02-1_plot_run_info_stat_cogsci2019.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let's box plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from run_info_utils import *\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment_name = 'cogsci2019'\n",
    "plot_dir = 'plot_figures'\n",
    "fig_ext = 'pdf'\n",
    "all_correct_case = 'init_all_correct'  # 'init_complete_all_correct', 'init_all_correct'\n",
    "font_size = {'xlabel':20, 'ylabel':20, 'xtick':20, 'ytick':10}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_fig(fig_name):\n",
    "    plot_fig_path = '{}/{}.{}'.format(plot_dir, fig_name, fig_ext)\n",
    "    plt.savefig(plot_fig_path, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_run_info = import_all_run_info(experiment_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "700"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(all_run_info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['init_all_correct_digit-2_epoch', 'init_all_correct_digit-3_epoch', 'network_output_dimension', 'hidden_activation', 'last_digit-7_accuracy', 'init_all_correct_carry-1_epoch', 'batch_size', 'last_carry-0_accuracy', 'last_digit-3_accuracy', 'init_complete_all_correct_digit-2_epoch', 'last_digit-6_accuracy', 'init_all_correct_digit-1_epoch', 'init_complete_all_correct_carry-2_epoch', 'init_all_correct_carry-2_epoch', 'test_set_size', 'init_all_correct_digit-4_epoch', 'result_bits', 'init_all_correct_carry-3_epoch', 'all_correct_stop', 'last_test_loss', 'init_complete_all_correct_digit-6_epoch', 'last_digit-3_wrong', 'last_carry-2_accuracy', 'hidden_dimensions', 'init_all_correct_digit-5_epoch', 'init_complete_all_correct_digit-1_epoch', 'last_test_accuracy', 'last_digit-8_accuracy', 'init_complete_all_correct_digit-4_epoch', 'last_digit-6_wrong', 'init_complete_all_correct_digit-5_epoch', 'network_input_dimension', 'last_tlu_test_loss', 'init_all_correct_digit-6_epoch', 'last_digit-5_wrong', 'last_tlu_op_wrong', 'last_digit-4_wrong', 'last_digit-4_accuracy', 'last_digit-2_wrong', 'init_complete_all_correct_digit-8_epoch', 'last_carry-2_wrong', 'last_carry-3_wrong', 'last_digit-7_wrong', 'last_epoch', 'operator', 'last_digit-1_wrong', 'operand_bits', 'last_test_op_wrong', 'init_complete_all_correct_carry-3_epoch', 'last_digit-8_wrong', 'init_all_correct_digit-8_epoch', 'run_id', 'last_carry-0_wrong', 'init_all_correct_epoch', 'last_carry-1_accuracy', 'init_complete_all_correct_digit-3_epoch', 'init_all_correct_carry-0_epoch', 'init_complete_all_correct_digit-7_epoch', 'last_tlu_test_accuracy', 'learning_rate', 'init_all_correct_digit-7_epoch', 'init_complete_all_correct_carry-1_epoch', 'train_set_size', 'experiment_name', 'last_digit-2_accuracy', 'last_carry-3_accuracy', 'init_complete_all_correct_carry-0_epoch', 'last_digit-5_accuracy', 'optimizer', 'dev_set_size', 'last_carry-1_wrong', 'last_digit-1_accuracy'])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_run_info[0].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_run_info_add = import_all_run_info(experiment_name, 'add', all_correct=True)\n",
    "all_run_info_subtract = import_all_run_info(experiment_name, 'subtract', all_correct=True)\n",
    "all_run_info_multiply = import_all_run_info(experiment_name, 'multiply', all_correct=True)\n",
    "all_run_info_divide = import_all_run_info(experiment_name, 'divide', all_correct=True)\n",
    "all_run_info_modulo = import_all_run_info(experiment_name, 'modulo', all_correct=True)\n",
    "all_run_info_fixed_random = import_all_run_info(experiment_name, 'fixed_random', all_correct=True)\n",
    "all_run_info_random = import_all_run_info(experiment_name, 'random', all_correct=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100\n",
      "100\n",
      "100\n",
      "100\n",
      "100\n",
      "100\n",
      "100\n"
     ]
    }
   ],
   "source": [
    "print(len(all_run_info_add))\n",
    "print(len(all_run_info_subtract))\n",
    "print(len(all_run_info_multiply))\n",
    "print(len(all_run_info_divide))\n",
    "print(len(all_run_info_modulo)) \n",
    "print(len(all_run_info_fixed_random)) \n",
    "print(len(all_run_info_random)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "700"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(get_all_run_info_files('cogsci2019'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['learning_rate', 'init_all_correct_digit-2_epoch', 'last_carry-4_accuracy', 'network_output_dimension', 'hidden_activation', 'last_digit-7_accuracy', 'init_all_correct_carry-1_epoch', 'init_all_correct_digit-4_epoch', 'init_complete_all_correct_carry-0_epoch', 'batch_size', 'init_all_correct_digit-6_epoch', 'optimizer', 'last_digit-1_accuracy', 'init_all_correct_digit-1_epoch', 'init_complete_all_correct_carry-2_epoch', 'init_all_correct_carry-2_epoch', 'test_set_size', 'last_digit-7_wrong', 'last_test_op_wrong', 'init_all_correct_carry-3_epoch', 'hidden_dimensions', 'last_test_loss', 'last_carry-4_wrong', 'init_complete_all_correct_digit-2_epoch', 'last_carry-2_accuracy', 'last_test_accuracy', 'init_all_correct_digit-5_epoch', 'last_digit-8_accuracy', 'last_carry-0_wrong', 'init_complete_all_correct_digit-1_epoch', 'operator', 'last_digit-6_wrong', 'init_complete_all_correct_digit-5_epoch', 'network_input_dimension', 'last_tlu_test_loss', 'init_all_correct_digit-8_epoch', 'last_tlu_op_wrong', 'init_all_correct_carry-4_epoch', 'last_digit-4_wrong', 'last_digit-4_accuracy', 'last_digit-2_wrong', 'init_complete_all_correct_digit-8_epoch', 'last_carry-3_wrong', 'experiment_name', 'init_all_correct_digit-3_epoch', 'init_complete_all_correct_digit-4_epoch', 'last_digit-1_wrong', 'operand_bits', 'last_digit-6_accuracy', 'init_complete_all_correct_carry-3_epoch', 'last_digit-8_wrong', 'run_id', 'init_all_correct_epoch', 'init_complete_all_correct_carry-4_epoch', 'last_carry-1_accuracy', 'init_complete_all_correct_digit-3_epoch', 'last_digit-2_accuracy', 'result_bits', 'init_all_correct_carry-0_epoch', 'last_digit-5_wrong', 'init_complete_all_correct_digit-7_epoch', 'last_tlu_test_accuracy', 'last_epoch', 'init_all_correct_digit-7_epoch', 'init_complete_all_correct_carry-1_epoch', 'train_set_size', 'last_digit-3_accuracy', 'last_carry-2_wrong', 'init_complete_all_correct_digit-6_epoch', 'last_carry-3_accuracy', 'all_correct_stop', 'last_digit-5_accuracy', 'last_carry-0_accuracy', 'dev_set_size', 'last_carry-1_wrong', 'last_digit-3_wrong'])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_run_info_add[0].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_add = pd.DataFrame.from_dict(all_run_info_add)\n",
    "df_run_info_subtract = pd.DataFrame.from_dict(all_run_info_subtract)\n",
    "df_run_info_multiply = pd.DataFrame.from_dict(all_run_info_multiply)\n",
    "df_run_info_divide = pd.DataFrame.from_dict(all_run_info_divide)\n",
    "df_run_info_modulo = pd.DataFrame.from_dict(all_run_info_modulo)\n",
    "df_run_info_fixed_random = pd.DataFrame.from_dict(all_run_info_fixed_random)\n",
    "df_run_info_random = pd.DataFrame.from_dict(all_run_info_random)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>all_correct_stop</th>\n",
       "      <th>batch_size</th>\n",
       "      <th>dev_set_size</th>\n",
       "      <th>experiment_name</th>\n",
       "      <th>hidden_activation</th>\n",
       "      <th>hidden_dimensions</th>\n",
       "      <th>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>learning_rate</th>\n",
       "      <th>network_input_dimension</th>\n",
       "      <th>network_output_dimension</th>\n",
       "      <th>operand_bits</th>\n",
       "      <th>operator</th>\n",
       "      <th>optimizer</th>\n",
       "      <th>result_bits</th>\n",
       "      <th>run_id</th>\n",
       "      <th>test_set_size</th>\n",
       "      <th>train_set_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>True</td>\n",
       "      <td>32</td>\n",
       "      <td>256</td>\n",
       "      <td>cogsci2019</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>[64]</td>\n",
       "      <td>2593.0</td>\n",
       "      <td>2356.0</td>\n",
       "      <td>2907.0</td>\n",
       "      <td>2814.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>add</td>\n",
       "      <td>Adam</td>\n",
       "      <td>8</td>\n",
       "      <td>20190105091719</td>\n",
       "      <td>256</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>True</td>\n",
       "      <td>32</td>\n",
       "      <td>256</td>\n",
       "      <td>cogsci2019</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>[64]</td>\n",
       "      <td>1463.0</td>\n",
       "      <td>1899.0</td>\n",
       "      <td>3052.0</td>\n",
       "      <td>3114.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>add</td>\n",
       "      <td>Adam</td>\n",
       "      <td>8</td>\n",
       "      <td>20190105140503</td>\n",
       "      <td>256</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>32</td>\n",
       "      <td>256</td>\n",
       "      <td>cogsci2019</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>[64]</td>\n",
       "      <td>1782.0</td>\n",
       "      <td>1858.0</td>\n",
       "      <td>2876.0</td>\n",
       "      <td>3070.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>add</td>\n",
       "      <td>Adam</td>\n",
       "      <td>8</td>\n",
       "      <td>20190105090505</td>\n",
       "      <td>256</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>True</td>\n",
       "      <td>32</td>\n",
       "      <td>256</td>\n",
       "      <td>cogsci2019</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>[64]</td>\n",
       "      <td>847.0</td>\n",
       "      <td>2045.0</td>\n",
       "      <td>3040.0</td>\n",
       "      <td>3037.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>add</td>\n",
       "      <td>Adam</td>\n",
       "      <td>8</td>\n",
       "      <td>20190105115539</td>\n",
       "      <td>256</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>True</td>\n",
       "      <td>32</td>\n",
       "      <td>256</td>\n",
       "      <td>cogsci2019</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>[64]</td>\n",
       "      <td>1514.0</td>\n",
       "      <td>1945.0</td>\n",
       "      <td>2425.0</td>\n",
       "      <td>2719.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>add</td>\n",
       "      <td>Adam</td>\n",
       "      <td>8</td>\n",
       "      <td>20190105075200</td>\n",
       "      <td>256</td>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 76 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   all_correct_stop  batch_size  dev_set_size experiment_name  \\\n",
       "0              True          32           256      cogsci2019   \n",
       "1              True          32           256      cogsci2019   \n",
       "2              True          32           256      cogsci2019   \n",
       "3              True          32           256      cogsci2019   \n",
       "4              True          32           256      cogsci2019   \n",
       "\n",
       "  hidden_activation hidden_dimensions  init_all_correct_carry-0_epoch  \\\n",
       "0           sigmoid              [64]                          2593.0   \n",
       "1           sigmoid              [64]                          1463.0   \n",
       "2           sigmoid              [64]                          1782.0   \n",
       "3           sigmoid              [64]                           847.0   \n",
       "4           sigmoid              [64]                          1514.0   \n",
       "\n",
       "   init_all_correct_carry-1_epoch  init_all_correct_carry-2_epoch  \\\n",
       "0                          2356.0                          2907.0   \n",
       "1                          1899.0                          3052.0   \n",
       "2                          1858.0                          2876.0   \n",
       "3                          2045.0                          3040.0   \n",
       "4                          1945.0                          2425.0   \n",
       "\n",
       "   init_all_correct_carry-3_epoch       ...        learning_rate  \\\n",
       "0                          2814.0       ...                0.001   \n",
       "1                          3114.0       ...                0.001   \n",
       "2                          3070.0       ...                0.001   \n",
       "3                          3037.0       ...                0.001   \n",
       "4                          2719.0       ...                0.001   \n",
       "\n",
       "   network_input_dimension  network_output_dimension  operand_bits  operator  \\\n",
       "0                        8                         8             4       add   \n",
       "1                        8                         8             4       add   \n",
       "2                        8                         8             4       add   \n",
       "3                        8                         8             4       add   \n",
       "4                        8                         8             4       add   \n",
       "\n",
       "   optimizer  result_bits          run_id  test_set_size  train_set_size  \n",
       "0       Adam            8  20190105091719            256             256  \n",
       "1       Adam            8  20190105140503            256             256  \n",
       "2       Adam            8  20190105090505            256             256  \n",
       "3       Adam            8  20190105115539            256             256  \n",
       "4       Adam            8  20190105075200            256             256  \n",
       "\n",
       "[5 rows x 76 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_add.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "epoch_keys_add = get_epoch_keys(all_run_info_add[0])\n",
    "epoch_keys_subtract = get_epoch_keys(all_run_info_subtract[0])\n",
    "epoch_keys_multiply = get_epoch_keys(all_run_info_multiply[0])\n",
    "epoch_keys_divide = get_epoch_keys(all_run_info_divide[0])\n",
    "epoch_keys_modulo = get_epoch_keys(all_run_info_modulo[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Add"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['init_all_correct_carry-0_epoch',\n",
       " 'init_all_correct_carry-1_epoch',\n",
       " 'init_all_correct_carry-2_epoch',\n",
       " 'init_all_correct_carry-3_epoch',\n",
       " 'init_all_correct_carry-4_epoch',\n",
       " 'init_all_correct_digit-1_epoch',\n",
       " 'init_all_correct_digit-2_epoch',\n",
       " 'init_all_correct_digit-3_epoch',\n",
       " 'init_all_correct_digit-4_epoch',\n",
       " 'init_all_correct_digit-5_epoch',\n",
       " 'init_all_correct_digit-6_epoch',\n",
       " 'init_all_correct_digit-7_epoch',\n",
       " 'init_all_correct_digit-8_epoch',\n",
       " 'init_all_correct_epoch',\n",
       " 'init_complete_all_correct_carry-0_epoch',\n",
