notebook/analyze-0-9-26.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d630a715-9081-4bb7-a970-84f4ec775115",
   "metadata": {},
   "source": [
    "# Detailed analysis for version 0.9.26 on Intel i7-12700\n",
    "\n",
    "Copyright 2022 https://www.math-crypto.com -- GNU General Public License"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "70e12d0b-b6f1-400d-bb71-742d6bd2ed76",
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import random\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib \n",
    "\n",
    "matplotlib.rc('xtick', labelsize=12) \n",
    "matplotlib.rc('ytick', labelsize=12)\n",
    "plt.rcParams.update({'font.size': 16})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "cb9f32ba-6e2f-4b08-a1f9-63b9600cb21b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Nice boxplot with x axis sorted according to median\n",
    "boxprops = dict(linewidth=1.5,color='darkblue')\n",
    "medianprops = dict(linewidth=2, color='firebrick')\n",
    "def boxplot_sorted(df, by, column, ax):\n",
    "    df2 = pd.DataFrame({col:vals[column] for col, vals in df.groupby(by)})\n",
    "    meds = df2.median().sort_values()\n",
    "    df2[meds.index].boxplot(rot=90, ax=ax, boxprops=boxprops, medianprops=medianprops)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75b06191-002f-4d50-97eb-a93fed994412",
   "metadata": {},
   "source": [
    "## Read parsed benchmark data. Clean up a bit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b2a4dd97-6f66-4142-931d-7055ebda52fa",
   "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>index</th>\n",
       "      <th>host</th>\n",
       "      <th>date</th>\n",
       "      <th>ver</th>\n",
       "      <th>nb_run</th>\n",
       "      <th>nb_build</th>\n",
       "      <th>cpu</th>\n",
       "      <th>BLAKE2-256</th>\n",
       "      <th>SR25519-Verify</th>\n",
       "      <th>Copy</th>\n",
       "      <th>Seq_Write</th>\n",
       "      <th>Rnd_Write</th>\n",
       "      <th>toolchain</th>\n",
       "      <th>arch</th>\n",
       "      <th>codegen</th>\n",
       "      <th>lto_ldd</th>\n",
       "      <th>profile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>work-pc</td>\n",
       "      <td>2022-Aug-02_06h09</td>\n",
       "      <td>0.9.26</td>\n",
       "      <td>6</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1006.87</td>\n",
       "      <td>22820.0</td>\n",
       "      <td>2990.0</td>\n",
       "      <td>1280.0</td>\n",
       "      <td>stable</td>\n",
       "      <td>skylake</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>release</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>work-pc</td>\n",
       "      <td>2022-Aug-02_06h09</td>\n",
       "      <td>0.9.26</td>\n",
       "      <td>15</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1013.17</td>\n",
       "      <td>23110.0</td>\n",
       "      <td>2980.0</td>\n",
       "      <td>1270.0</td>\n",
       "      <td>nightly</td>\n",
       "      <td>alderlake</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>release</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>work-pc</td>\n",
       "      <td>2022-Aug-02_06h09</td>\n",
       "      <td>0.9.26</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1340.0</td>\n",
       "      <td>977.82</td>\n",
       "      <td>22900.0</td>\n",
       "      <td>2990.0</td>\n",
       "      <td>1280.0</td>\n",
       "      <td>nightly</td>\n",
       "      <td>none</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>release</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>work-pc</td>\n",
       "      <td>2022-Aug-02_06h09</td>\n",
       "      <td>0.9.26</td>\n",
       "      <td>16</td>\n",
       "      <td>24</td>\n",
       "      <td>0.1</td>\n",
       "      <td>1340.0</td>\n",
       "      <td>996.89</td>\n",
       "      <td>23920.0</td>\n",
       "      <td>2920.0</td>\n",
       "      <td>1280.0</td>\n",
       "      <td>nightly</td>\n",
       "      <td>none</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>release</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>work-pc</td>\n",
       "      <td>2022-Aug-02_06h09</td>\n",
       "      <td>0.9.26</td>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1015.56</td>\n",
       "      <td>22690.0</td>\n",
       "      <td>2960.0</td>\n",
       "      <td>1270.0</td>\n",
       "      <td>stable</td>\n",
       "      <td>skylake</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>release</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index     host               date     ver  nb_run nb_build  cpu  \\\n",
       "0      0  work-pc  2022-Aug-02_06h09  0.9.26       6       17  0.0   \n",
