hyperloglog-rs 0.1.56

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9da8f0d7",
   "metadata": {},
   "source": [
    "# Union evaluations\n",
    "In this notebook, we will evaluate the Jaccard similarity of the different methods. The first will be the implementation as provided from this Rust crate HyperLogLog couters, testing all of the available bits and precisions. Note that, since there are no Python bindings for the Rust crate (yet) we run the cose as one of the test in the crate test suite. \n",
    "\n",
    "The second one will be MinHash, using the implementation provided by the datasketch library. We will compare the performance of the two methods for the same amount of memory used.\n",
    "\n",
    "## What is an HyperLogLog counter?\n",
    "An HyperLogLog counter is a probabilistic data structure used to estimate the cardinality of a set. It is based on the observation that the cardinality of a set can be estimated by the maximum number of leading zeros in the binary representation of the hashes of the elements of the set. The HyperLogLog counter is a data structure that stores the maximum number of leading zeros for a set of hashes. The counter is initialized with a number of bits, which determines the maximum number of leading zeros that can be stored. The counter is then updated with the hashes of the elements of the set. The estimate of the cardinality is then given by the harmonic mean of the values stored in the counter. HyperLogLog counters can be used to compute the cardinality of the union of two sets by taking the maximum of the values stored in the two counters, and therefore we can also compute the Jaccard similarity of two sets.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "dd95a263",
   "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>precision</th>\n",
       "      <th>bits</th>\n",
       "      <th>exact</th>\n",
       "      <th>old</th>\n",
       "      <th>recent</th>\n",
       "      <th>hash_name</th>\n",
       "      <th>memory</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>44199</td>\n",
       "      <td>18.064531</td>\n",
       "      <td>18.064531</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>44199</td>\n",
       "      <td>86.144000</td>\n",
       "      <td>86.144000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>44199</td>\n",
       "      <td>1378.304000</td>\n",
       "      <td>1378.304000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>44199</td>\n",
       "      <td>31895.666000</td>\n",
       "      <td>31895.666000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>44199</td>\n",
       "      <td>31895.666000</td>\n",
       "      <td>31895.666000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167995</th>\n",
       "      <td>17</td>\n",
       "      <td>2</td>\n",
       "      <td>29069</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>262144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167996</th>\n",
       "      <td>17</td>\n",
       "      <td>3</td>\n",
       "      <td>29069</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>393216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167997</th>\n",
       "      <td>17</td>\n",
       "      <td>4</td>\n",
       "      <td>29069</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>524288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167998</th>\n",
       "      <td>17</td>\n",
       "      <td>5</td>\n",
       "      <td>29069</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>655360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167999</th>\n",
       "      <td>17</td>\n",
       "      <td>6</td>\n",
       "      <td>29069</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>29055.014000</td>\n",
       "      <td>siphasher::sip::SipHasher13</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>168000 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        precision  bits  exact           old        recent  \\\n",
       "0               4     1  44199     18.064531     18.064531   \n",
       "1               4     2  44199     86.144000     86.144000   \n",
       "2               4     3  44199   1378.304000   1378.304000   \n",
       "3               4     4  44199  31895.666000  31895.666000   \n",
       "4               4     5  44199  31895.666000  31895.666000   \n",
       "...           ...   ...    ...           ...           ...   \n",
       "167995         17     2  29069  29055.014000  29055.014000   \n",
       "167996         17     3  29069  29055.014000  29055.014000   \n",
       "167997         17     4  29069  29055.014000  29055.014000   \n",
       "167998         17     5  29069  29055.014000  29055.014000   \n",
       "167999         17     6  29069  29055.014000  29055.014000   \n",
       "\n",
       "                          hash_name  memory  \n",
       "0       siphasher::sip::SipHasher13      16  \n",
