported-business",
   "metadata": {},
   "source": [
    "### access dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "automatic-catholic",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    {'S': 40, 'M': 50, 'L': 60}\n",
       "1    {'S': 45, 'M': 60, 'L': 70}\n",
       "2    {'S': 39, 'M': 49, 'L': 59}\n",
       "Name: price_j, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_j']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "unsigned-feature",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'S': 45, 'M': 60, 'L': 70}"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_j'][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "further-irish",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'S': 45, 'M': 60, 'L': 70}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[1, 'price_j']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "compact-algebra",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "60"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_j'][1]['M']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "respected-german",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    50\n",
       "1    60\n",
       "2    49\n",
       "Name: price_j, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_j'].apply(lambda v: v['M'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "equipped-conflict",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['price_medium']=df['price_j'].apply(lambda v: v['M'])\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "homeless-record",
   "metadata": {},
   "source": [
    "## list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "downtown-startup",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    [70, 10, 100]\n",
       "1     [40, 20, 80]\n",
       "2     [120, 5, 50]\n",
       "Name: orders, dtype: object"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['orders'].str.split(',')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "unlimited-brass",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "str"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(df['orders'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "tough-angola",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "      <th>x</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium              x  \n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  [70, 10, 100]  \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60   [40, 20, 80]  \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49   [120, 5, 50]  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['x']=pd.eval(df['orders'].str.split(','))\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "organizational-judge",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "180"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(df['x'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "intensive-second",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    [70, 10, 100]\n",
       "1     [40, 20, 80]\n",
       "2     [120, 5, 50]\n",
       "Name: c1, dtype: object"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['c1']=df['orders'].str.split(',').apply(lambda v: list(map(int, v)))\n",
    "df['c1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "executed-sapphire",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[70, 10, 100]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['c1'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "angry-smoke",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "      <th>x</th>\n",
       "      <th>c1</th>\n",
       "      <th>c2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium              x             c1  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  [70, 10, 100]  [70, 10, 100]   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60   [40, 20, 80]   [40, 20, 80]   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49   [120, 5, 50]   [120, 5, 50]   \n",
       "\n",
       "              c2  \n",
       "0  [70, 10, 100]  \n",
       "1   [40, 20, 80]  \n",
       "2   [120, 5, 50]  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['c2']=pd.eval(df['orders2'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "peripheral-provision",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[70, 10, 100]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['c2'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "classified-victorian",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    10\n",
       "1    20\n",
       "2     5\n",
       "Name: c2, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['c2'].apply(lambda v: v[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "historic-space",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "      <th>x</th>\n",
       "      <th>c1</th>\n",
       "      <th>c2</th>\n",
       "      <th>d1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[hot, cold, frappe]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[hot, cold]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[hot]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium              x             c1  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  [70, 10, 100]  [70, 10, 100]   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60   [40, 20, 80]   [40, 20, 80]   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49   [120, 5, 50]   [120, 5, 50]   \n",
       "\n",
       "              c2                   d1  \n",
       "0  [70, 10, 100]  [hot, cold, frappe]  \n",
       "1   [40, 20, 80]          [hot, cold]  \n",
       "2   [120, 5, 50]                [hot]  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['d1']=df['serve'].str.split(',')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "southern-gardening",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "      <th>x</th>\n",
       "      <th>c1</th>\n",
       "      <th>c2</th>\n",
       "      <th>d1</th>\n",