       "1      1  work-pc  2022-Aug-02_06h09  0.9.26      15       32  0.0   \n",
       "2      2  work-pc  2022-Aug-02_06h09  0.9.26      12       23  0.0   \n",
       "3      3  work-pc  2022-Aug-02_06h09  0.9.26      16       24  0.1   \n",
       "4      4  work-pc  2022-Aug-02_06h09  0.9.26      18       16  0.0   \n",
       "\n",
       "   BLAKE2-256  SR25519-Verify     Copy  Seq_Write  Rnd_Write toolchain  \\\n",
       "0      1410.0         1006.87  22820.0     2990.0     1280.0    stable   \n",
       "1      1410.0         1013.17  23110.0     2980.0     1270.0   nightly   \n",
       "2      1340.0          977.82  22900.0     2990.0     1280.0   nightly   \n",
       "3      1340.0          996.89  23920.0     2920.0     1280.0   nightly   \n",
       "4      1410.0         1015.56  22690.0     2960.0     1270.0    stable   \n",
       "\n",
       "        arch  codegen  lto_ldd  profile  \n",
       "0    skylake     True     True  release  \n",
       "1  alderlake     True    False  release  \n",
       "2       none    False     True  release  \n",
       "3       none     True    False  release  \n",
       "4    skylake     True    False  release  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_feather(\"../processed/todo/0.9.26_i7-12700_2022-Aug-02_06h09.feather\")\n",
    "\n",
    "df['arch'] = df['arch'].fillna('none')\n",
    "df['host'] = df['host'].astype('category')\n",
    "df['arch'] = df['arch'].astype('category')\n",
    "df['ver'] = df['ver'].astype('category')\n",
    "df['profile'] = df['profile'].astype('category')\n",
    "df['codegen'] = df['codegen'].astype('bool')\n",
    "df['lto_ldd'] = df['lto_ldd'].astype('bool')\n",
    "df['nb_run'] = df['nb_run'].astype('int')  \n",
    "\n",
    "df['SR25519-Verify'] = df['SR25519-Verify']*1000 # same as in benchmark palette\n",
    "\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6aa70bf0-1348-46a0-a298-deb762851ce3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check for highest CPU percentage (TODO remove outliers if any)\n",
    "max(df.loc[:,\"cpu\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "669cd939-e078-44df-8b0f-a1ed34fd78af",
   "metadata": {},
   "source": [
    "## Boxplot of all scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e150d99e-217a-473a-8d9c-7c27a552e343",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1224x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2, figsize=(17, 5))\n",
    "fig.suptitle(\"CPU scores for i7-12700\")\n",
    "boxplot_sorted(df, by=\"nb_build\", column=\"BLAKE2-256\", ax=ax[0])\n",
    "ax[0].set_title('BLAKE2-256')\n",
    "ax[0].set_ylabel('GiB/s')\n",
    "boxplot_sorted(df, by=\"nb_build\", column=\"SR25519-Verify\", ax=ax[1])\n",
    "ax[1].set_title('SR25519-Verify')\n",
    "ax[1].set_ylabel('KiB/s')\n",
    "fig.savefig('images/scores_cpu_i7_12700.svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d1a973b8-62f8-48a9-900f-ed7c1f7c895e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1224x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2, figsize=(17, 5))\n",
    "fig.suptitle(\"Disk scores for i7-12700 with NVMe\")\n",
    "boxplot_sorted(df, by=\"nb_build\", column=\"Seq_Write\", ax=ax[0])\n",
    "ax[0].set_title('Seq Write')\n",
    "ax[0].set_ylabel('MiB/s')\n",
    "boxplot_sorted(df, by=\"nb_build\", column=\"Rnd_Write\", ax=ax[1])\n",
    "ax[1].set_title('Rnd Write')\n",
    "ax[1].set_ylabel('MiB/s')\n",
    "fig.savefig('images/scores_disk_i7_12700.svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "25e451e4-e216-496f-a288-9c495ed0c3d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, figsize=(7.5, 5))\n",
    "fig.suptitle(\"Memory score for i7-12700\")\n",
    "boxplot_sorted(df, by=\"nb_build\", column=\"Copy\", ax=ax)\n",
    "ax.set_title('Copy')\n",
    "ax.set_ylabel('GiB/s')\n",
    "fig.savefig('images/scores_mem_i7_12700.svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ed9ab28-855d-4857-b05e-508eb96cdfc1",
   "metadata": {},
   "source": [
    "## Table of build options"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "52c607cf-0f85-473f-bce9-0a5676b1b4ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "|   nb_build | arch      | toolchain   | codegen   | lto_ldd   | profile    |\n",
      "|-----------:|:----------|:------------|:----------|:----------|:-----------|\n",
      "|         10 | none      | stable      | False     | False     | release    |\n",
      "|         11 | none      | stable      | False     | True      | release    |\n",
      "|         12 | none      | stable      | True      | False     | release    |\n",
      "|         13 | none      | stable      | True      | True      | release    |\n",
      "|         14 | skylake   | stable      | False     | False     | release    |\n",