       "1       siphasher::sip::SipHasher13      32  \n",
       "2       siphasher::sip::SipHasher13      48  \n",
       "3       siphasher::sip::SipHasher13      64  \n",
       "4       siphasher::sip::SipHasher13      80  \n",
       "...                             ...     ...  \n",
       "167995  siphasher::sip::SipHasher13  262144  \n",
       "167996  siphasher::sip::SipHasher13  393216  \n",
       "167997  siphasher::sip::SipHasher13  524288  \n",
       "167998  siphasher::sip::SipHasher13  655360  \n",
       "167999  siphasher::sip::SipHasher13  786432  \n",
       "\n",
       "[168000 rows x 7 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv(\"union_cardinality_benchmark.tsv\", sep=\"\\t\")\n",
    "df[\"memory\"] = 2**df.precision * df.bits\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1c3eed05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "713cfb043bc245a89cb356a377e49a38",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/100000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>precision</th>\n",
       "      <th>bits</th>\n",
       "      <th>exact</th>\n",
       "      <th>hll</th>\n",
       "      <th>nn</th>\n",
       "      <th>memory</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>397062</td>\n",
       "      <td>426806.470</td>\n",
       "      <td>391157.78</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>450075</td>\n",
       "      <td>478586.780</td>\n",
       "      <td>436966.63</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>105663</td>\n",
       "      <td>119218.484</td>\n",
       "      <td>108914.68</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>346393</td>\n",
       "      <td>366406.220</td>\n",
       "      <td>336592.90</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>265464</td>\n",
       "      <td>283791.660</td>\n",
       "      <td>259825.06</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>370409</td>\n",
       "      <td>394773.470</td>\n",
       "      <td>360504.88</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>308056</td>\n",
       "      <td>305078.840</td>\n",
       "      <td>279168.88</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>331971</td>\n",
       "      <td>367946.720</td>\n",
       "      <td>337073.16</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>320244</td>\n",
       "      <td>333806.600</td>\n",
       "      <td>304779.56</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>224682</td>\n",
       "      <td>245158.900</td>\n",
       "      <td>225524.06</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100000 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    precision  bits   exact         hll         nn  memory\n",
       "0           8     6  397062  426806.470  391157.78    1536\n",
       "0           8     6  450075  478586.780  436966.63    1536\n",
       "0           8     6  105663  119218.484  108914.68    1536\n",
       "0           8     6  346393  366406.220  336592.90    1536\n",
       "0           8     6  265464  283791.660  259825.06    1536\n",
       "..        ...   ...     ...         ...        ...     ...\n",
       "0           8     6  370409  394773.470  360504.88    1536\n",
       "0           8     6  308056  305078.840  279168.88    1536\n",
       "0           8     6  331971  367946.720  337073.16    1536\n",
       "0           8     6  320244  333806.600  304779.56    1536\n",
       "0           8     6  224682  245158.900  225524.06    1536\n",
       "\n",
       "[100000 rows x 6 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from glob import glob\n",
    "import pandas as pd\n",
    "from tqdm.auto import tqdm\n",
    "\n",
    "df = pd.concat([\n",
    "    pd.read_csv(path, sep=\"\\t\")\n",
    "    for path in tqdm(glob(\"union_test/*.tsv.gz\"))\n",
    "])\n",
    "\n",
    "df[\"memory\"] = 2**df.precision * df.bits\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "745dfb45",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13306"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "df[\"mse_hll\"] = (df.exact - df.hll)**2\n",
    "df[\"mse_nn\"] = (df.exact - df.nn)**2\n",
    "df[\"mse_rates\"] = np.log((df.exact - df.nn)**2 / (df.exact - df.hll)**2)\n",
    "\n",
    "(df[\"mse_rates\"] > 1.0).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "58e7ea9d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "df.mse_rates.hist(log=True, bins=1000)\n",
    "#plt.xscale(\"log\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5e9bfe8",
   "metadata": {},
   "source": [
    "We determine the number of u64 words to use for the MinHash to execute versions of MinHash with comparable memory usage to the HyperLogLog counters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ef240523",
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "\n",
    "bits = df.bits.unique()\n",
    "precision = df.precision.unique()\n",
    "\n",