       "      <th>d2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[hot, cold, frappe]</td>\n",
       "      <td>[hot, cold, frappe]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[hot, cold]</td>\n",
       "      <td>[hot, cold]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[hot]</td>\n",
       "      <td>[hot]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium              x             c1  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  [70, 10, 100]  [70, 10, 100]   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60   [40, 20, 80]   [40, 20, 80]   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49   [120, 5, 50]   [120, 5, 50]   \n",
       "\n",
       "              c2                   d1                   d2  \n",
       "0  [70, 10, 100]  [hot, cold, frappe]  [hot, cold, frappe]  \n",
       "1   [40, 20, 80]          [hot, cold]          [hot, cold]  \n",
       "2   [120, 5, 50]                [hot]                [hot]  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['d2']=pd.eval(df['serve2'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "integrated-stuart",
   "metadata": {},
   "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>menu</th>\n",
       "      <th>price</th>\n",
       "      <th>price2</th>\n",
       "      <th>orders</th>\n",
       "      <th>orders2</th>\n",
       "      <th>serve</th>\n",
       "      <th>serve2</th>\n",
       "      <th>price_j</th>\n",
       "      <th>price_k</th>\n",
       "      <th>price_j2</th>\n",
       "      <th>price_medium</th>\n",
       "      <th>x</th>\n",
       "      <th>c1</th>\n",
       "      <th>c2</th>\n",
       "      <th>d1</th>\n",
       "      <th>d2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mocha</td>\n",
       "      <td>{\"S\":40, \"M\":50, \"L\":60}</td>\n",
       "      <td>{'S':40, 'M':50, 'L':60}</td>\n",
       "      <td>70,10,100</td>\n",
       "      <td>[70,10,100]</td>\n",
       "      <td>hot,cold,frappe</td>\n",
       "      <td>['hot', 'cold', 'frappe']</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>{'S': 40, 'M': 50, 'L': 60}</td>\n",
       "      <td>50</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[70, 10, 100]</td>\n",
       "      <td>[hot, cold, frappe]</td>\n",
       "      <td>[hot, cold, frappe]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>latte</td>\n",
       "      <td>{\"S\":45, \"M\":60, \"L\":70}</td>\n",
       "      <td>{'S':45, 'M':60, 'L':70}</td>\n",
       "      <td>40,20,80</td>\n",
       "      <td>[40,20,80]</td>\n",
       "      <td>hot,cold</td>\n",
       "      <td>['hot', 'cold']</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>{'S': 45, 'M': 60, 'L': 70}</td>\n",
       "      <td>60</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[40, 20, 80]</td>\n",
       "      <td>[hot, cold]</td>\n",
       "      <td>[hot, cold]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>espresso</td>\n",
       "      <td>{\"S\":39, \"M\":49, \"L\":59}</td>\n",
       "      <td>{'S':39, 'M':49, 'L':59}</td>\n",
       "      <td>120,5,50</td>\n",
       "      <td>[120,5,50]</td>\n",
       "      <td>hot</td>\n",
       "      <td>['hot']</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>{'S': 39, 'M': 49, 'L': 59}</td>\n",
       "      <td>49</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[120, 5, 50]</td>\n",
       "      <td>[hot]</td>\n",
       "      <td>[hot]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       menu                     price                    price2     orders  \\\n",
       "0     mocha  {\"S\":40, \"M\":50, \"L\":60}  {'S':40, 'M':50, 'L':60}  70,10,100   \n",
       "1     latte  {\"S\":45, \"M\":60, \"L\":70}  {'S':45, 'M':60, 'L':70}   40,20,80   \n",
       "2  espresso  {\"S\":39, \"M\":49, \"L\":59}  {'S':39, 'M':49, 'L':59}   120,5,50   \n",
       "\n",
       "       orders2            serve                     serve2  \\\n",
       "0  [70,10,100]  hot,cold,frappe  ['hot', 'cold', 'frappe']   \n",
       "1   [40,20,80]         hot,cold            ['hot', 'cold']   \n",
       "2   [120,5,50]              hot                    ['hot']   \n",
       "\n",
       "                       price_j                      price_k  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}  {'S': 40, 'M': 50, 'L': 60}   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}  {'S': 45, 'M': 60, 'L': 70}   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}  {'S': 39, 'M': 49, 'L': 59}   \n",
       "\n",
       "                      price_j2  price_medium              x             c1  \\\n",
       "0  {'S': 40, 'M': 50, 'L': 60}            50  [70, 10, 100]  [70, 10, 100]   \n",
       "1  {'S': 45, 'M': 60, 'L': 70}            60   [40, 20, 80]   [40, 20, 80]   \n",
       "2  {'S': 39, 'M': 49, 'L': 59}            49   [120, 5, 50]   [120, 5, 50]   \n",
       "\n",
       "              c2                   d1                   d2  \n",
       "0  [70, 10, 100]  [hot, cold, frappe]  [hot, cold, frappe]  \n",
       "1   [40, 20, 80]          [hot, cold]          [hot, cold]  \n",
       "2   [120, 5, 50]                [hot]                [hot]  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['d2']=df['serve2'].map(ast.literal_eval)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "altered-present",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['hot', 'cold', 'frappe']"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['d2'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "favorite-myanmar",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'frappe'"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['d2'][0][2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "scheduled-apparel",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df.to_csv('out.tsv', sep='\\t', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}