      "|         15 | skylake   | stable      | False     | True      | release    |\n",
      "|         16 | skylake   | stable      | True      | False     | release    |\n",
      "|         17 | skylake   | stable      | True      | True      | release    |\n",
      "|         18 | alderlake | stable      | False     | False     | release    |\n",
      "|         19 | alderlake | stable      | False     | True      | release    |\n",
      "|         20 | alderlake | stable      | True      | False     | release    |\n",
      "|         21 | alderlake | stable      | True      | True      | release    |\n",
      "|         22 | none      | nightly     | False     | False     | release    |\n",
      "|         23 | none      | nightly     | False     | True      | release    |\n",
      "|         24 | none      | nightly     | True      | False     | release    |\n",
      "|         25 | none      | nightly     | True      | True      | release    |\n",
      "|         26 | skylake   | nightly     | False     | False     | release    |\n",
      "|         27 | skylake   | nightly     | False     | True      | release    |\n",
      "|         28 | skylake   | nightly     | True      | False     | release    |\n",
      "|         29 | skylake   | nightly     | True      | True      | release    |\n",
      "|         30 | alderlake | nightly     | False     | False     | release    |\n",
      "|         31 | alderlake | nightly     | False     | True      | release    |\n",
      "|         32 | alderlake | nightly     | True      | False     | release    |\n",
      "|         33 | alderlake | nightly     | True      | True      | release    |\n",
      "|         34 | none      | stable      | False     | False     | production |\n",
      "|         35 | none      | stable      | True      | False     | production |\n",
      "|         36 | skylake   | stable      | False     | False     | production |\n",
      "|         37 | skylake   | stable      | True      | False     | production |\n",
      "|         38 | alderlake | stable      | False     | False     | production |\n",
      "|         39 | alderlake | stable      | True      | False     | production |\n",
      "|         40 | none      | nightly     | False     | False     | production |\n",
      "|         41 | none      | nightly     | True      | False     | production |\n",
      "|         42 | skylake   | nightly     | False     | False     | production |\n",
      "|         43 | skylake   | nightly     | True      | False     | production |\n",
      "|         44 | alderlake | nightly     | False     | False     | production |\n",
      "|         45 | alderlake | nightly     | True      | False     | production |\n"
     ]
    }
   ],
   "source": [
    "cat_cols = ['nb_build', 'arch', 'toolchain', 'codegen', 'lto_ldd', 'profile']\n",
    "print( df[cat_cols].set_index('nb_build').sort_index().drop_duplicates().to_markdown() )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ba27b99-12b5-4df3-8ca5-9034653ee53a",
   "metadata": {},
   "source": [
    "## Find winning build options (close to Pareto efficient)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e080f8f9-1f2d-4bde-8c03-f7f8123f48e5",
   "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>BLAKE2-256</th>\n",
       "      <th>SR25519-Verify</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_build</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1330.0</td>\n",
       "      <td>978.825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1330.0</td>\n",
       "      <td>984.075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1330.0</td>\n",
       "      <td>988.160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1340.0</td>\n",
       "      <td>999.890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>990.820</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          BLAKE2-256  SR25519-Verify\n",
       "nb_build                            \n",
       "10            1330.0         978.825\n",
       "11            1330.0         984.075\n",
       "12            1330.0         988.160\n",
       "13            1340.0         999.890\n",
       "14            1410.0         990.820"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate medians to find winning builds\n",
    "medians = df[[\"nb_build\", \"BLAKE2-256\", \"SR25519-Verify\"]].groupby(\"nb_build\").median()\n",
    "medians.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "34fd7faf-5738-4723-87e3-e5ee8da3fca6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# https://www.tutorialspoint.com/how-to-annotate-the-points-on-a-scatter-plot-with-automatically-placed-arrows-in-matplotlib\n",