       " 'init_complete_all_correct_carry-1_epoch',\n",
       " 'init_complete_all_correct_carry-2_epoch',\n",
       " 'init_complete_all_correct_carry-3_epoch',\n",
       " 'init_complete_all_correct_carry-4_epoch',\n",
       " 'init_complete_all_correct_digit-1_epoch',\n",
       " 'init_complete_all_correct_digit-2_epoch',\n",
       " 'init_complete_all_correct_digit-3_epoch',\n",
       " 'init_complete_all_correct_digit-4_epoch',\n",
       " 'init_complete_all_correct_digit-5_epoch',\n",
       " 'init_complete_all_correct_digit-6_epoch',\n",
       " 'init_complete_all_correct_digit-7_epoch',\n",
       " 'init_complete_all_correct_digit-8_epoch',\n",
       " 'last_epoch']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_keys_add"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>init_all_correct_carry-4_epoch</th>\n",
       "      <th>init_all_correct_digit-1_epoch</th>\n",
       "      <th>init_all_correct_digit-2_epoch</th>\n",
       "      <th>init_all_correct_digit-3_epoch</th>\n",
       "      <th>init_all_correct_digit-4_epoch</th>\n",
       "      <th>init_all_correct_digit-5_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>init_complete_all_correct_carry-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-1_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-2_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-5_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-6_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-7_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-8_epoch</th>\n",
       "      <th>last_epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1522.610000</td>\n",
       "      <td>2189.360000</td>\n",
       "      <td>2856.440000</td>\n",
       "      <td>3005.110000</td>\n",
       "      <td>3361.960000</td>\n",
       "      <td>449.930000</td>\n",
       "      <td>1587.050000</td>\n",
       "      <td>2597.570000</td>\n",
       "      <td>3361.460000</td>\n",
       "      <td>289.520000</td>\n",
       "      <td>...</td>\n",
       "      <td>3364.030000</td>\n",
       "      <td>593.650000</td>\n",
       "      <td>1599.720000</td>\n",
       "      <td>2621.070000</td>\n",
       "      <td>3363.86000</td>\n",
       "      <td>369.540000</td>\n",
       "      <td>1.770000</td>\n",
       "      <td>1.810000</td>\n",
       "      <td>1.730000</td>\n",
       "      <td>3377.110000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>640.697661</td>\n",
       "      <td>335.355006</td>\n",
       "      <td>422.022555</td>\n",
       "      <td>384.627299</td>\n",
       "      <td>441.523028</td>\n",
       "      <td>65.217624</td>\n",
       "      <td>184.089042</td>\n",
       "      <td>298.192577</td>\n",
       "      <td>448.810739</td>\n",
       "      <td>72.266829</td>\n",
       "      <td>...</td>\n",
       "      <td>439.527282</td>\n",
       "      <td>207.749527</td>\n",
       "      <td>184.652066</td>\n",
       "      <td>293.404924</td>\n",
       "      <td>445.32721</td>\n",
       "      <td>80.629771</td>\n",
       "      <td>1.003579</td>\n",
       "      <td>1.021931</td>\n",
       "      <td>1.117853</td>\n",
       "      <td>431.658884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>530.000000</td>\n",
       "      <td>1106.000000</td>\n",
       "      <td>1954.000000</td>\n",
       "      <td>2211.000000</td>\n",
       "      <td>2392.000000</td>\n",
       "      <td>329.000000</td>\n",
       "      <td>1263.000000</td>\n",
       "      <td>1883.000000</td>\n",
       "      <td>2201.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2412.000000</td>\n",
       "      <td>342.000000</td>\n",
       "      <td>1272.000000</td>\n",
       "      <td>1974.000000</td>\n",
       "      <td>2256.00000</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2412.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>925.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>2594.000000</td>\n",
       "      <td>2754.250000</td>\n",
       "      <td>3053.750000</td>\n",
       "      <td>413.500000</td>\n",
       "      <td>1445.750000</td>\n",
       "      <td>2407.500000</td>\n",
       "      <td>3053.750000</td>\n",
       "      <td>244.500000</td>\n",
       "      <td>...</td>\n",
       "      <td>3053.750000</td>\n",
       "      <td>433.250000</td>\n",
       "      <td>1454.000000</td>\n",
       "      <td>2424.500000</td>\n",
       "      <td>3053.75000</td>\n",
       "      <td>303.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3053.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1509.000000</td>\n",
       "      <td>2211.500000</td>\n",
       "      <td>2814.500000</td>\n",
       "      <td>2964.500000</td>\n",
       "      <td>3325.500000</td>\n",
       "      <td>443.000000</td>\n",
       "      <td>1575.000000</td>\n",
       "      <td>2594.500000</td>\n",
       "      <td>3328.500000</td>\n",
       "      <td>268.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3325.500000</td>\n",
       "      <td>478.000000</td>\n",
       "      <td>1594.500000</td>\n",
       "      <td>2609.000000</td>\n",
       "      <td>3328.50000</td>\n",
       "      <td>350.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3336.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1997.750000</td>\n",
       "      <td>2387.500000</td>\n",
       "      <td>3117.750000</td>\n",
       "      <td>3225.750000</td>\n",
       "      <td>3622.250000</td>\n",
       "      <td>477.000000</td>\n",
       "      <td>1694.750000</td>\n",
       "      <td>2764.000000</td>\n",
       "      <td>3631.750000</td>\n",
       "      <td>309.250000</td>\n",
       "      <td>...</td>\n",
       "      <td>3622.250000</td>\n",
       "      <td>808.500000</td>\n",
       "      <td>1701.750000</td>\n",
       "      <td>2783.000000</td>\n",
       "      <td>3631.75000</td>\n",
       "      <td>420.750000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3665.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3058.000000</td>\n",
       "      <td>3068.000000</td>\n",
       "      <td>4255.000000</td>\n",
       "      <td>4238.000000</td>\n",
       "      <td>4593.000000</td>\n",
       "      <td>886.000000</td>\n",
       "      <td>2131.000000</td>\n",
       "      <td>3748.000000</td>\n",
       "      <td>4593.000000</td>\n",
       "      <td>537.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4593.000000</td>\n",
       "      <td>1145.000000</td>\n",
       "      <td>2137.000000</td>\n",
       "      <td>3748.000000</td>\n",
       "      <td>4593.00000</td>\n",
       "      <td>561.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4593.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       init_all_correct_carry-0_epoch  init_all_correct_carry-1_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      1522.610000                     2189.360000   \n",
       "std                        640.697661                      335.355006   \n",
       "min                        530.000000                     1106.000000   \n",
       "25%                        925.000000                     2017.000000   \n",
       "50%                       1509.000000                     2211.500000   \n",
       "75%                       1997.750000                     2387.500000   \n",
       "max                       3058.000000                     3068.000000   \n",
       "\n",
       "       init_all_correct_carry-2_epoch  init_all_correct_carry-3_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2856.440000                     3005.110000   \n",
       "std                        422.022555                      384.627299   \n",
       "min                       1954.000000                     2211.000000   \n",
       "25%                       2594.000000                     2754.250000   \n",
       "50%                       2814.500000                     2964.500000   \n",
       "75%                       3117.750000                     3225.750000   \n",
       "max                       4255.000000                     4238.000000   \n",
       "\n",
       "       init_all_correct_carry-4_epoch  init_all_correct_digit-1_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      3361.960000                      449.930000   \n",
       "std                        441.523028                       65.217624   \n",
       "min                       2392.000000                      329.000000   \n",
       "25%                       3053.750000                      413.500000   \n",
       "50%                       3325.500000                      443.000000   \n",
       "75%                       3622.250000                      477.000000   \n",
       "max                       4593.000000                      886.000000   \n",
       "\n",
       "       init_all_correct_digit-2_epoch  init_all_correct_digit-3_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      1587.050000                     2597.570000   \n",
       "std                        184.089042                      298.192577   \n",
       "min                       1263.000000                     1883.000000   \n",
       "25%                       1445.750000                     2407.500000   \n",
       "50%                       1575.000000                     2594.500000   \n",
       "75%                       1694.750000                     2764.000000   \n",
       "max                       2131.000000                     3748.000000   \n",
       "\n",
       "       init_all_correct_digit-4_epoch  init_all_correct_digit-5_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      3361.460000                      289.520000   \n",
       "std                        448.810739                       72.266829   \n",
       "min                       2201.000000                      202.000000   \n",
       "25%                       3053.750000                      244.500000   \n",
       "50%                       3328.500000                      268.000000   \n",
       "75%                       3631.750000                      309.250000   \n",
       "max                       4593.000000                      537.000000   \n",
       "\n",
       "          ...       init_complete_all_correct_carry-4_epoch  \\\n",
       "count     ...                                    100.000000   \n",
       "mean      ...                                   3364.030000   \n",
       "std       ...                                    439.527282   \n",
       "min       ...                                   2412.000000   \n",
       "25%       ...                                   3053.750000   \n",
       "50%       ...                                   3325.500000   \n",
       "75%       ...                                   3622.250000   \n",
       "max       ...                                   4593.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-1_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                593.650000   \n",
       "std                                 207.749527   \n",
       "min                                 342.000000   \n",
       "25%                                 433.250000   \n",
       "50%                                 478.000000   \n",
       "75%                                 808.500000   \n",
       "max                                1145.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-2_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               1599.720000   \n",
       "std                                 184.652066   \n",
       "min                                1272.000000   \n",
       "25%                                1454.000000   \n",
       "50%                                1594.500000   \n",
       "75%                                1701.750000   \n",
       "max                                2137.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-3_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               2621.070000   \n",
       "std                                 293.404924   \n",
       "min                                1974.000000   \n",
       "25%                                2424.500000   \n",
       "50%                                2609.000000   \n",
       "75%                                2783.000000   \n",
       "max                                3748.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-4_epoch  \\\n",
       "count                                100.00000   \n",
       "mean                                3363.86000   \n",
       "std                                  445.32721   \n",
       "min                                 2256.00000   \n",
       "25%                                 3053.75000   \n",
       "50%                                 3328.50000   \n",
       "75%                                 3631.75000   \n",
       "max                                 4593.00000   \n",
       "\n",
       "       init_complete_all_correct_digit-5_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                369.540000   \n",
       "std                                  80.629771   \n",
       "min                                 255.000000   \n",
       "25%                                 303.000000   \n",
       "50%                                 350.500000   \n",
       "75%                                 420.750000   \n",
       "max                                 561.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-6_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.770000   \n",
       "std                                   1.003579   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   2.000000   \n",
       "max                                   6.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-7_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.810000   \n",
       "std                                   1.021931   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   2.000000   \n",
       "max                                   4.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-8_epoch   last_epoch  \n",
       "count                               100.000000   100.000000  \n",
       "mean                                  1.730000  3377.110000  \n",
       "std                                   1.117853   431.658884  \n",
       "min                                   1.000000  2412.000000  \n",
       "25%                                   1.000000  3053.750000  \n",
       "50%                                   1.000000  3336.000000  \n",
       "75%                                   2.000000  3665.750000  \n",
       "max                                   5.000000  4593.000000  \n",
       "\n",
       "[8 rows x 28 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_add[epoch_keys_add].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_add_plot = df_run_info_add[epoch_keys_add].rename(index=str, columns={'init_all_correct_epoch':'all_correct',\n",