    "number_of_words = {\n",
    "    # We divide again by 2 because the number of permutations used is a u64\n",
    "    # and by two again as \n",
    "    math.ceil(b * 2**p / 32) // 2 // 2\n",
    "    for b in bits\n",
    "    for p in precision\n",
    "}\n",
    "\n",
    "# Some HyperLogLog counters require less than 64 bits, so there will be\n",
    "# values in the list that are zero. We remove them.\n",
    "number_of_words = [\n",
    "    word\n",
    "    for word in number_of_words\n",
    "    if word > 0\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b8c5238",
   "metadata": {},
   "source": [
    "We parallelize the computation of the Jaccard similarity using MinHash, which is significantly slower than the HyperLogLog counters as while the former is an extensively optimized Rust implementation, the latter is more didactical Python implementation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b8106fcc",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>precision</th>\n",
       "      <th>bits</th>\n",
       "      <th>memory</th>\n",
       "      <th colspan=\"2\" halign=\"left\">old_squared_error</th>\n",
       "      <th colspan=\"2\" halign=\"left\">recent_squared_error</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>1536</td>\n",
       "      <td>1.961102e+09</td>\n",
       "      <td>2.419204e+09</td>\n",
       "      <td>3.716497e+08</td>\n",
       "      <td>6.508915e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  precision bits memory old_squared_error               recent_squared_error  \\\n",
       "                                     mean           std                 mean   \n",
       "0         8    6   1536      1.961102e+09  2.419204e+09         3.716497e+08   \n",
       "\n",
       "                 \n",
       "            std  \n",
       "0  6.508915e+08  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"old_squared_error\"] = (df.exact - df.hll)**2\n",
    "df[\"recent_squared_error\"] = (df.exact - df.nn)**2\n",
    "#df[\"bounded_squared_error\"] = (df.exact - df.bounded_approximation)**2\n",
    "columns = [\n",
    "    \"old_squared_error\",\n",
    "    \"recent_squared_error\",\n",
    "    #\"bounded_squared_error\"\n",
    "]\n",
    "data_hll = df.groupby([\"precision\", \"bits\", \"memory\",])[columns].agg([\"mean\", \"std\"])\n",
    "data_hll = data_hll.reset_index()\n",
    "data_hll"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "49d6a372",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.189511\n",
       "Name: mean, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_hll.recent_squared_error[\"mean\"] / data_hll.old_squared_error[\"mean\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a53f306",
   "metadata": {},
   "source": [
    "We visualize two versions of the HyperLogLog counters, one including also **EXTREMELY SMALL** registers of 1 and 2 bits, which really push the limits of the HyperLogLog counters, and one without them, which is more representative of the performance of the HyperLogLog counters for the more common use cases. We include these tiny registers as exploring the limits of the HyperLogLog counters is one of the main goals of this project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9a370282",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1920x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1920x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "bits_to_discard = (1, 2, 3, 4, 5)\n",
    "\n",
    "hash_name_line_style = {\n",
    "    \"Hasher64_1\": \"-\",\n",
    "    \"MetroHasher\": \"-\",\n",
    "    \"SipHasher13\": \"--\",\n",
    "    \"SipHasher24\": \"-.\",\n",
    "    \"HighwayHasher\": \":\",\n",
    "}\n",
    "\n",
    "hash_name_marker_style = {\n",
    "    \"Hasher64_1\": \"o\",\n",
    "    \"MetroHasher\": \"o\",\n",
    "    \"SipHasher13\": \"^\",\n",
    "    \"SipHasher24\": \"s\",\n",
    "    \"HighwayHasher\": \"x\",\n",
    "}\n",
    "\n",
    "for yscale in (\"linear\", \"log\"):\n",
    "    fig, axes = plt.subplots(dpi=300)\n",
    "    for bits in data_hll[\"bits\"].unique():\n",
    "        if bits in bits_to_discard:\n",
    "            continue\n",
    "        for column in columns:\n",
    "            filtered = data_hll[data_hll.bits == bits]\n",
    "\n",
    "            plt.errorbar(\n",
    "                filtered.memory,\n",
    "                filtered[column][\"mean\"],\n",
    "                filtered[column][\"std\"],\n",
    "                alpha=0.5,\n",
    "                label=f\"{bits}b, {column.split('_')[0]}\"\n",
    "            )\n",
    "    plt.legend()\n",
    "    plt.xscale(\"log\")\n",
    "    plt.yscale(yscale)\n",
    "    plt.ylabel(f\"Union MSE, 100k sets ({yscale})\")\n",
    "    plt.xlabel(\"Memory required (bits)\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5085bf55",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}