    "fig, ax = plt.subplots(1, figsize=(7.5, 5))\n",
    "plt.rcParams[\"figure.figsize\"] = [7.00, 5.50]\n",
    "plt.rcParams[\"figure.autolayout\"] = True\n",
    "xpoints = medians[\"BLAKE2-256\"]\n",
    "ypoints = medians[\"SR25519-Verify\"]\n",
    "labels = medians.index\n",
    "plt.scatter(xpoints, ypoints) #, c=xpoints)\n",
    "# These labels where found by looking manually at ``medians``\n",
    "labels_to_plot = [\"official\", \"docker\", \"45\", \"39\", \"32\", \"17\", \"33\", \"20\", \"21\"]\n",
    "ii = 0\n",
    "# angles\n",
    "ts = np.linspace(0,2*math.pi,int(len(labels_to_plot)*1.2))\n",
    "np.random.seed(1)\n",
    "np.random.shuffle(ts)\n",
    "xts = np.cos(ts)\n",
    "yts = np.sin(ts)\n",
    "for label, x, y in zip(labels, xpoints, ypoints):\n",
    "    if label in labels_to_plot:\n",
    "        plt.annotate(\n",
    "          label,\n",
    "          xy=(x, y), xytext=(xts[ii]*60, yts[ii]*60),\n",
    "          textcoords='offset points', va='center', ha='center',\n",
    "          bbox=dict(boxstyle='round,pad=0.5', fc='lightgreen', alpha=0.5),\n",
    "          arrowprops=dict(arrowstyle='->', connectionstyle='arc3,rad=0')\n",
    "        )\n",
    "        ii = ii + 1\n",
    "plt.xlabel(\"BLAKE2-256\")\n",
    "plt.ylabel(\"SR25519-Verify\")\n",
    "plt.show()\n",
    "fig.savefig('images/scatter_cpu_i7_12700.svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b1898637-168b-4172-be9d-b24b151b04a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# From the figure we select these (almost) Pareto efficient builds\n",
    "pareto = [\"45\", \"39\", \"32\", \"17\", \"33\", \"20\", \"21\"]\n",
    "df_sel = df[df[\"nb_build\"].isin(pareto)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "680d23ad-0bd1-4c88-b391-7dfed5dcd1f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1224x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2, figsize=(17, 5))\n",
    "fig.suptitle(\"CPU scores for i7-12700\")\n",
    "boxplot_sorted(df_sel, by=\"nb_build\", column=\"BLAKE2-256\", ax=ax[0])\n",
    "ax[0].set_title('BLAKE2-256')\n",
    "ax[0].set_ylabel('GiB/s')\n",
    "boxplot_sorted(df_sel, by=\"nb_build\", column=\"SR25519-Verify\", ax=ax[1])\n",
    "ax[1].set_title('SR25519-Verify')\n",
    "ax[1].set_ylabel('KiB/s')\n",
    "fig.savefig('images/pareto_scores_cpu_i7_12700.svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "521331ff-0e4e-45da-884d-d44fd8c465fc",
   "metadata": {},
   "source": [
    "## Summary table of winning builds\n",
    "\n",
    "On the overview page of our website, we have a simple table with the median of the winning build. Here we check whether it is OK to simply that that value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "86086056-e131-4d90-b195-e9518390a242",
   "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>nb_build</th>\n",
       "      <th>BLAKE2-256</th>\n",
       "      <th>SR25519-Verify</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1006.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>32</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1013.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>21</td>\n",
       "      <td>1410.0</td>\n",
       "      <td>1005.42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>official</td>\n",
       "      <td>1340.0</td>\n",
       "      <td>997.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>45</td>\n",
       "      <td>1400.0</td>\n",
       "      <td>1022.29</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    nb_build  BLAKE2-256  SR25519-Verify\n",
       "0         17      1410.0         1006.87\n",
       "1         32      1410.0         1013.17\n",
       "8         21      1410.0         1005.42\n",
       "12  official      1340.0          997.85\n",
       "16        45      1400.0         1022.29"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sel = pareto + [\"docker\",  \"official\"]\n",
    "df_sel2 = df[df[\"nb_build\"].isin(sel)].loc[:,[\"nb_build\", \"BLAKE2-256\", \"SR25519-Verify\"]]\n",
    "df_sel2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f192bcd0-93f7-4b11-8756-96521981ee1c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<AxesSubplot:title={'center':'BLAKE2-256'}>,\n",
       "        <AxesSubplot:title={'center':'SR25519-Verify'}>]], dtype=object)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 504x396 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# observation are not normally distributed, but let's ignore that for simplicity when calculating standard errors\n",
    "df_sel2[df_sel2[\"nb_build\"]=='official'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "276bd9a8-3c03-49e7-abdf-021e4cc76ed5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<AxesSubplot:title={'center':'BLAKE2-256'}>,\n",