    "                                                                                 '{}_carry-0_epoch'.format(all_correct_case):'carry-0',\n",
    "                                                                                 '{}_carry-1_epoch'.format(all_correct_case):'carry-1',\n",
    "                                                                                 '{}_carry-2_epoch'.format(all_correct_case):'carry-2',\n",
    "                                                                                 '{}_carry-3_epoch'.format(all_correct_case):'carry-3',\n",
    "                                                                                 '{}_carry-4_epoch'.format(all_correct_case):'carry-4',\n",
    "                                                                                 '{}_digit-1_epoch'.format(all_correct_case):'digit-1',\n",
    "                                                                                 '{}_digit-2_epoch'.format(all_correct_case):'digit-2',\n",
    "                                                                                 '{}_digit-3_epoch'.format(all_correct_case):'digit-3',\n",
    "                                                                                 '{}_digit-4_epoch'.format(all_correct_case):'digit-4',\n",
    "                                                                                 '{}_digit-5_epoch'.format(all_correct_case):'digit-5'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "column_list = ['carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "df_run_info_add_plot.boxplot(column=column_list)\n",
    "df_run_info_add_plot[column_list].to_csv('result_statistics/epoch_add_carry.csv', index=False)\n",
    "save_fig('epoch_add_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "column_list = ['carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "df_run_info_add_plot[column_list]\n",
    "df_run_info_add_plot.boxplot(column=column_list)\n",
    "\n",
    "df_run_info_add_plot[column_list].to_csv('result_statistics/epoch_add_carry.csv', index=False)\n",
    "save_fig('epoch_add_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "column_list = ['carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4']\n",
    "\n",
    "x = ['0', '1', '2', '3', '4']\n",
    "y = df_run_info_add_plot[column_list].mean()\n",
    "e = df_run_info_add_plot[column_list].std()\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.xlabel('Carries', fontsize=font_size['xlabel'])\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_add_carry')\n",
    "\n",
    "df_plot = df_run_info_add_plot[column_list]\n",
    "df_plot.columns = x\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_add_carry.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_add_plot.boxplot(column=['all_correct', 'digit-1', 'digit-2', 'digit-3', 'digit-4', 'digit-5'])\n",
    "save_fig('epoch_add_digit')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Subtract"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['init_all_correct_carry-0_epoch',\n",
       " 'init_all_correct_carry-1_epoch',\n",
       " 'init_all_correct_carry-2_epoch',\n",
       " 'init_all_correct_carry-3_epoch',\n",
       " 'init_all_correct_digit-1_epoch',\n",
       " 'init_all_correct_digit-2_epoch',\n",
       " 'init_all_correct_digit-3_epoch',\n",
       " 'init_all_correct_digit-4_epoch',\n",
       " 'init_all_correct_digit-5_epoch',\n",
       " 'init_all_correct_digit-6_epoch',\n",
       " 'init_all_correct_digit-7_epoch',\n",
       " 'init_all_correct_digit-8_epoch',\n",
       " 'init_all_correct_epoch',\n",
       " 'init_complete_all_correct_carry-0_epoch',\n",
       " 'init_complete_all_correct_carry-1_epoch',\n",
       " 'init_complete_all_correct_carry-2_epoch',\n",
       " 'init_complete_all_correct_carry-3_epoch',\n",
       " 'init_complete_all_correct_digit-1_epoch',\n",
       " 'init_complete_all_correct_digit-2_epoch',\n",
       " 'init_complete_all_correct_digit-3_epoch',\n",
       " 'init_complete_all_correct_digit-4_epoch',\n",
       " 'init_complete_all_correct_digit-5_epoch',\n",
       " 'init_complete_all_correct_digit-6_epoch',\n",
       " 'init_complete_all_correct_digit-7_epoch',\n",
       " 'init_complete_all_correct_digit-8_epoch',\n",
       " 'last_epoch']"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_keys_subtract"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>init_all_correct_digit-1_epoch</th>\n",
       "      <th>init_all_correct_digit-2_epoch</th>\n",
       "      <th>init_all_correct_digit-3_epoch</th>\n",
       "      <th>init_all_correct_digit-4_epoch</th>\n",
       "      <th>init_all_correct_digit-5_epoch</th>\n",
       "      <th>init_all_correct_digit-6_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>init_complete_all_correct_carry-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-1_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-2_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-5_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-6_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-7_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-8_epoch</th>\n",
       "      <th>last_epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.0000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2364.13000</td>\n",
       "      <td>2322.950000</td>\n",
       "      <td>2685.620000</td>\n",
       "      <td>2648.300000</td>\n",
       "      <td>1041.290000</td>\n",
       "      <td>2563.650000</td>\n",
       "      <td>2557.320000</td>\n",
       "      <td>667.450000</td>\n",
       "      <td>2.560000</td>\n",
       "      <td>2.650000</td>\n",
       "      <td>...</td>\n",
       "      <td>2677.330000</td>\n",
       "      <td>1123.130000</td>\n",
       "      <td>2573.390000</td>\n",
       "      <td>2590.1800</td>\n",
       "      <td>729.850000</td>\n",
       "      <td>2.560000</td>\n",
       "      <td>2.650000</td>\n",
       "      <td>2.810000</td>\n",
       "      <td>2.860000</td>\n",
       "      <td>2748.080000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>260.36752</td>\n",
       "      <td>212.003925</td>\n",
       "      <td>271.618299</td>\n",
       "      <td>251.249584</td>\n",
       "      <td>91.600725</td>\n",
       "      <td>368.416578</td>\n",
       "      <td>223.991472</td>\n",
       "      <td>82.396599</td>\n",
       "      <td>2.109718</td>\n",
       "      <td>2.203647</td>\n",
       "      <td>...</td>\n",
       "      <td>238.468263</td>\n",
       "      <td>294.707136</td>\n",
       "      <td>358.915366</td>\n",
       "      <td>209.4946</td>\n",
       "      <td>99.509948</td>\n",
       "      <td>2.109718</td>\n",
       "      <td>2.203647</td>\n",
       "      <td>2.236949</td>\n",
       "      <td>2.165011</td>\n",
       "      <td>245.501411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1665.00000</td>\n",
       "      <td>1819.000000</td>\n",
       "      <td>2161.000000</td>\n",
       "      <td>1986.000000</td>\n",
       "      <td>852.000000</td>\n",
       "      <td>1791.000000</td>\n",
       "      <td>1984.000000</td>\n",
       "      <td>489.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2011.000000</td>\n",
       "      <td>874.000000</td>\n",
       "      <td>1810.000000</td>\n",
       "      <td>2035.0000</td>\n",
       "      <td>551.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2221.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2199.50000</td>\n",
       "      <td>2150.750000</td>\n",
       "      <td>2468.750000</td>\n",
       "      <td>2462.750000</td>\n",
       "      <td>983.000000</td>\n",
       "      <td>2310.250000</td>\n",
       "      <td>2419.500000</td>\n",
       "      <td>606.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2495.000000</td>\n",
       "      <td>989.000000</td>\n",
       "      <td>2326.250000</td>\n",
       "      <td>2466.7500</td>\n",
       "      <td>652.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2579.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2356.50000</td>\n",
       "      <td>2301.000000</td>\n",
       "      <td>2625.500000</td>\n",
       "      <td>2626.000000</td>\n",
       "      <td>1035.500000</td>\n",
       "      <td>2551.000000</td>\n",
       "      <td>2575.500000</td>\n",
       "      <td>662.500000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2644.500000</td>\n",
       "      <td>1045.000000</td>\n",
       "      <td>2565.000000</td>\n",
       "      <td>2598.5000</td>\n",
       "      <td>720.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2713.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2518.00000</td>\n",
       "      <td>2493.000000</td>\n",
       "      <td>2907.750000</td>\n",
       "      <td>2832.250000</td>\n",
       "      <td>1101.250000</td>\n",
       "      <td>2835.750000</td>\n",
       "      <td>2690.250000</td>\n",
       "      <td>708.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2860.750000</td>\n",
       "      <td>1121.750000</td>\n",
       "      <td>2838.750000</td>\n",
       "      <td>2715.2500</td>\n",
       "      <td>795.250000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2958.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3070.00000</td>\n",
       "      <td>2770.000000</td>\n",
       "      <td>3381.000000</td>\n",
       "      <td>3230.000000</td>\n",
       "      <td>1344.000000</td>\n",
       "      <td>3381.000000</td>\n",
       "      <td>3195.000000</td>\n",
       "      <td>998.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3248.000000</td>\n",
       "      <td>2539.000000</td>\n",
       "      <td>3381.000000</td>\n",
       "      <td>3195.0000</td>\n",
       "      <td>1060.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>3381.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       init_all_correct_carry-0_epoch  init_all_correct_carry-1_epoch  \\\n",
       "count                       100.00000                      100.000000   \n",
       "mean                       2364.13000                     2322.950000   \n",
       "std                         260.36752                      212.003925   \n",
       "min                        1665.00000                     1819.000000   \n",
       "25%                        2199.50000                     2150.750000   \n",
       "50%                        2356.50000                     2301.000000   \n",
       "75%                        2518.00000                     2493.000000   \n",
       "max                        3070.00000                     2770.000000   \n",
       "\n",
       "       init_all_correct_carry-2_epoch  init_all_correct_carry-3_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2685.620000                     2648.300000   \n",
       "std                        271.618299                      251.249584   \n",
       "min                       2161.000000                     1986.000000   \n",
       "25%                       2468.750000                     2462.750000   \n",
       "50%                       2625.500000                     2626.000000   \n",
       "75%                       2907.750000                     2832.250000   \n",
       "max                       3381.000000                     3230.000000   \n",
       "\n",
       "       init_all_correct_digit-1_epoch  init_all_correct_digit-2_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      1041.290000                     2563.650000   \n",
       "std                         91.600725                      368.416578   \n",
       "min                        852.000000                     1791.000000   \n",
       "25%                        983.000000                     2310.250000   \n",
       "50%                       1035.500000                     2551.000000   \n",
       "75%                       1101.250000                     2835.750000   \n",
       "max                       1344.000000                     3381.000000   \n",
       "\n",
       "       init_all_correct_digit-3_epoch  init_all_correct_digit-4_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2557.320000                      667.450000   \n",
       "std                        223.991472                       82.396599   \n",
       "min                       1984.000000                      489.000000   \n",
       "25%                       2419.500000                      606.750000   \n",
       "50%                       2575.500000                      662.500000   \n",
       "75%                       2690.250000                      708.000000   \n",
       "max                       3195.000000                      998.000000   \n",
       "\n",
       "       init_all_correct_digit-5_epoch  init_all_correct_digit-6_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                         2.560000                        2.650000   \n",
       "std                          2.109718                        2.203647   \n",
       "min                          1.000000                        1.000000   \n",
       "25%                          1.000000                        1.000000   \n",
       "50%                          2.000000                        1.000000   \n",
       "75%                          4.000000                        4.000000   \n",
       "max                         12.000000                       10.000000   \n",
       "\n",
       "          ...       init_complete_all_correct_carry-3_epoch  \\\n",
       "count     ...                                    100.000000   \n",
       "mean      ...                                   2677.330000   \n",
       "std       ...                                    238.468263   \n",
       "min       ...                                   2011.000000   \n",
       "25%       ...                                   2495.000000   \n",
       "50%       ...                                   2644.500000   \n",
       "75%       ...                                   2860.750000   \n",
       "max       ...                                   3248.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-1_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               1123.130000   \n",
       "std                                 294.707136   \n",
       "min                                 874.000000   \n",
       "25%                                 989.000000   \n",
       "50%                                1045.000000   \n",
       "75%                                1121.750000   \n",
       "max                                2539.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-2_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               2573.390000   \n",
       "std                                 358.915366   \n",
       "min                                1810.000000   \n",
       "25%                                2326.250000   \n",
       "50%                                2565.000000   \n",
       "75%                                2838.750000   \n",
       "max                                3381.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-3_epoch  \\\n",
       "count                                 100.0000   \n",
       "mean                                 2590.1800   \n",
       "std                                   209.4946   \n",
       "min                                  2035.0000   \n",
       "25%                                  2466.7500   \n",
       "50%                                  2598.5000   \n",
       "75%                                  2715.2500   \n",
       "max                                  3195.0000   \n",
       "\n",
       "       init_complete_all_correct_digit-4_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                729.850000   \n",
       "std                                  99.509948   \n",
       "min                                 551.000000   \n",
       "25%                                 652.500000   \n",
       "50%                                 720.000000   \n",
       "75%                                 795.250000   \n",
       "max                                1060.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-5_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  2.560000   \n",
       "std                                   2.109718   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   2.000000   \n",
       "75%                                   4.000000   \n",
       "max                                  12.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-6_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  2.650000   \n",
       "std                                   2.203647   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   4.000000   \n",
       "max                                  10.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-7_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  2.810000   \n",
       "std                                   2.236949   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   2.000000   \n",
       "75%                                   4.000000   \n",
       "max                                  10.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-8_epoch   last_epoch  \n",
       "count                               100.000000   100.000000  \n",
       "mean                                  2.860000  2748.080000  \n",
       "std                                   2.165011   245.501411  \n",
       "min                                   1.000000  2221.000000  \n",
       "25%                                   1.000000  2579.250000  \n",
       "50%                                   2.000000  2713.000000  \n",
       "75%                                   4.000000  2958.500000  \n",
       "max                                  10.000000  3381.000000  \n",
       "\n",
       "[8 rows x 26 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_subtract[epoch_keys_subtract].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_subtract_plot = df_run_info_subtract[epoch_keys_subtract].rename(index=str, columns={'init_all_correct_epoch':'all_correct',\n",
    "                                                                                             '{}_carry-0_epoch'.format(all_correct_case):'carry-0',\n",
    "                                                                                             '{}_carry-1_epoch'.format(all_correct_case):'carry-1',\n",
    "                                                                                             '{}_carry-2_epoch'.format(all_correct_case):'carry-2',\n",
    "                                                                                             '{}_carry-3_epoch'.format(all_correct_case):'carry-3',\n",
    "                                                                                             '{}_digit-1_epoch'.format(all_correct_case):'digit-1',\n",
    "                                                                                             '{}_digit-2_epoch'.format(all_correct_case):'digit-2',\n",
    "                                                                                             '{}_digit-3_epoch'.format(all_correct_case):'digit-3',\n",
    "                                                                                             '{}_digit-4_epoch'.format(all_correct_case):'digit-4'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Epochs')\n",
    "df_run_info_subtract_plot.boxplot(column=['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3'])\n",
    "df_run_info_subtract_plot[['carry-0', 'carry-1', 'carry-2', 'carry-3']].to_csv('result_statistics/epoch_subtract_carry.csv', index=False)\n",
    "save_fig('epoch_subtract_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "column_list = ['carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "\n",
    "x = ['0', '1', '2', '3']\n",
    "y = df_run_info_subtract_plot[column_list].mean()\n",
    "e = df_run_info_subtract_plot[column_list].std()\n",
    "\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.xlabel('Carries', fontsize=font_size['xlabel'])\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_subtract_carry')\n",
    "\n",
    "df_plot = df_run_info_subtract_plot[column_list]\n",
    "df_plot.columns = x\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_subtract_carry.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_subtract_plot.boxplot(column=['all_correct', 'digit-1', 'digit-2', 'digit-3', 'digit-4'])\n",
    "save_fig('epoch_subtract_digit')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multiply"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['init_all_correct_carry-0_epoch',\n",
       " 'init_all_correct_carry-12_epoch',\n",
       " 'init_all_correct_carry-1_epoch',\n",
       " 'init_all_correct_carry-2_epoch',\n",
       " 'init_all_correct_carry-3_epoch',\n",
       " 'init_all_correct_carry-4_epoch',\n",
       " 'init_all_correct_carry-5_epoch',\n",
       " 'init_all_correct_carry-6_epoch',\n",
       " 'init_all_correct_carry-8_epoch',\n",
       " 'init_all_correct_digit-1_epoch',\n",
       " 'init_all_correct_digit-2_epoch',\n",
       " 'init_all_correct_digit-3_epoch',\n",
       " 'init_all_correct_digit-4_epoch',\n",
       " 'init_all_correct_digit-5_epoch',\n",
       " 'init_all_correct_digit-6_epoch',\n",
       " 'init_all_correct_digit-7_epoch',\n",
       " 'init_all_correct_digit-8_epoch',\n",
       " 'init_all_correct_epoch',\n",
       " 'init_complete_all_correct_carry-0_epoch',\n",
       " 'init_complete_all_correct_carry-12_epoch',\n",
       " 'init_complete_all_correct_carry-1_epoch',\n",
       " 'init_complete_all_correct_carry-2_epoch',\n",
       " 'init_complete_all_correct_carry-3_epoch',\n",
       " 'init_complete_all_correct_carry-4_epoch',\n",
       " 'init_complete_all_correct_carry-5_epoch',\n",
       " 'init_complete_all_correct_carry-6_epoch',\n",
       " 'init_complete_all_correct_carry-8_epoch',\n",
       " 'init_complete_all_correct_digit-1_epoch',\n",
       " 'init_complete_all_correct_digit-2_epoch',\n",
       " 'init_complete_all_correct_digit-3_epoch',\n",
       " 'init_complete_all_correct_digit-4_epoch',\n",
       " 'init_complete_all_correct_digit-5_epoch',\n",
       " 'init_complete_all_correct_digit-6_epoch',\n",
       " 'init_complete_all_correct_digit-7_epoch',\n",
       " 'init_complete_all_correct_digit-8_epoch',\n",
       " 'last_epoch']"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_keys_multiply"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-12_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>init_all_correct_carry-4_epoch</th>\n",
       "      <th>init_all_correct_carry-5_epoch</th>\n",
       "      <th>init_all_correct_carry-6_epoch</th>\n",
       "      <th>init_all_correct_carry-8_epoch</th>\n",
       "      <th>init_all_correct_digit-1_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>init_complete_all_correct_carry-8_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-1_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-2_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-5_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-6_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-7_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-8_epoch</th>\n",
       "      <th>last_epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4556.030000</td>\n",
       "      <td>1449.910000</td>\n",
       "      <td>5630.980000</td>\n",
       "      <td>5324.210000</td>\n",
       "      <td>4969.07000</td>\n",
       "      <td>6033.210000</td>\n",
       "      <td>4729.870000</td>\n",
       "      <td>3988.070000</td>\n",
       "      <td>2608.390000</td>\n",
       "      <td>126.310000</td>\n",
       "      <td>...</td>\n",
       "      <td>3396.050000</td>\n",
       "      <td>126.910000</td>\n",
       "      <td>1295.130000</td>\n",
       "      <td>2215.840000</td>\n",
       "      <td>6087.900000</td>\n",
       "      <td>5569.960000</td>\n",
       "      <td>4725.650000</td>\n",
       "      <td>2330.410000</td>\n",
       "      <td>1242.21000</td>\n",
       "      <td>6127.650000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>613.664493</td>\n",
       "      <td>676.075366</td>\n",
       "      <td>416.650714</td>\n",
       "      <td>406.629986</td>\n",
       "      <td>447.59391</td>\n",
       "      <td>494.276745</td>\n",
       "      <td>472.497355</td>\n",
       "      <td>743.729494</td>\n",
       "      <td>1174.449872</td>\n",
       "      <td>22.023722</td>\n",
       "      <td>...</td>\n",
       "      <td>1116.443337</td>\n",
       "      <td>22.320935</td>\n",
       "      <td>138.407731</td>\n",
       "      <td>162.684594</td>\n",
       "      <td>456.238496</td>\n",
       "      <td>419.960266</td>\n",
       "      <td>393.394073</td>\n",
       "      <td>477.354158</td>\n",
       "      <td>287.54363</td>\n",
       "      <td>436.063552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3381.000000</td>\n",
       "      <td>1035.000000</td>\n",
       "      <td>4458.000000</td>\n",
       "      <td>4510.000000</td>\n",
       "      <td>4036.00000</td>\n",
       "      <td>4843.000000</td>\n",
       "      <td>3661.000000</td>\n",
       "      <td>2328.000000</td>\n",
       "      <td>1392.000000</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1402.000000</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>1035.000000</td>\n",
       "      <td>1853.000000</td>\n",
       "      <td>5113.000000</td>\n",
       "      <td>4690.000000</td>\n",
       "      <td>3671.000000</td>\n",
       "      <td>1634.000000</td>\n",
       "      <td>658.00000</td>\n",
       "      <td>5232.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4106.750000</td>\n",
       "      <td>1204.250000</td>\n",
       "      <td>5368.750000</td>\n",
       "      <td>5062.750000</td>\n",
       "      <td>4645.25000</td>\n",
       "      <td>5675.000000</td>\n",
       "      <td>4402.500000</td>\n",
       "      <td>3409.750000</td>\n",
       "      <td>1705.250000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2549.250000</td>\n",
       "      <td>113.750000</td>\n",
       "      <td>1198.500000</td>\n",
       "      <td>2094.750000</td>\n",
       "      <td>5811.500000</td>\n",
       "      <td>5290.750000</td>\n",
       "      <td>4455.250000</td>\n",
       "      <td>1979.250000</td>\n",
       "      <td>1047.50000</td>\n",
       "      <td>5833.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4556.500000</td>\n",
       "      <td>1294.500000</td>\n",
       "      <td>5644.500000</td>\n",
       "      <td>5321.000000</td>\n",
       "      <td>4926.00000</td>\n",
       "      <td>6007.500000</td>\n",
       "      <td>4700.500000</td>\n",
       "      <td>3898.000000</td>\n",
       "      <td>1908.500000</td>\n",
       "      <td>125.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3429.500000</td>\n",
       "      <td>125.500000</td>\n",
       "      <td>1288.500000</td>\n",
       "      <td>2207.000000</td>\n",
       "      <td>6055.500000</td>\n",
       "      <td>5534.500000</td>\n",
       "      <td>4756.000000</td>\n",
       "      <td>2262.000000</td>\n",
       "      <td>1223.00000</td>\n",
       "      <td>6084.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4977.250000</td>\n",
       "      <td>1374.250000</td>\n",
       "      <td>5854.250000</td>\n",
       "      <td>5583.000000</td>\n",
       "      <td>5272.50000</td>\n",
       "      <td>6338.250000</td>\n",
       "      <td>5060.000000</td>\n",
       "      <td>4429.750000</td>\n",
       "      <td>3529.750000</td>\n",
       "      <td>139.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4183.750000</td>\n",
       "      <td>139.250000</td>\n",
       "      <td>1374.750000</td>\n",
       "      <td>2314.000000</td>\n",
       "      <td>6377.750000</td>\n",
       "      <td>5754.750000</td>\n",
       "      <td>4996.250000</td>\n",
       "      <td>2564.250000</td>\n",
       "      <td>1425.50000</td>\n",
       "      <td>6396.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6292.000000</td>\n",
       "      <td>5451.000000</td>\n",
       "      <td>7009.000000</td>\n",
       "      <td>6424.000000</td>\n",
       "      <td>6132.00000</td>\n",
       "      <td>7324.000000</td>\n",
       "      <td>5880.000000</td>\n",
       "      <td>6633.000000</td>\n",
       "      <td>5376.000000</td>\n",
       "      <td>181.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>5427.000000</td>\n",
       "      <td>182.000000</td>\n",
       "      <td>1630.000000</td>\n",
       "      <td>2804.000000</td>\n",
       "      <td>7076.000000</td>\n",
       "      <td>7324.000000</td>\n",
       "      <td>5592.000000</td>\n",
       "      <td>4167.000000</td>\n",
       "      <td>2078.00000</td>\n",
       "      <td>7324.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       init_all_correct_carry-0_epoch  init_all_correct_carry-12_epoch  \\\n",
       "count                      100.000000                       100.000000   \n",
       "mean                      4556.030000                      1449.910000   \n",
       "std                        613.664493                       676.075366   \n",
       "min                       3381.000000                      1035.000000   \n",
       "25%                       4106.750000                      1204.250000   \n",
       "50%                       4556.500000                      1294.500000   \n",