       "        <AxesSubplot:title={'center':'SR25519-Verify'}>]], dtype=object)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 504x396 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_sel2[df_sel2[\"nb_build\"]=='17'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "e1ed9a61-d507-45cc-8d5c-cb6f94235af3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Summary stats with ± mean based on 95% CI\n",
    "def statistics_table(df, score):\n",
    "    stats = df.groupby(\"nb_build\")[score].agg(['median', 'mean', 'sem'])\n",
    "    # stats['ci95_lo'] = stats['mean'] - 1.96* stats['sem']\n",
    "    # stats['ci95_hi'] = stats['mean'] + 1.96* stats['sem']\n",
    "    stats['± mean'] = 1.96* stats['sem']\n",
    "    stats['± mean pct'] = stats['± mean']/stats['mean']*100\n",
    "    median_official = stats['median']['official']\n",
    "    stats['impr med off'] = (stats['median'] - median_official)\n",
    "    stats['impr med off pct'] = stats['impr med off']/median_official*100\n",
    "    return stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "03b2849c-7d57-4759-8360-8fc115ac6c49",
   "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>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>sem</th>\n",
       "      <th>± mean</th>\n",
       "      <th>± mean pct</th>\n",
       "      <th>impr med off</th>\n",
       "      <th>impr med off pct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_build</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>1409.5</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.980000</td>\n",
       "      <td>0.069528</td>\n",
       "      <td>70.0</td>\n",
       "      <td>5.223881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>1407.5</td>\n",
       "      <td>0.993399</td>\n",
       "      <td>1.947063</td>\n",
       "      <td>0.138335</td>\n",
       "      <td>70.0</td>\n",
       "      <td>5.223881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>1409.5</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.980000</td>\n",
       "      <td>0.069528</td>\n",
       "      <td>70.0</td>\n",
       "      <td>5.223881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>1409.0</td>\n",
       "      <td>0.688247</td>\n",
       "      <td>1.348965</td>\n",
       "      <td>0.095739</td>\n",
       "      <td>70.0</td>\n",
       "      <td>5.223881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>1420.0</td>\n",
       "      <td>1416.0</td>\n",
       "      <td>1.123903</td>\n",
       "      <td>2.202850</td>\n",
       "      <td>0.155568</td>\n",
       "      <td>80.0</td>\n",
       "      <td>5.970149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>1400.0</td>\n",
       "      <td>1396.5</td>\n",
       "      <td>1.312892</td>\n",
       "      <td>2.573267</td>\n",
       "      <td>0.184265</td>\n",
       "      <td>60.0</td>\n",
       "      <td>4.477612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>1400.0</td>\n",
       "      <td>1398.5</td>\n",
       "      <td>0.819178</td>\n",
       "      <td>1.605589</td>\n",
       "      <td>0.114808</td>\n",
       "      <td>60.0</td>\n",
       "      <td>4.477612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>docker</th>\n",
       "      <td>1340.0</td>\n",
       "      <td>1340.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>official</th>\n",
       "      <td>1340.0</td>\n",
       "      <td>1337.5</td>\n",
       "      <td>1.428101</td>\n",
       "      <td>2.799079</td>\n",
       "      <td>0.209277</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          median    mean       sem    ± mean  ± mean pct  impr med off  \\\n",
       "nb_build                                                                 \n",
       "17        1410.0  1409.5  0.500000  0.980000    0.069528          70.0   \n",
       "20        1410.0  1407.5  0.993399  1.947063    0.138335          70.0   \n",
       "21        1410.0  1409.5  0.500000  0.980000    0.069528          70.0   \n",
       "32        1410.0  1409.0  0.688247  1.348965    0.095739          70.0   \n",
       "33        1420.0  1416.0  1.123903  2.202850    0.155568          80.0   \n",
       "39        1400.0  1396.5  1.312892  2.573267    0.184265          60.0   \n",
       "45        1400.0  1398.5  0.819178  1.605589    0.114808          60.0   \n",
       "docker    1340.0  1340.0  0.000000  0.000000    0.000000           0.0   \n",
       "official  1340.0  1337.5  1.428101  2.799079    0.209277           0.0   \n",
       "\n",
       "          impr med off pct  \n",
       "nb_build                    \n",
       "17                5.223881  \n",