       "75%                       4977.250000                      1374.250000   \n",
       "max                       6292.000000                      5451.000000   \n",
       "\n",
       "       init_all_correct_carry-1_epoch  init_all_correct_carry-2_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      5630.980000                     5324.210000   \n",
       "std                        416.650714                      406.629986   \n",
       "min                       4458.000000                     4510.000000   \n",
       "25%                       5368.750000                     5062.750000   \n",
       "50%                       5644.500000                     5321.000000   \n",
       "75%                       5854.250000                     5583.000000   \n",
       "max                       7009.000000                     6424.000000   \n",
       "\n",
       "       init_all_correct_carry-3_epoch  init_all_correct_carry-4_epoch  \\\n",
       "count                       100.00000                      100.000000   \n",
       "mean                       4969.07000                     6033.210000   \n",
       "std                         447.59391                      494.276745   \n",
       "min                        4036.00000                     4843.000000   \n",
       "25%                        4645.25000                     5675.000000   \n",
       "50%                        4926.00000                     6007.500000   \n",
       "75%                        5272.50000                     6338.250000   \n",
       "max                        6132.00000                     7324.000000   \n",
       "\n",
       "       init_all_correct_carry-5_epoch  init_all_correct_carry-6_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      4729.870000                     3988.070000   \n",
       "std                        472.497355                      743.729494   \n",
       "min                       3661.000000                     2328.000000   \n",
       "25%                       4402.500000                     3409.750000   \n",
       "50%                       4700.500000                     3898.000000   \n",
       "75%                       5060.000000                     4429.750000   \n",
       "max                       5880.000000                     6633.000000   \n",
       "\n",
       "       init_all_correct_carry-8_epoch  init_all_correct_digit-1_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2608.390000                      126.310000   \n",
       "std                       1174.449872                       22.023722   \n",
       "min                       1392.000000                       79.000000   \n",
       "25%                       1705.250000                      113.000000   \n",
       "50%                       1908.500000                      125.000000   \n",
       "75%                       3529.750000                      139.000000   \n",
       "max                       5376.000000                      181.000000   \n",
       "\n",
       "          ...       init_complete_all_correct_carry-8_epoch  \\\n",
       "count     ...                                    100.000000   \n",
       "mean      ...                                   3396.050000   \n",
       "std       ...                                   1116.443337   \n",
       "min       ...                                   1402.000000   \n",
       "25%       ...                                   2549.250000   \n",
       "50%       ...                                   3429.500000   \n",
       "75%       ...                                   4183.750000   \n",
       "max       ...                                   5427.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-1_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                126.910000   \n",
       "std                                  22.320935   \n",
       "min                                  79.000000   \n",
       "25%                                 113.750000   \n",
       "50%                                 125.500000   \n",
       "75%                                 139.250000   \n",
       "max                                 182.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-2_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               1295.130000   \n",
       "std                                 138.407731   \n",
       "min                                1035.000000   \n",
       "25%                                1198.500000   \n",
       "50%                                1288.500000   \n",
       "75%                                1374.750000   \n",
       "max                                1630.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-3_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               2215.840000   \n",
       "std                                 162.684594   \n",
       "min                                1853.000000   \n",
       "25%                                2094.750000   \n",
       "50%                                2207.000000   \n",
       "75%                                2314.000000   \n",
       "max                                2804.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-4_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               6087.900000   \n",
       "std                                 456.238496   \n",
       "min                                5113.000000   \n",
       "25%                                5811.500000   \n",
       "50%                                6055.500000   \n",
       "75%                                6377.750000   \n",
       "max                                7076.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-5_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               5569.960000   \n",
       "std                                 419.960266   \n",
       "min                                4690.000000   \n",
       "25%                                5290.750000   \n",
       "50%                                5534.500000   \n",
       "75%                                5754.750000   \n",
       "max                                7324.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-6_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               4725.650000   \n",
       "std                                 393.394073   \n",
       "min                                3671.000000   \n",
       "25%                                4455.250000   \n",
       "50%                                4756.000000   \n",
       "75%                                4996.250000   \n",
       "max                                5592.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-7_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               2330.410000   \n",
       "std                                 477.354158   \n",
       "min                                1634.000000   \n",
       "25%                                1979.250000   \n",
       "50%                                2262.000000   \n",
       "75%                                2564.250000   \n",
       "max                                4167.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-8_epoch   last_epoch  \n",
       "count                                100.00000   100.000000  \n",
       "mean                                1242.21000  6127.650000  \n",
       "std                                  287.54363   436.063552  \n",
       "min                                  658.00000  5232.000000  \n",
       "25%                                 1047.50000  5833.000000  \n",
       "50%                                 1223.00000  6084.000000  \n",
       "75%                                 1425.50000  6396.250000  \n",
       "max                                 2078.00000  7324.000000  \n",
       "\n",
       "[8 rows x 36 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_multiply[epoch_keys_multiply].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_multiply_plot = df_run_info_multiply[epoch_keys_multiply].rename(index=str, columns={'init_all_correct_epoch':'all_correct',\n",
    "                                                                                                '{}_carry-0_epoch'.format(all_correct_case):'carry-0',\n",
    "                                                                                                '{}_carry-1_epoch'.format(all_correct_case):'carry-1',\n",
    "                                                                                                '{}_carry-2_epoch'.format(all_correct_case):'carry-2',\n",
    "                                                                                                '{}_carry-3_epoch'.format(all_correct_case):'carry-3',\n",
    "                                                                                                '{}_carry-4_epoch'.format(all_correct_case):'carry-4',\n",
    "                                                                                                '{}_carry-5_epoch'.format(all_correct_case):'carry-5',\n",
    "                                                                                                '{}_carry-6_epoch'.format(all_correct_case):'carry-6',\n",
    "                                                                                                '{}_carry-8_epoch'.format(all_correct_case):'carry-8',\n",
    "                                                                                                '{}_carry-12_epoch'.format(all_correct_case):'carry-12',\n",
    "                                                                                                '{}_digit-1_epoch'.format(all_correct_case):'digit-1',\n",
    "                                                                                                '{}_digit-2_epoch'.format(all_correct_case):'digit-2',\n",
    "                                                                                                '{}_digit-3_epoch'.format(all_correct_case):'digit-3',\n",
    "                                                                                                '{}_digit-4_epoch'.format(all_correct_case):'digit-4',\n",
    "                                                                                                '{}_digit-5_epoch'.format(all_correct_case):'digit-5',\n",
    "                                                                                                '{}_digit-6_epoch'.format(all_correct_case):'digit-6',\n",
    "                                                                                                '{}_digit-7_epoch'.format(all_correct_case):'digit-7',\n",
    "                                                                                                '{}_digit-8_epoch'.format(all_correct_case):'digit-8'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf4AAAEeCAYAAABrK3enAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X2cXGV9/vHPlQQEifLsgokQVMSFKJZEQV1tliiCiKg/oS5qA11FW4xaiya42sjDauiDVmtFkUWDyqaIUiIgmKa71W1VIIoYWC2RQgER1EAkPJmE7++Pc0+YLLubM8memTk71/v1mtfM3HPmzLWTzX7Puc997qOIwMzMzFrDlEYHMDMzs/px4TczM2shLvxmZmYtxIXfzMyshbjwm5mZtRAXfjMzsxbiwm9mZtZCXPjNzMxaiAu/mZlZC5nW6ABF2GeffWLWrFmFrPvhhx9mt912K2TdRShbXnDmeihbXihf5rLlBWeuhyLzrl69+ncRse82F4yISXebM2dOFGVgYKCwdRehbHkjnLkeypY3onyZy5Y3wpnroci8wI2Ro0a6q9/MzKyFuPCbmZm1EBd+MzOzFuLCb2Zm1kJc+M3MzFqIC39O/f39zJ49m/nz5zN79mz6+/sbHcnMzKxmk/I8/onW399PT08PfX19bN68malTp9Ld3Q1AV1dXg9OZmZnl5z3+HHp7e+nr66Ozs5Np06bR2dlJX18fvb29jY5mZmZWE+/x5zA8PExHR8dWbR0dHQwPDzcokdnkJamm5bN5S8wsL+/x59De3s7Q0NBWbUNDQ7S3tzcokdnkNdZsYwcuumrUdjOrjQt/Dj09PXR3dzMwMMCmTZsYGBigu7ubnp6eRkczMzOribv6c6gM4Fu4cCHDw8O0t7fT29vrgX1mZlY6Lvw5dXV10dXVxeDgIPPmzWt0HDMzs+3irn4zM7MW4sJvZmbWQgor/JIOkXRT1e0Pkj4oaS9JKyXdlu73TMtL0uckrZV0s6Qjqta1IC1/m6QFRWUekX/UW2dn56jttuM8O6KZWfEKO8YfEb8EXgIgaSpwD3AFsBhYFRFLJS1OzxcBxwEHp9uRwAXAkZL2ApYAc4EAVktaEREPFJU95R+1fdbiq7lj6fFFfnRL8uyIZmb1Ua+u/vnAryLiTuBEYFlqXwa8KT0+EbgkMj8C9pC0P/A6YGVErEvFfiVwbJ1yW514dkQzs/qo16j+twGVftu2iLg3Pf4N0JYezwDuqnrP3altrPatSDodOB2gra2NwcHBicr+FEWue6Jt2LChFHmHh4fZvHkzg4ODWzJv3ryZ4eHhUuQvy/dcUba8FWXKXMbv2JmL1wx5Cy/8knYG3gicNfK1iAhJEzL1VkRcCFwIMHfu3CjslLtrry7V6XxlOf2wvb2dqVOnMm/evC2ZBwYGaG9vL0X+snzPFWXLC/j/Xh04c/GaIW89uvqPA34SEfel5/elLnzS/f2p/R7gOVXvm5naxmq3ScSzI5qZ1Uc9uvq7eLKbH2AFsABYmu6vrGp/n6TlZIP71kfEvZKuAz5ZGf0PHMMovQdWbp4d0cysPgot/JJ2A14LvKeqeSlwmaRu4E7g5NR+DfB6YC3wCHAaQESsk3QucENa7pyIWFdkbmsMz45oZla8Qgt/RDwM7D2i7fdko/xHLhvAGWOs52Lg4iIympmZtRLP3GdmZtZCXPjNzMxaiAu/mVmL83TZrcWX5TUza2GeLrv1eI/fzKyFebrs1uPCb2bWwoaHh+no6NiqraOjg+Hh4QYlsqK58JuZtbD29naGhoa2ahsaGqK9vb1BiaxoLvxmZi3M02W3Hg/uMzNrYZ4uu/V4j98aRtKot87OzjFfazY+Dcomg66uLtasWcOqVatYs2aNi/4k5z1+a5hsluanmrX4au5Yenyd09TOp0GZWRl5j99sO/k0KDMrIxd+s+3k06DMrIxavqv/8LO/x/pHN9b0nlmLr8613O677sTPlhyzPbF2WH9/P729vVsG6/T09Lj7eYJVToPq7Ozc0ubToMys2bV84V//6MaajifXcq34vBsIE83HnuujchpU5XuunAblrn4za2YtX/gno+pjz5UNlb6+PhYuXOjCP4F8GtSOq7XHrQy9bWbNzoV/EvKx5+KMd0rhLbfcwimnnMIpp5yypW2sMxcsU0uPWxl628zKwIP7JiFPwVmciBj1duCiq0ZtNzNrNi78k5Cn4DQzs7G4q38S8rFnMzMbiwv/JNXV1UVXV1dNx0XNzGzyc+E3m+RqvcaBxyaYTW6FHuOXtIekyyX9QtKwpJdL2kvSSkm3pfs907KS9DlJayXdLOmIqvUsSMvfJmlBkZnNJptaBiO66JtNfkXv8X8WuDYi3ippZ+DpwEeBVRGxVNJiYDGwCDgOODjdjgQuAI6UtBewBJgLBLBa0oqIeGAiAj6jfTEvWra4tjcty7tugOa/2IyZmbWOwgq/pN2BVwOnAkTEH4E/SjoRmJcWWwYMkhX+E4FLItvl+FHqLdg/LbsyItal9a4EjgUm5PqnDw0vnXQz95mZmY2lyD3+g4DfAl+RdDiwGvgA0BYR96ZlfgO0pcczgLuq3n93ahurfSuSTgdOB2hra2NwcDB30FqW3bBhQ2HrLkKteZuFMxevGfLW3ONWQ2/b4OBu2xdqgpTx/54zF68p8o51nG9Hb2Rd85uAI9PzzwLnAg+OWO6BdH8V0FHVviqt40zgY1XtHwfOHO+z58yZE3kduOiq3MtGRAwMDBS27h1Fdigk961Z1ft7mwhly9wseWvJ0cz/90ZTS95m4czFKzIvcGPkqM9FDu67G7g7In6cnl8OHAHcl7rwSff3p9fvAZ5T9f6ZqW2sdhthtH9gD+IyM7NqhRX+iPgNcJekQ1LTfOBWYAVQGZm/ALgyPV4B/Hka3X8UsD6yQwLXAcdI2jOdAXBMajMzM7MaFT2qfyHwjTSi/3bgNLKNjcskdQN3AienZa8BXg+sBR5JyxIR6ySdC9yQljsn0kA/M7Nm4LkSrEwKLfwRcRPZcfqR5o+ybABnjLGei4GLJzad1ZMvv2qT2WiFfNbiq2s6Y8isXjxzH9tx2t21+YuSZXz5VTOz5tDyhb/WLXJvxZuZWZm1fOE3M2tFHpfQugqdq9/MzJqTT/9tXS78ZmZmLcRd/VYXRU7N6gshlVtNAzQ9sNZsh7nwW13UcjEkj+pvHbUMlPXAWrOJ4cJvNopa5x0Azz1gZuXgwm82ilrmHQD3UphZeXhwn5mZWcH6+/uZPXs28+fPZ/bs2fT39zcsi/f4zczMCtTf309PTw99fX1s3ryZqVOn0t3dDUBXV1fd83iP38zMrEC9vb309fXR2dnJtGnT6OzspK+vj97e3obkceE3MzMr0PDwMB0dHVu1dXR0MDw83JA8LvxmZmYFam9vZ2hoaKu2oaEh2tvbG5Jnm8f4Jb0A+DBwYPXyEXF0gblsEvJELWa2vcp8bYGenh66u7u3HOMfGBigu7u7YV39eQb3fRP4IvBlYHOxcWyy8kQtZrYjxirkZfh7URnAt3DhQoaHh2lvb6e3t7chA/sgX+HfFBEXFJ7ErInUPMUweJphMxtTV1cXXV1dNc35UZQxC7+kvdLD70j6K+AK4PHK6xGxruBsZg1TyxTD0BwT+NQ626BnGjRrTePt8a8GAqgcWPlw1WsBPLeoUM1gvONJOv+pbc10PMlaUy2zDTbDhoqZNcaYhT8iDqpnkGYzViFvhm4aMzOz7bXN0/kknSFpj6rne6aufzMzMyuZPOfxvzsiHqw8iYgHgHcXF8nMzMyKkqfwT1XVAW9JU4Gdi4tkZmZmRclT+K8F/lXSfEnzgf7Utk2S7pD0c0k3Sboxte0laaWk29L9nqldkj4naa2kmyUdUbWeBWn52yQtqP3HNDMzM8h3Hv8i4D3AX6bnK4GLaviMzoj4XdXzxcCqiFgqaXF6vgg4Djg43Y4ELgCOTKcVLgHmkp1NsFrSinTIwawwNY9m92yDk15Rp0yCT5u0+tlm4Y+IJyT1AUNkhfeXEbEjM/idCMxLj5cBg2SF/0TgksiG0/9I0h6S9k/LrqzMGyBpJXAsWc+DWSFqnQmsDLOH2Y4r6pRJ8GmTVj955uqfR1ag7yA7p/85khZExPdzrD+A70kK4EsRcSHQFhH3ptd/A7SlxzOAu6ree3dqG6t9ZM7TgdMB2traGBwczBGvdhs2bChs3XmdsephHs6/05H7D8puO8G/zN9tO1NNrEZ/x9ujGTLnzVDr73Ez/GzQHDmK+o5rWXeRmiFDrcqUuRlqCBEx7o1sIp9Dqp6/AFi9rfelZWek+2cBPwNeDTw4YpkH0v1VQEdV+yqy7v0zgY9VtX8cOHO8z50zZ04UZWBgoLB153XgoqtyL1tL3lrWW6RmyVGLZsjs34vmylDr34qy/XzNomyZi6whwI2RozbnOca/U0T8smpD4X8k5TpIGRH3pPv7JV0BvAy4T9L+EXFv6sq/Py1+D/CcqrfPTG338OShgUr7YJ7PN2slNV9fwNcWMGtJeQr/jZIuAr6enr8duHFbb5K0GzAlIh5Kj48BzgFWAAuApen+yvSWFcD7JC0nG9y3Pm0cXAd8sjL6P63nrFw/nTW1WqdFBk+NPJ5ari/gKXvNWleewv+XwBnA+9PzHwBfyPG+NuCK9Md9GnBpRFwr6QbgMkndwJ3AyWn5a4DXA2uBR4DTILsYkKRzgRvScueELxA0KYxVxD0tsplZcfKM6n9c0ufJjrk/QTaq/4853nc7cPgo7b8H5o/SHmQbGKOt62Lg4m19ppmVny+QZVasPHP1Hw/8Cvgs8HlgraTjig5mZq1prAFJAwMDYw0iNrMa5Onq/0eySXjWAkh6HnA18N0ig5mZmdnEyzNl70OVop/cDjxUUB4zMzMrUN5R/dcAl5FNyHMScIOktwBExLcLzGdm1jSKOmUyWzf4tEmrhzyFfxfgPuBP0/PfArsCJ5BtCLjwm1lLKOqUSfBpk1Y/eUb1n1aPIGa242oqHr6okFlLyjNX/wvIrpTXFhGzJb0YeGNEnFd4OjPLrZaLBPmiQmatK8/gvi+TzZS3ESAibgbeVmQoMzMzK0aeY/xPj4jrR0yqsamgPJbDZJ2Tvb+/n97eXoaHh2lvb6enp4eurq6G5TEzm4zyFP7fpXP3A0DSW4F7x3+LFWkyzsne399PT08PfX19bN68malTp9Ld3Q3QVMXfs8qZWdnl6eo/A/gS8EJJ9wAfBN5baCprOb29vfT19dHZ2cm0adPo7Oykr6+P3t7eRkfbimeVM7OyyzOq/3bgNdVX2ys+lrWa4eFhOjo6tmrr6OhgeHi4QYnMzCanPF39AETEw0UGsdbW3t7O0NAQnZ2dW9qGhoZob29vYCqz8jv87O+x/tGNuZfPe8hv91134mdLjtneWNZAuQu/WZF6enro7u7ecox/YGCA7u7upuvqNytirgQobr6E9Y9unHRjgmzH5DmP/2kR8fi22sx2RGUA38KFC7eM6u/t7W2qgX1mnivBJoM8g/t+mLPNbId0dXWxZs0aVq1axZo1a1z0zcwKMOYev6T9gBnArpL+BKicx/RM4Ol1yGZmZmYTbLyu/tcBpwIzgX/kycL/EPDRYmOZmZlZEcYs/BGxDFgm6f9FxLfqmMnMzMwKkucY/0xJz1TmIkk/keRzOMzMzEooT+H/i4j4A3AMsDfwTmBpoanMzMysEHkKf+XY/uuBSyLilqo2MzMz24b+/n5mz57N/PnzmT17Nv39/Q3LkmcCn9WSvgccBJwl6RnAE3k/QNJU4Ebgnoh4g6SDgOVkvQergXdGxB8lPQ24BJgD/B74s4i4I63jLKAb2Ay8PyKuy/v5k1URk4gUNYGImVkra7aLkOUp/N3AS4DbI+IRSXsDp9XwGR8AhslOAwQ4H/hMRCyX9MW0/gvS/QMR8XxJb0vL/ZmkQ4G3AYcBzwb+XdILImJzDRkmFU8iYmZWHr29vZxyyilbTVB2yimnNGySsjyFP4BDgTcA5wC7AbvkWbmkmWQXeO8FPqTsmqZHA6ekRZYBnyAr/CemxwCXA59Py58ILE8zBf6vpLXAy/AkQma5jHUp4dEuIwy+lLDZRLv11lt55JFHnrLHf8cddzQkT57C/wWyrv2jyQr/Q8C3gJfmeO8/AR8BnpGe7w08GBGb0vO7ySYJIt3fBRARmyStT8vPAH5Utc7q92wh6XTgdIC2tjYGBwdzxKvdhg0bClt3UcqWt4zfcTNnHhgYeErbhg0bmD59+qjLN+vP0czf8ViaJW/eHLV+x2X7+Rpl2rRpHHvssUjiscceY/r06Rx77LFcdNFFjck+1vXFq64n/pN0/9Oqtp/leN8bgC+kx/OAq4B9gLVVyzwHWJMerwFmVr32q7T854F3VLX3AW8d77PnzJkTRRkYGChs3UU4cNFVjY5Qs7J9xxHly1y2vBHly9ws//dqyVHLd1zGn69RJMU+++wTs2bNiilTpsSsWbNin332CUkT+jnAjbGN2hwRufb4N6YBegEgaV/yDe57JfBGSa8nOzTwTOCzwB6SpkW21z8TuCctf0/aELhb0jRgd7JBfpX2iur3mJmZNbUZM2bw0EMPAU8eStu4cSMzZjyl87ou8hT+zwFXAM+S1Au8FfjYtt4UEWcBZwFImgecGRFvl/TNtI7lwALgyvSWFen5D9Pr/5G2lFYAl0r6NNngvoOB63P/hGZmLewZ7Yt50bLF+d+wLO96IRvCZXk8/elP5+KLL95yjP/tb397w7Jss/BHxDckrQbmk52//6aIGN6Bz1wELJd0HvBTsq570v3X0uC9dWQj+YmIWyRdBtwKbALOiBYe0W9mVouHhpfmPrNncHCQefPm5Vq2plOKW9yvf/1rvvrVr241qv/888/n1FNPbUiebRZ+SQcAjwDfqW6LiP/L+yERMQgMpse3k43KH7nMY8BJY7y/l+zMADMzs1Jpb29n5syZrFmzZsvG1cDAAO3t7Q3Jk6er/2qy4/siO1Z/EPBLsvPqzczMbBw9PT10d3dvOZ1vYGCA7u5uensbsz+bp6v/RdXPJR0B/FVhiczMzCaRyiQ91V39jZq8B/Lt8W8lIn4i6cgiwpiZmU1GXV1ddHV11TSOoih5jvF/qOrpFOAI4NeFJTIzM7PC5Lk63zOqbk8jO+Z/YpGhzMzMJpNSXZ0vIs6uRxAzs7Ly9RBsPM12db5t7vFL+o6kFWPd6hHSzKyZjTYt6sDAwHhTmlsL6e3tpa+vj87OTqZNm0ZnZyd9fX3NO6ofuB3YD/h6et4F3Af8W1GhzMzMJovh4WE6Ojq2auvo6GB4eEfmwtt+eQr/KyNibtXz70i6MSL+uqhQZmZmk0V7eztDQ0N0dnZuaRsaGmrqCXx2k/TcNOMekg4Cdis2lpmZtarDz/4e6x/dmHv5vNMH777rTvxsyTHbG2u7lW4CH+CvgUFJt5PN3ncg8J5CU5mZWcta/+jGSXV9gdJN4BMR10o6GHhhavpFRDxebCwzM7PJo1QT+CRzgFlp+cMlERGXFJbKzMzMCpFn5r6vAc8DbgIql8MNwIXfzKwEaurivjb/8XIrpzx7/HOBQ8MnnzY9TyJiZiPlPVYO2QZCLctbOeWZsncN2Xn81uQ8iYiZmW1Lnj3+fYBbJV0PbBnUFxFvLCyVmZmZFSJP4f9E0SHMzMysPvKczvefktqAl6am6yPi/mJjmZmZWRHyXKTnZOB64CTgZODHkt5adDAzMzObeHm6+nuAl1b28iXtC/w7cHmRwczMzGzi5RnVP2VE1/7vc77PzMzMmkyeAn6tpOsknSrpVOBq4LvbepOkXSRdL+lnkm6RdHZqP0jSjyWtlfSvknZO7U9Lz9em12dVreus1P5LSa/bnh/UzMzMchT+iPgw8CXgxel2YUR8JMe6HweOjojDgZcAx0o6Cjgf+ExEPB94AOhOy3cDD6T2z6TlkHQo8DbgMOBY4AuSpub/Ec3MzKxizMIv6fmSXgkQEd+OiA9FxIeA30p63rZWHJkN6elO6RbA0Tw5PmAZ8Kb0+MT0nPT6fGVT0Z0ILI+IxyPif4G1wMtq+SHNzMwsM94e/z8BfxilfX16bZskTZV0E3A/sBL4FfBgRGxKi9wNzEiPZwB3AaTX1wN7V7eP8h4zMzOrwXij+tsi4ucjGyPi59XH38cTEZuBl0jaA7iCJy/tO+EknQ6cDtDW1sbg4GAhn7Nhw4bC1l2EsuUFZ66HsuWF8mUuW96KZsmcN0et33Ojf76m+L0YZx7328Z5be1Yr43znr8FPgz8DpiW2l4OXJceXwe8PD2elpYTcBZwVtV6tiw31m3OnDlRlIGBgcLWXYSy5Y1w5nooW96I8mUuW96IiAMXXdXoCBFRW45avudG/nyXXnppHHbYYTFlypQ47LDD4tJLL53wzwBujBz1eLw9/hslvTsivlzdKOldwOptbVCk8/03RsSDknYFXks2YG8AeCuwHFgAXJnesiI9/2F6/T8iIiStAC6V9Gng2cDBZBMKmZmZNb3+/n56enro6+tj8+bNTJ06le7ubFx7V1dX3fOMV/g/CFwh6e08WejnAjsDb86x7v2BZWkE/hTgsoi4StKtwHJJ5wE/BfrS8n3A1yStBdaRjeQnIm6RdBlwK7AJOCOyQwhmZmZNr7e3l76+Pjo7OxkcHGTevHn09fWxcOHC5ir8EXEf8ApJncDs1Hx1RPxHnhVHxM3An4zSfjujjMqPiMfIpgUebV29QG+ezzUzM2smw8PDdHR0bNXW0dHB8PBwQ/LkuUjPAFn3vJmZmdWovb2doaEhOjs7t7QNDQ3R3t7ekDyeetfMzKxAPT09dHd3MzAwwKZNmxgYGKC7u5uenp6G5MlzkR4zMzPbTpXj+AsXLmR4eJj29nZ6e3sbcnwfXPjNzMwK19XVRVdX15bBfY3krn4zM7MW4sJvZmbWQlz4zczMWogLv5mZWQtx4TczM2shLvxmZmYtxIXfzMyshbjwm5mZtRAXfjMzsxbiwm9mZtZCXPjNzMxaiAu/mZlZC/FFeszMWpCk0dvPH335iCgwjdWT9/jNzFpQRDzlNjAwMGq7i/7k4sJvZmbWQlz4zczMWogLv5mZWQtx4TczM2shLvxmZmYtpLDCL+k5kgYk3SrpFkkfSO17SVop6bZ0v2dql6TPSVor6WZJR1Sta0Fa/jZJC4rKbGZmNtkVuce/CfibiDgUOAo4Q9KhwGJgVUQcDKxKzwGOAw5Ot9OBCyDbUACWAEcCLwOWVDYWzMzMrDaFFf6IuDcifpIePwQMAzOAE4FlabFlwJvS4xOBSyLzI2APSfsDrwNWRsS6iHgAWAkcW1RuMzOzyUz1mJhB0izg+8Bs4P8iYo/ULuCBiNhD0lXA0ogYSq+tAhYB84BdIuK81P5x4NGI+IcRn3E6WU8BbW1tc5YvX17Iz7JhwwamT59eyLqLULa84Mz1ULa8UL7MZcsLzZP51Gsf5qvH7pZr2Voy17LeohT5HXd2dq6OiLnbWq7wKXslTQe+BXwwIv5QPU1kRISkCdnyiIgLgQsB5s6dG/PmzZuI1T7F4OAgRa27CGXLC85cD2XLC+XLXLa80ESZr706d46aMtew3qI0w3dc6Kh+STuRFf1vRMS3U/N9qQufdH9/ar8HeE7V22emtrHazczMrEaF7fGnbvw+YDgiPl310gpgAbA03V9Z1f4+ScvJBvKtj4h7JV0HfLJqQN8xwFlF5TYzs8Z6RvtiXrRs8bYXrFi27UWy9QIcvz2RJpUiu/pfCbwT+Lmkm1LbR8kK/mWSuoE7gZPTa9cArwfWAo8ApwFExDpJ5wI3pOXOiYh1BeY2M7MGemh4KXcszVega+k6n7X46h1INXkUVvjTIL3Rr/sI80dZPoAzxljXxcDFE5fOzMysNXnmPjMzsxbiwm9mZtZCXPjNzMxaSOHn8ZuZmdWqpoF41+Zbdvddd9rONJOLC7+ZmTWVvCP6IdtAqGV5c1e/mZlZS/Eev5mZWQGqp6jPox7XzgHv8ZuZmRUiIp5yO3DRVaO216vogwu/mZlZS3HhNzMzayEu/GZmZi3Ehd/MzKyFuPCbmZm1EJ/OZ2ZmTW+8U+N0/lPb6jlKvmxc+M3MrOmNVcgHBweZN29efcOMcPjZ32P9oxtzL1/LdMS777oTP1tyzPbEGpMLv5mZ2Q5Y/+jG3NMG17qhUtM1C3Jy4TczM9sBz2hfzIuWLc7/hmW1rBtgYq9F4MJvZma2Ax4aXlqqPX6P6jczM2sh3uM3MzPbQTXtmV9b2+C+iebCb2ZmtgPydvNDtoFQy/JFcFe/mZlZCyms8Eu6WNL9ktZUte0laaWk29L9nqldkj4naa2kmyUdUfWeBWn52yQtKCqvmZlZKyhyj/+rwLEj2hYDqyLiYGBVeg5wHHBwup0OXADZhgKwBDgSeBmwpLKxYGZmZrUrrPBHxPeBdSOaT+TJMxiXAW+qar8kMj8C9pC0P/A6YGVErIuIB4CVPHVjwszMrOlIesrtzvPfMGr7eFMST3iuIuczljQLuCoiZqfnD0bEHumxgAciYg9JVwFLI2IovbYKWATMA3aJiPNS+8eBRyPiH0b5rNPJegtoa2ubs3z58kJ+pg0bNjB9+vRC1l2EsuUFZ66HsuWF8mUuW15w5nooMm9nZ+fqiJi7reUaNqo/IkLShG11RMSFwIUAc+fOjaLmbm6GeaFrUba84Mz1ULa8UL7MZcsLzlwPzZC33qP670td+KT7+1P7PcBzqpabmdrGajczM7PtUO/CvwKojMxfAFxZ1f7naXT/UcD6iLgXuA44RtKeaVDfManNzMzMtkNhXf2S+smO0e8j6W6y0flLgcskdQN3Aienxa8BXg+sBR4BTgOIiHWSzgVuSMudExEjBwyamZlZToUV/ojoGuOl+aMsG8AZY6znYuDiCYxmZmbWsjxzn5mZWQtx4TczM2shLvxmZmYtpNAJfBpF0m/JBg8WYR/gdwWtuwhlywvOXA9lywvly1y2vODM9VBk3gMjYt9tLTQpC3+RJN2YZ2akZlG2vODM9VC2vFC+zGXLC85cD82Q1139ZmZmLcSF38zMrIW48NfuwkYHqFGVpzMqAAAPeElEQVTZ8oIz10PZ8kL5MpctLzhzPTQ8r4/xm5mZtRDv8ZuZmbUQF34zM7MW4sJvZmbWQlz4W5ikaSOeq1FZJitJUxudoVaSdmp0hlpImt7oDLWStI+kZzc6Ry0kHSDp6EbnyEvSrpKe1ugcO0rShNdpF/46qS6qzVBgJbUDl0o6U9K7IbtKYjNkG42kAyUd2ugctZD0AuDLkhZJekej8+Qh6RCy34slkk5vdJ5tkbQ/cLukNzY6S17p/95VQGkKv6TZwBXAs8uwMSvphWSXe/9iurR7qUjaT9ILJT07Ip6Y6OLvwl8wSU+XtGsqqi+ELZchbmSmGcA3gR8DdwN/Lqmvkq3Zir+kg4HbgQslvajRefJI/9bfAtYCjwLHSnplY1ONLxX9frLfi7uADkk7V73eVL8Xyd7AVOCTZSj+aWPwX4F/iYgbR3m96b5jSbsC5wCfi4ivR8TmRmcaj6RZZL/HlwEXAM+v3vNvxu+4WvrbsRI4CxiWNCcV/wnLPW3bi9gOejlwiqTrgI9JelNE3N7gTLsCt0TEPwJIugL4saS+iOhu9IZJtbR3MR/4CLAb8AlJZwM/b6ac1dIfyg+R/XH/oqTdgXbgucB/NTTcGFLmDwD/HBFfkfRc4FTgBEmKiMub8fuOiDWSzgfuBT4n6UHgh+m1jQ0NN0LaiPor4O6I+Fpq+wiwHtgcERc143cMVDL9F0DaSXic7HTwv0xtaqLss4CBiLhA0jOBQ4C/krR7RHyisnPTRHm3SId/LgP+ISKWSXof8CVJR0fEHybqc7zHX7CIWEW2V/IN4KyIuL0Juso2ATtX9UA8DrwUeKGkDzQ02Qhp7+Jq4EsRcQ7wP8DfAi9ugu9xVBHxKPAV4Lr0fD3wM6Bp9/hT5vNT0Z8C/B1ZQdoXeG8zdpdKmprGqcwB1gAnAV8H1tGE33VE/BH4NvBTSR+XdANZUdoDeKekJQ0NOIaIeAz4CXCopE8CfwD+GWiXdElaptmK6KvToaoBYBj4OfAWSV+GpsxbsS/Zxvey9PwrwP8Bf5zID3HhL0ilW0bSfsCVZFtxH5D0rEZ3lUXEHWT/kf9N0p6pbSPQS3blqKYSEXcBj6THZwG3AUuAvSV1SPqLRuarVvl3j4gfRsT/Vr30B2CvtMzcZuqWrspcuaLlM4BvRMQJEfFF4GzgWY3KN5aI2BwRm8i6zveLiBuAJ0i/K80oIr4PXAu8ABhMPWznk23M7tfQcOO7g6zreQbw6YgYJuuJOyD9jWsaETEIfB54DPhNRLw9Iv4dOBrYJ/UCNKthst8PACLiYbL/ewcATFR2F/6CpO6kE4C/B74fEW8nu1TwNyVNkfSK1I1TV5VBIhFxLtkv2I9Tty7ATsDL0mjYpjoOlo5x7ZQen0XWnXsVcDlwfyOzVRu5J6Enz5y4G7hN0vPJpuycsG67HVWdOXWBro+IK6oW2Rdoa8bfi+Rh4DxJvwA+CbwNuCQN/GsaVRtYPwDOJdt4rdgX2K/ZvuOqzF8HlgOvAo6StBswG3g62cZWU6j6+/ZVsr8PD0qqbLTOBtposkPcSgOX0/+9P6YdnUqP1lSyQ5wbJL0C+I6kvXb4QyPCtwJuPNn9eNSI9ouA7wG/AN7UoGxPq3r812Rd0n1kA9FOaPR3N0reqdX36fFcYDPwhvRcjc45XmbgCLKu85ub8Tse53t+NdlhiuManW+UvNOqHi8F/rrq+X6NzjdG5p3TvaraXtWs33HKV/334s/INlwvBG5s1N+wGn4vFgO/AT4G3Fr5e9EsN+Dg9HdsCHjRGMt8HjgD+G/gzRPyuY3+wSfrjWyv48L0WJX/8On54cALK6/VKc8hwEsqn0c2AOai9PjFwJ8AL61nphrzzgS+CuxOtgX8fuDEZsmbI/NLyPbyj250zpyZnwkcBvxnM21cjZL3ALLjoM+uWmZKM2TN+XvxPOC7wPGNzrmNzAeS7Rwo5Z8FzG7i34sDq/6+/TnwDmBes+RNOaYC7wX+huxQz7fS32KNWO4assMWr5uo/E3V5VFmo4wSvQfYJR3/uj8i/ijp1cDeUdWNOuI9heQi+0P4QWB9RNwkaW9gGbAiZbi5+j1FZxrPOHm/AVwR2UA5JF0YEY81Q7dozsw3SToiItY2w4jiHJn/ANwiqSsift3ozOPk/TrwrYj4dWXZiGiKruecvxfrJf1FRNzbyKwV42S+BLgy/Q7cXf2eJv29uAT4Tsp3SfV7Gv1/ryIiNku6GnggIjZI+hTZBsC5km6JbPwKZGNY/i6ysQsTk7/RWz2T4caTW5mvITtd5zSygTrfBD4MHA+8Avgl0FHnbFPS/UHAD8gOQcygqouOJtkCLmPePJmrfj+2um/mzM12m4y/F814K9v3XLa84/0M6fGnyM78eBbZIaCu6p9jon4W7/FPgIgISa8FPgMsJDtm/l7g42TdTF1k3XpnRsRQvXKl0/UOkbQ2Im6RNADMjIjVZD0STXX+bdnyQr7MFZXcjc5ftu+5bHnBmeuhbHnHEmngckRsjIizJH2YbGDiAcC7q5absJ/Do/onQOpuOgn4C7LzLW8GVkbELyLioxHxDuDUiPhO0V3TlfVLOgb4B7JBZZdIegOwM3Cm0il80BRFqFR5wZnroWx5wZnroWx5t0VPzkVSfXhqgKzn4l2F1YxGd3OU/UY2UciLyEZdfoFsdqvnpdfeDZyUHtetuwk4ElgFzEnP55GNer6AbIDZ21L7lHplmkx5ndl5ndl5dyB/wwcuu6t/B0iaC3waOIVs0pDjyY7J/ErSi8mmQP0A1G/LU9m5y+8F1kXW5UVEDEr6Idl5+o8CbwKWRxMMgipbXnDmeihbXnDmeihb3mo5B3vWZeCyu/q3k7KrVb0b+I+I+FVEfIVsgouPSPom2akvPZFN2VuvTK8B3kc2K99Bkt5a9fITEbEhIj4EzFB2hbCGKltecOZ6KFtecOZ6KFveUSiyWVv/DnilpDnALsBnIuKfYMuYhMcg21ksaofRe/zbQdKRwJ+STcG6t6RDI+LWiFgk6SXARrJ/t1vrNbgk9TAsAd5FNkPgY8BrJW2MiCsjXbBE2dXt9gMeKDrTeMqWN2Vx5oKVLW/K4swFK1vekdRkAxFd+Guk7Jrw/0zWrX8F0AMcJ2lTRPxPRNxUvXydiv4BZMeFHo+IX6a2a8gGjLxV0tSI+HZa/DfAqyPiN0XnGkvZ8oIz10PZ8oIz10PZ8lZUCrmygYjvB1YDf6vsYkyVgYjfj4gHoL4DEd3VX4O01fn3wFBE/DYibiObTvFQ4GRl87A3wm/JrqG+SdI7lZ0acg/ZaYVDZBe1ASDlbvRkIWXLC85cD2XLC85cD2XLC2w5zftIYBGwJCKWkM3S10E2kO9w4HUAStcYqBfv8dfmV2SX/DxM0izgzoi4UdKXyM7fr8tgkqotyQ5gOvD7iPiypCeAlwEbJV0eEXdL+mpkl91tmLLlBWd23tE5s/PmpSYeiOg9/nFIW84Znatsgp6DgAVkU1aeSTaIRBFxPXBGRNxej1zpP8UbyXobDgM+JeldEdFHNvDlNWQX06AZ/lOULS84cz2ULS84cz2ULe9o1OQDEb3HP470C3g8cB7ZP+D+ZFfRehfwZeAT6XY38FC9ckmaCfwlcALZNab3BE6StGtE/LOyS8HeNN466qlsecGZ66FsecGZ66FseUdSGQYiRhNMaNCMN7J5kXchuzLSvNS2D9lUiu9Lr10OHFqvPOn+VWSTBT0XOIrsP8CBZNcHuBP4UKO/uzLmdWbndWbnnYCf4wCyS6//e1XbDKAb+Brwlqr2fYH9G5HTXf0jVLr3gV0iO59yM1DpTvo98C/AC9JrfxYRt9YjV0SEpBOAzwK3RnZYYX/gGxFxJ9m13i8nmzmw4cqWF5y5HsqWF5y5HsqWdxzlGIjY6C2kZrrx5Fbn0UBvevxh4BbggPT8BOBqskEnU+uYbTrZ6YNHVbW9mewX6W/ITmN5VaO/w7LmdWbndWbn3Y7clZrRARwLvDQ97yY77fttwLTU9rRG563cfIy/SkREGpTxBbJjTETE30vaCfiBpK+QXWnvgxGxod7xyLqGngnZ6R8RcYWkQ8i2Mt8ZET+oc6bxlC0vOHM9lC0vOHM9lC0vsNVAxHPIuvKPk7Q8Ii6SdBrZQMSpZD0XTTMQ0V39iTLTgNcDH42IVangExGfJLvy3n+TXWXvu/XOFxEPA/8KvEJSe2SXcnw58ApgICJW1jvTeMqWF5y5HsqWF5y5HsqWt2LEQMTf8eRAxIWRTeP+Y5pwIGKlm8ISSeeRDdzrqWyhKbsYz68j4tcNzjYDeA/QSXa86GRgYURc08hcYylbXnDmeihbXnDmeihL3nQKd0h6FfBi4LvAs4AvAieSHSr+BPDZiPh0w4KOw4V/BGXXdT4a+G5ErFQ29/5nyX4Bb25sOpC0G/BSoA24IyJ+3OBI4ypbXnDmeihbXnDmeihL3jQQ8WzgbyJiQNKbgeenQ8NvIbtc+2VNm7+VC7+kaRGxaUTbTsCHyKbhnQHsDZwTEVc0IKKZmTURSdPJjuefHxE/Sm1vJrvq3hfJBoSf1IxjEipabnCfsgs+nBsRCyJik7ILPGxOr02JiI2S/p7snP0ZwIaIuK3SvdPI7GZm1nClHIhYreUG90XE/5ENIPlmer5Z0tT0+IlU4J+IiPsj4qeRXYjHzMystAMRq7VM4Zc0S9JigIg4GDhQ0hXpeXXxj7T8tHQ/pbrdzMxa3rfJ6ueFkj4FfB34YkTc1dhY+bTMMX5JBwI/IvvHOTu1XQ/cExFvTs+nVjYC0v2ewMXAO9JWnpmZWWkGIo6mJQp/ZRBfKv7fAVZExMfSa9eTXV73pPR8p3Scfw/gMuBTETHQsPBmZmYTaNIX/qpzLg8B1gFPI7vwzrcj4hNpmVuAX0XEG9PzPcmK/jnNPkjDzMysFpN+VH8q+ieQXVr3DmAt8BHgM5I2RcR5EXGYpFdC1t1PdkrGp1z0zcxssmmFPf6jyC6WcDzwWuBC4J+AH6THX4mIJVXLTwV2j4h1DYhrZmZWqFYo/DPJLu+4J9le/ynAl8iu9nQ5sC4i/rNxCc3MzOpn0p/OFxF3R8QNwJ+SXSFpLbAMeCGwOiL+U5IaGtLMzKxOJv0x/io/B96TpuR9C/D+NJmPz9E3M7OW0UqF/xqyEf1vBHoj4r8anMfMzKzuJv0x/pGqzun33PtmZtZyJv0x/lFsBnfvm5lZa2q5PX4zM7NW1op7/GZmZi3Lhd/MzKyFuPCbmZm1EBd+MzOzFuLCb2Zm1kL+P3ap28zoFPZZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_multiply_plot.boxplot(column=['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4', 'carry-5', 'carry-6', 'carry-8', 'carry-12'], rot=45)\n",
    "df_run_info_multiply_plot[['carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4', 'carry-5', 'carry-6', 'carry-8', 'carry-12']].to_csv(\n",
    "    'result_statistics/epoch_multiply_carry.csv', index=False)\n",
    "save_fig('epoch_multiply_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4', 'carry-5', 'carry-6', 'carry-8', 'carry-12']\n",
    "column_list = ['carry-0', 'carry-1', 'carry-2', 'carry-3', 'carry-4', 'carry-5', 'carry-6', 'carry-8', 'carry-12']\n",
    "\n",
    "x = ['0', '1', '2', '3', '4', '5', '6', '8', '12']\n",
    "y = df_run_info_multiply_plot[column_list].mean()\n",
    "e = df_run_info_multiply_plot[column_list].std()\n",
    "\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.xlabel('Carries', fontsize=font_size['xlabel'])\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_multiply_carry')\n",
    "\n",
    "df_plot = df_run_info_multiply_plot[column_list]\n",
    "df_plot.columns = x\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_multiply_carry.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_multiply_plot.boxplot(column=['all_correct', 'digit-1', 'digit-2', 'digit-3', 'digit-4', 'digit-5', 'digit-6', 'digit-7', 'digit-8'])\n",
    "save_fig('epoch_multiply_digit')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Divide"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['init_all_correct_carry-0_epoch',\n",
       " 'init_all_correct_carry-1_epoch',\n",
       " 'init_all_correct_carry-2_epoch',\n",
       " 'init_all_correct_carry-3_epoch',\n",
       " 'init_all_correct_digit-1_epoch',\n",
       " 'init_all_correct_digit-2_epoch',\n",
       " 'init_all_correct_digit-3_epoch',\n",
       " 'init_all_correct_digit-4_epoch',\n",
       " 'init_all_correct_digit-5_epoch',\n",
       " 'init_all_correct_digit-6_epoch',\n",
       " 'init_all_correct_digit-7_epoch',\n",
       " 'init_all_correct_digit-8_epoch',\n",
       " 'init_all_correct_epoch',\n",
       " 'init_complete_all_correct_carry-0_epoch',\n",
       " 'init_complete_all_correct_carry-1_epoch',\n",
       " 'init_complete_all_correct_carry-2_epoch',\n",
       " 'init_complete_all_correct_carry-3_epoch',\n",
       " 'init_complete_all_correct_digit-1_epoch',\n",
       " 'init_complete_all_correct_digit-2_epoch',\n",
       " 'init_complete_all_correct_digit-3_epoch',\n",
       " 'init_complete_all_correct_digit-4_epoch',\n",
       " 'init_complete_all_correct_digit-5_epoch',\n",
       " 'init_complete_all_correct_digit-6_epoch',\n",