       "20                5.223881  \n",
       "21                5.223881  \n",
       "32                5.223881  \n",
       "33                5.970149  \n",
       "39                4.477612  \n",
       "45                4.477612  \n",
       "docker            0.000000  \n",
       "official          0.000000  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats_score1 = statistics_table(df_sel2, 'BLAKE2-256')\n",
    "stats_score1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3942f388-95f3-46a2-ba3a-957d01b17106",
   "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>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>sem</th>\n",
       "      <th>± mean</th>\n",
       "      <th>± mean pct</th>\n",
       "      <th>impr med off</th>\n",
       "      <th>impr med off pct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_build</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1011.210</td>\n",
       "      <td>1009.1315</td>\n",
       "      <td>0.713452</td>\n",
       "      <td>1.398366</td>\n",
       "      <td>0.138571</td>\n",
       "      <td>14.080</td>\n",
       "      <td>1.412053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1010.020</td>\n",
       "      <td>1010.1400</td>\n",
       "      <td>1.470067</td>\n",
       "      <td>2.881332</td>\n",
       "      <td>0.285241</td>\n",
       "      <td>12.890</td>\n",
       "      <td>1.292710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1010.040</td>\n",
       "      <td>1008.2360</td>\n",
       "      <td>0.651780</td>\n",
       "      <td>1.277489</td>\n",
       "      <td>0.126705</td>\n",
       "      <td>12.910</td>\n",
       "      <td>1.294716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1012.940</td>\n",
       "      <td>1011.0875</td>\n",
       "      <td>0.702205</td>\n",
       "      <td>1.376321</td>\n",
       "      <td>0.136123</td>\n",
       "      <td>15.810</td>\n",
       "      <td>1.585551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>998.625</td>\n",
       "      <td>991.7380</td>\n",
       "      <td>7.788737</td>\n",
       "      <td>15.265924</td>\n",
       "      <td>1.539310</td>\n",
       "      <td>1.495</td>\n",
       "      <td>0.149930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>1021.870</td>\n",
       "      <td>1014.3980</td>\n",
       "      <td>2.424813</td>\n",
       "      <td>4.752634</td>\n",
       "      <td>0.468518</td>\n",
       "      <td>24.740</td>\n",
       "      <td>2.481121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>1022.550</td>\n",
       "      <td>1016.7925</td>\n",
       "      <td>2.232816</td>\n",
       "      <td>4.376320</td>\n",
       "      <td>0.430404</td>\n",
       "      <td>25.420</td>\n",
       "      <td>2.549317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>docker</th>\n",
       "      <td>996.225</td>\n",
       "      <td>996.3450</td>\n",
       "      <td>0.188422</td>\n",
       "      <td>0.369307</td>\n",
       "      <td>0.037066</td>\n",
       "      <td>-0.905</td>\n",
       "      <td>-0.090760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>official</th>\n",
       "      <td>997.130</td>\n",
       "      <td>996.7150</td>\n",
       "      <td>0.332262</td>\n",
       "      <td>0.651234</td>\n",
       "      <td>0.065338</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            median       mean       sem     ± mean  ± mean pct  impr med off  \\\n",
       "nb_build                                                                       \n",
       "17        1011.210  1009.1315  0.713452   1.398366    0.138571        14.080   \n",
       "20        1010.020  1010.1400  1.470067   2.881332    0.285241        12.890   \n",
       "21        1010.040  1008.2360  0.651780   1.277489    0.126705        12.910   \n",
       "32        1012.940  1011.0875  0.702205   1.376321    0.136123        15.810   \n",
       "33         998.625   991.7380  7.788737  15.265924    1.539310         1.495   \n",
       "39        1021.870  1014.3980  2.424813   4.752634    0.468518        24.740   \n",
       "45        1022.550  1016.7925  2.232816   4.376320    0.430404        25.420   \n",
       "docker     996.225   996.3450  0.188422   0.369307    0.037066        -0.905   \n",
       "official   997.130   996.7150  0.332262   0.651234    0.065338         0.000   \n",
       "\n",
       "          impr med off pct  \n",
       "nb_build                    \n",
       "17                1.412053  \n",
       "20                1.292710  \n",
       "21                1.294716  \n",
       "32                1.585551  \n",
       "33                0.149930  \n",
       "39                2.481121  \n",
       "45                2.549317  \n",
       "docker           -0.090760  \n",
       "official          0.000000  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats_score2 = statistics_table(df_sel2, 'SR25519-Verify')\n",
    "stats_score2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd8ca788-fcd9-48ce-9427-ad0551dd7e3a",
   "metadata": {},
   "source": [
    "**The 95% CI on each mean is quite small relative to the value itself (< 1% except for one value).** We can thus safely ignore these errors on the empirical mean and treat the stated mean *as is*. In addition, due to the skewness of the distribution (see figure above), we prefer to take the median of the score."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27b14a8b-be0d-4f94-8b05-5063e21cef35",
   "metadata": {},
   "source": [
    "Let's finally make a simple summary table for one build so we can compare versions and CPUs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "504ca460-8fc2-4cc2-818a-229fa941a563",
   "metadata": {},
   "outputs": [],
   "source": [
    "build = '32' # reasons explained on website\n",
    "stats1 = stats_score1.loc[[build,'official', 'docker'], ['median', 'impr med off pct']]\n",
    "stats1 = stats1.rename(columns={\"median\": \"BLAKE2-256\", \"impr med off pct\": \"relative diff (%)\"})\n",
    "stats2 = stats_score2.loc[[build,'official', 'docker'], ['median', 'impr med off pct']]\n",
    "stats2 = stats2.rename(columns={\"median\": \"SR25519-Verify\", \"impr med off pct\": \"relative diff (%)\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c9a6f7b7-61be-407e-9f31-7b39e21074d9",
   "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>SR25519-Verify</th>\n",
       "      <th>relative diff (%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_build</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1012.940</td>\n",
       "      <td>1.585551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>official</th>\n",
       "      <td>997.130</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>docker</th>\n",
       "      <td>996.225</td>\n",
       "      <td>-0.090760</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          SR25519-Verify  relative diff (%)\n",
       "nb_build                                   \n",
       "32              1012.940           1.585551\n",
       "official         997.130           0.000000\n",
       "docker           996.225          -0.090760"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d3f4050b-da49-40a5-b5a4-3a9caee42973",
   "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>BLAKE2-256</th>\n",
       "      <th>relative diff (%)</th>\n",
       "      <th>SR25519-Verify</th>\n",
       "      <th>relative diff (%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_build</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1410.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>1012.9</td>\n",
       "      <td>1.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>official</th>\n",
       "      <td>1340.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>997.1</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>docker</th>\n",
       "      <td>1340.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>996.2</td>\n",
       "      <td>-0.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          BLAKE2-256  relative diff (%)  SR25519-Verify  relative diff (%)\n",
       "nb_build                                                                  \n",
       "32            1410.0                5.2          1012.9                1.6\n",
       "official      1340.0                0.0           997.1                0.0\n",
       "docker        1340.0                0.0           996.2               -0.1"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum_stats_build = pd.concat([stats1, stats2], axis=1).round(1)\n",
    "sum_stats_build"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "38ebbe9c-be11-4927-b61b-e7e228a2d78a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "| nb_build   |   BLAKE2-256 |   relative diff (%) |   SR25519-Verify |   relative diff (%) |\n",
      "|:-----------|-------------:|--------------------:|-----------------:|--------------------:|\n",
      "| 32         |         1410 |                 5.2 |           1012.9 |                 1.6 |\n",
      "| official   |         1340 |                 0   |            997.1 |                 0   |\n",
      "| docker     |         1340 |                 0   |            996.2 |                -0.1 |\n"
     ]
    }
   ],
   "source": [
    "print( sum_stats_build.to_markdown() )"
   ]
  }
 ],
 "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": "3.10.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}