       " 'init_complete_all_correct_digit-7_epoch',\n",
       " 'init_complete_all_correct_digit-8_epoch',\n",
       " 'last_epoch']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_keys_divide"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>init_all_correct_digit-1_epoch</th>\n",
       "      <th>init_all_correct_digit-2_epoch</th>\n",
       "      <th>init_all_correct_digit-3_epoch</th>\n",
       "      <th>init_all_correct_digit-4_epoch</th>\n",
       "      <th>init_all_correct_digit-5_epoch</th>\n",
       "      <th>init_all_correct_digit-6_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>init_complete_all_correct_carry-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-1_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-2_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-5_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-6_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-7_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-8_epoch</th>\n",
       "      <th>last_epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2420.560000</td>\n",
       "      <td>1012.660000</td>\n",
       "      <td>2119.040000</td>\n",
       "      <td>1580.320000</td>\n",
       "      <td>2421.460000</td>\n",
       "      <td>1801.970000</td>\n",
       "      <td>785.890000</td>\n",
       "      <td>416.390000</td>\n",
       "      <td>1.910000</td>\n",
       "      <td>1.920000</td>\n",
       "      <td>...</td>\n",
       "      <td>1819.870000</td>\n",
       "      <td>2421.460000</td>\n",
       "      <td>1826.670000</td>\n",
       "      <td>824.330000</td>\n",
       "      <td>416.610000</td>\n",
       "      <td>1.910000</td>\n",
       "      <td>1.920000</td>\n",
       "      <td>1.810000</td>\n",
       "      <td>1.950000</td>\n",
       "      <td>2421.460000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>108.310564</td>\n",
       "      <td>115.437352</td>\n",
       "      <td>139.744928</td>\n",
       "      <td>571.006694</td>\n",
       "      <td>107.789771</td>\n",
       "      <td>89.480256</td>\n",
       "      <td>44.638366</td>\n",
       "      <td>31.760116</td>\n",
       "      <td>1.436149</td>\n",
       "      <td>1.220035</td>\n",
       "      <td>...</td>\n",
       "      <td>127.802298</td>\n",
       "      <td>107.789771</td>\n",
       "      <td>97.005368</td>\n",
       "      <td>48.525127</td>\n",
       "      <td>31.687201</td>\n",
       "      <td>1.436149</td>\n",
       "      <td>1.220035</td>\n",
       "      <td>1.134358</td>\n",
       "      <td>1.104399</td>\n",
       "      <td>107.789771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2071.000000</td>\n",
       "      <td>839.000000</td>\n",
       "      <td>1811.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2071.000000</td>\n",
       "      <td>1624.000000</td>\n",
       "      <td>691.000000</td>\n",
       "      <td>344.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1515.000000</td>\n",
       "      <td>2071.000000</td>\n",
       "      <td>1630.000000</td>\n",
       "      <td>700.000000</td>\n",
       "      <td>344.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2071.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2355.750000</td>\n",
       "      <td>952.500000</td>\n",
       "      <td>2025.750000</td>\n",
       "      <td>1646.500000</td>\n",
       "      <td>2355.750000</td>\n",
       "      <td>1744.000000</td>\n",
       "      <td>753.500000</td>\n",
       "      <td>395.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1726.000000</td>\n",
       "      <td>2355.750000</td>\n",
       "      <td>1765.750000</td>\n",
       "      <td>789.750000</td>\n",
       "      <td>395.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2355.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2428.500000</td>\n",
       "      <td>991.500000</td>\n",
       "      <td>2111.000000</td>\n",
       "      <td>1765.500000</td>\n",
       "      <td>2429.500000</td>\n",
       "      <td>1793.500000</td>\n",
       "      <td>785.500000</td>\n",
       "      <td>414.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1822.000000</td>\n",
       "      <td>2429.500000</td>\n",
       "      <td>1818.000000</td>\n",
       "      <td>826.500000</td>\n",
       "      <td>414.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2429.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2493.000000</td>\n",
       "      <td>1037.250000</td>\n",
       "      <td>2202.750000</td>\n",
       "      <td>1853.250000</td>\n",
       "      <td>2493.000000</td>\n",
       "      <td>1850.250000</td>\n",
       "      <td>812.250000</td>\n",
       "      <td>437.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1914.500000</td>\n",
       "      <td>2493.000000</td>\n",
       "      <td>1876.750000</td>\n",
       "      <td>858.250000</td>\n",
       "      <td>437.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2493.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2647.000000</td>\n",
       "      <td>1681.000000</td>\n",
       "      <td>2476.000000</td>\n",
       "      <td>2139.000000</td>\n",
       "      <td>2647.000000</td>\n",
       "      <td>2066.000000</td>\n",
       "      <td>913.000000</td>\n",
       "      <td>513.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2153.000000</td>\n",
       "      <td>2647.000000</td>\n",
       "      <td>2099.000000</td>\n",
       "      <td>933.000000</td>\n",
       "      <td>513.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>2647.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       init_all_correct_carry-0_epoch  init_all_correct_carry-1_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2420.560000                     1012.660000   \n",
       "std                        108.310564                      115.437352   \n",
       "min                       2071.000000                      839.000000   \n",
       "25%                       2355.750000                      952.500000   \n",
       "50%                       2428.500000                      991.500000   \n",
       "75%                       2493.000000                     1037.250000   \n",
       "max                       2647.000000                     1681.000000   \n",
       "\n",
       "       init_all_correct_carry-2_epoch  init_all_correct_carry-3_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2119.040000                     1580.320000   \n",
       "std                        139.744928                      571.006694   \n",
       "min                       1811.000000                        1.000000   \n",
       "25%                       2025.750000                     1646.500000   \n",
       "50%                       2111.000000                     1765.500000   \n",
       "75%                       2202.750000                     1853.250000   \n",
       "max                       2476.000000                     2139.000000   \n",
       "\n",
       "       init_all_correct_digit-1_epoch  init_all_correct_digit-2_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      2421.460000                     1801.970000   \n",
       "std                        107.789771                       89.480256   \n",
       "min                       2071.000000                     1624.000000   \n",
       "25%                       2355.750000                     1744.000000   \n",
       "50%                       2429.500000                     1793.500000   \n",
       "75%                       2493.000000                     1850.250000   \n",
       "max                       2647.000000                     2066.000000   \n",
       "\n",
       "       init_all_correct_digit-3_epoch  init_all_correct_digit-4_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                       785.890000                      416.390000   \n",
       "std                         44.638366                       31.760116   \n",
       "min                        691.000000                      344.000000   \n",
       "25%                        753.500000                      395.000000   \n",
       "50%                        785.500000                      414.500000   \n",
       "75%                        812.250000                      437.000000   \n",
       "max                        913.000000                      513.000000   \n",
       "\n",
       "       init_all_correct_digit-5_epoch  init_all_correct_digit-6_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                         1.910000                        1.920000   \n",
       "std                          1.436149                        1.220035   \n",
       "min                          1.000000                        1.000000   \n",
       "25%                          1.000000                        1.000000   \n",
       "50%                          1.000000                        1.000000   \n",
       "75%                          3.000000                        3.000000   \n",
       "max                          7.000000                        7.000000   \n",
       "\n",
       "          ...       init_complete_all_correct_carry-3_epoch  \\\n",
       "count     ...                                    100.000000   \n",
       "mean      ...                                   1819.870000   \n",
       "std       ...                                    127.802298   \n",
       "min       ...                                   1515.000000   \n",
       "25%       ...                                   1726.000000   \n",
       "50%       ...                                   1822.000000   \n",
       "75%       ...                                   1914.500000   \n",
       "max       ...                                   2153.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-1_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               2421.460000   \n",
       "std                                 107.789771   \n",
       "min                                2071.000000   \n",
       "25%                                2355.750000   \n",
       "50%                                2429.500000   \n",
       "75%                                2493.000000   \n",
       "max                                2647.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-2_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               1826.670000   \n",
       "std                                  97.005368   \n",
       "min                                1630.000000   \n",
       "25%                                1765.750000   \n",
       "50%                                1818.000000   \n",
       "75%                                1876.750000   \n",
       "max                                2099.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-3_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                824.330000   \n",
       "std                                  48.525127   \n",
       "min                                 700.000000   \n",
       "25%                                 789.750000   \n",
       "50%                                 826.500000   \n",
       "75%                                 858.250000   \n",
       "max                                 933.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-4_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                416.610000   \n",
       "std                                  31.687201   \n",
       "min                                 344.000000   \n",
       "25%                                 395.000000   \n",
       "50%                                 414.500000   \n",
       "75%                                 437.000000   \n",
       "max                                 513.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-5_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.910000   \n",
       "std                                   1.436149   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   3.000000   \n",
       "max                                   7.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-6_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.920000   \n",
       "std                                   1.220035   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   3.000000   \n",
       "max                                   7.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-7_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.810000   \n",
       "std                                   1.134358   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   2.000000   \n",
       "max                                   5.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-8_epoch   last_epoch  \n",
       "count                               100.000000   100.000000  \n",
       "mean                                  1.950000  2421.460000  \n",
       "std                                   1.104399   107.789771  \n",
       "min                                   1.000000  2071.000000  \n",
       "25%                                   1.000000  2355.750000  \n",
       "50%                                   2.000000  2429.500000  \n",
       "75%                                   3.000000  2493.000000  \n",
       "max                                   6.000000  2647.000000  \n",
       "\n",
       "[8 rows x 26 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_divide[epoch_keys_divide].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_divide_plot = df_run_info_divide[epoch_keys_divide].rename(index=str, columns={'init_all_correct_epoch':'all_correct',\n",
    "                                                                                                '{}_carry-0_epoch'.format(all_correct_case):'carry-0',\n",
    "                                                                                                '{}_carry-1_epoch'.format(all_correct_case):'carry-1',\n",
    "                                                                                                '{}_carry-2_epoch'.format(all_correct_case):'carry-2',\n",
    "                                                                                                '{}_carry-3_epoch'.format(all_correct_case):'carry-3',\n",
    "                                                                                                '{}_digit-1_epoch'.format(all_correct_case):'digit-1',\n",
    "                                                                                                '{}_digit-2_epoch'.format(all_correct_case):'digit-2',\n",
    "                                                                                                '{}_digit-3_epoch'.format(all_correct_case):'digit-3',\n",
    "                                                                                                '{}_digit-4_epoch'.format(all_correct_case):'digit-4'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_divide_plot.boxplot(column=['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3'])\n",
    "df_run_info_divide_plot[['carry-0', 'carry-1', 'carry-2', 'carry-3']].to_csv('result_statistics/epoch_divide_carry.csv', index=False)\n",
    "save_fig('epoch_divide_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "column_list =['carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "\n",
    "x = ['0', '1', '2', '3']\n",
    "y = df_run_info_divide_plot[column_list].mean()\n",
    "e = df_run_info_divide_plot[column_list].std()\n",
    "\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.xlabel('Carries', fontsize=font_size['xlabel'])\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_divide_carry')\n",
    "\n",
    "df_plot = df_run_info_divide_plot[column_list]\n",
    "df_plot.columns = x\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_divide_carry.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_divide_plot.boxplot(column=['all_correct', 'digit-1', 'digit-2', 'digit-3', 'digit-4'])\n",
    "save_fig('epoch_divide_digit')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Modulo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['init_all_correct_carry-0_epoch',\n",
       " 'init_all_correct_carry-1_epoch',\n",
       " 'init_all_correct_carry-2_epoch',\n",
       " 'init_all_correct_carry-3_epoch',\n",
       " 'init_all_correct_digit-1_epoch',\n",
       " 'init_all_correct_digit-2_epoch',\n",
       " 'init_all_correct_digit-3_epoch',\n",
       " 'init_all_correct_digit-4_epoch',\n",
       " 'init_all_correct_digit-5_epoch',\n",
       " 'init_all_correct_digit-6_epoch',\n",
       " 'init_all_correct_digit-7_epoch',\n",
       " 'init_all_correct_digit-8_epoch',\n",
       " 'init_all_correct_epoch',\n",
       " 'init_complete_all_correct_carry-0_epoch',\n",
       " 'init_complete_all_correct_carry-1_epoch',\n",
       " 'init_complete_all_correct_carry-2_epoch',\n",
       " 'init_complete_all_correct_carry-3_epoch',\n",
       " 'init_complete_all_correct_digit-1_epoch',\n",
       " 'init_complete_all_correct_digit-2_epoch',\n",
       " 'init_complete_all_correct_digit-3_epoch',\n",
       " 'init_complete_all_correct_digit-4_epoch',\n",
       " 'init_complete_all_correct_digit-5_epoch',\n",
       " 'init_complete_all_correct_digit-6_epoch',\n",
       " 'init_complete_all_correct_digit-7_epoch',\n",
       " 'init_complete_all_correct_digit-8_epoch',\n",
       " 'last_epoch']"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_keys_modulo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>init_all_correct_carry-0_epoch</th>\n",
       "      <th>init_all_correct_carry-1_epoch</th>\n",
       "      <th>init_all_correct_carry-2_epoch</th>\n",
       "      <th>init_all_correct_carry-3_epoch</th>\n",
       "      <th>init_all_correct_digit-1_epoch</th>\n",
       "      <th>init_all_correct_digit-2_epoch</th>\n",
       "      <th>init_all_correct_digit-3_epoch</th>\n",
       "      <th>init_all_correct_digit-4_epoch</th>\n",
       "      <th>init_all_correct_digit-5_epoch</th>\n",
       "      <th>init_all_correct_digit-6_epoch</th>\n",
       "      <th>...</th>\n",
       "      <th>init_complete_all_correct_carry-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-1_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-2_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-3_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-4_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-5_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-6_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-7_epoch</th>\n",
       "      <th>init_complete_all_correct_digit-8_epoch</th>\n",
       "      <th>last_epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4425.390000</td>\n",
       "      <td>4540.300000</td>\n",
       "      <td>3989.270000</td>\n",
       "      <td>3646.380000</td>\n",
       "      <td>4400.570000</td>\n",
       "      <td>4548.930000</td>\n",
       "      <td>3956.450000</td>\n",
       "      <td>1066.500000</td>\n",
       "      <td>2.030000</td>\n",
       "      <td>2.060000</td>\n",
       "      <td>...</td>\n",
       "      <td>3716.700000</td>\n",
       "      <td>4464.820000</td>\n",
       "      <td>4574.190000</td>\n",
       "      <td>4017.280000</td>\n",
       "      <td>1150.940000</td>\n",
       "      <td>2.030000</td>\n",
       "      <td>2.060000</td>\n",
       "      <td>1.860000</td>\n",
       "      <td>1.890000</td>\n",
       "      <td>4658.730000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>304.410472</td>\n",
       "      <td>273.112198</td>\n",
       "      <td>409.204053</td>\n",
       "      <td>308.930667</td>\n",
       "      <td>339.937108</td>\n",
       "      <td>264.570788</td>\n",
       "      <td>280.736131</td>\n",
       "      <td>113.743331</td>\n",
       "      <td>1.282792</td>\n",
       "      <td>1.269852</td>\n",
       "      <td>...</td>\n",
       "      <td>307.718568</td>\n",
       "      <td>307.053721</td>\n",
       "      <td>251.649054</td>\n",
       "      <td>274.343075</td>\n",
       "      <td>113.200976</td>\n",
       "      <td>1.282792</td>\n",
       "      <td>1.269852</td>\n",
       "      <td>1.146096</td>\n",
       "      <td>1.262473</td>\n",
       "      <td>258.331783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3949.000000</td>\n",
       "      <td>3951.000000</td>\n",
       "      <td>3142.000000</td>\n",
       "      <td>2838.000000</td>\n",
       "      <td>3680.000000</td>\n",
       "      <td>4010.000000</td>\n",
       "      <td>3295.000000</td>\n",
       "      <td>778.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2946.000000</td>\n",
       "      <td>3761.000000</td>\n",
       "      <td>4029.000000</td>\n",
       "      <td>3376.000000</td>\n",
       "      <td>904.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4030.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4166.250000</td>\n",
       "      <td>4394.250000</td>\n",
       "      <td>3649.000000</td>\n",
       "      <td>3441.500000</td>\n",
       "      <td>4115.500000</td>\n",
       "      <td>4399.250000</td>\n",
       "      <td>3776.750000</td>\n",
       "      <td>1001.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3495.250000</td>\n",
       "      <td>4228.500000</td>\n",
       "      <td>4414.750000</td>\n",
       "      <td>3857.750000</td>\n",
       "      <td>1072.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4466.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4367.500000</td>\n",
       "      <td>4527.500000</td>\n",
       "      <td>3969.000000</td>\n",
       "      <td>3656.000000</td>\n",
       "      <td>4355.000000</td>\n",
       "      <td>4546.000000</td>\n",
       "      <td>3970.500000</td>\n",
       "      <td>1074.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3723.500000</td>\n",
       "      <td>4451.500000</td>\n",
       "      <td>4562.000000</td>\n",
       "      <td>4022.500000</td>\n",
       "      <td>1148.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4654.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4620.250000</td>\n",
       "      <td>4695.000000</td>\n",
       "      <td>4285.750000</td>\n",
       "      <td>3839.250000</td>\n",
       "      <td>4655.750000</td>\n",
       "      <td>4695.000000</td>\n",
       "      <td>4153.500000</td>\n",
       "      <td>1138.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>3942.500000</td>\n",
       "      <td>4670.500000</td>\n",
       "      <td>4709.000000</td>\n",
       "      <td>4195.750000</td>\n",
       "      <td>1223.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.250000</td>\n",
       "      <td>4841.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5529.000000</td>\n",
       "      <td>5312.000000</td>\n",
       "      <td>4955.000000</td>\n",
       "      <td>4443.000000</td>\n",
       "      <td>5529.000000</td>\n",
       "      <td>5312.000000</td>\n",
       "      <td>4876.000000</td>\n",
       "      <td>1476.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4522.000000</td>\n",
       "      <td>5529.000000</td>\n",
       "      <td>5312.000000</td>\n",
       "      <td>4876.000000</td>\n",
       "      <td>1534.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>5529.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       init_all_correct_carry-0_epoch  init_all_correct_carry-1_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      4425.390000                     4540.300000   \n",
       "std                        304.410472                      273.112198   \n",
       "min                       3949.000000                     3951.000000   \n",
       "25%                       4166.250000                     4394.250000   \n",
       "50%                       4367.500000                     4527.500000   \n",
       "75%                       4620.250000                     4695.000000   \n",
       "max                       5529.000000                     5312.000000   \n",
       "\n",
       "       init_all_correct_carry-2_epoch  init_all_correct_carry-3_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      3989.270000                     3646.380000   \n",
       "std                        409.204053                      308.930667   \n",
       "min                       3142.000000                     2838.000000   \n",
       "25%                       3649.000000                     3441.500000   \n",
       "50%                       3969.000000                     3656.000000   \n",
       "75%                       4285.750000                     3839.250000   \n",
       "max                       4955.000000                     4443.000000   \n",
       "\n",
       "       init_all_correct_digit-1_epoch  init_all_correct_digit-2_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      4400.570000                     4548.930000   \n",
       "std                        339.937108                      264.570788   \n",
       "min                       3680.000000                     4010.000000   \n",
       "25%                       4115.500000                     4399.250000   \n",
       "50%                       4355.000000                     4546.000000   \n",
       "75%                       4655.750000                     4695.000000   \n",
       "max                       5529.000000                     5312.000000   \n",
       "\n",
       "       init_all_correct_digit-3_epoch  init_all_correct_digit-4_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                      3956.450000                     1066.500000   \n",
       "std                        280.736131                      113.743331   \n",
       "min                       3295.000000                      778.000000   \n",
       "25%                       3776.750000                     1001.000000   \n",
       "50%                       3970.500000                     1074.000000   \n",
       "75%                       4153.500000                     1138.500000   \n",
       "max                       4876.000000                     1476.000000   \n",
       "\n",
       "       init_all_correct_digit-5_epoch  init_all_correct_digit-6_epoch  \\\n",
       "count                      100.000000                      100.000000   \n",
       "mean                         2.030000                        2.060000   \n",
       "std                          1.282792                        1.269852   \n",
       "min                          1.000000                        1.000000   \n",
       "25%                          1.000000                        1.000000   \n",
       "50%                          2.000000                        2.000000   \n",
       "75%                          3.000000                        3.000000   \n",
       "max                          7.000000                        7.000000   \n",
       "\n",
       "          ...       init_complete_all_correct_carry-3_epoch  \\\n",
       "count     ...                                    100.000000   \n",
       "mean      ...                                   3716.700000   \n",
       "std       ...                                    307.718568   \n",
       "min       ...                                   2946.000000   \n",
       "25%       ...                                   3495.250000   \n",
       "50%       ...                                   3723.500000   \n",
       "75%       ...                                   3942.500000   \n",
       "max       ...                                   4522.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-1_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               4464.820000   \n",
       "std                                 307.053721   \n",
       "min                                3761.000000   \n",
       "25%                                4228.500000   \n",
       "50%                                4451.500000   \n",
       "75%                                4670.500000   \n",
       "max                                5529.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-2_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               4574.190000   \n",
       "std                                 251.649054   \n",
       "min                                4029.000000   \n",
       "25%                                4414.750000   \n",
       "50%                                4562.000000   \n",
       "75%                                4709.000000   \n",
       "max                                5312.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-3_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               4017.280000   \n",
       "std                                 274.343075   \n",
       "min                                3376.000000   \n",
       "25%                                3857.750000   \n",
       "50%                                4022.500000   \n",
       "75%                                4195.750000   \n",
       "max                                4876.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-4_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                               1150.940000   \n",
       "std                                 113.200976   \n",
       "min                                 904.000000   \n",
       "25%                                1072.750000   \n",
       "50%                                1148.000000   \n",
       "75%                                1223.500000   \n",
       "max                                1534.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-5_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  2.030000   \n",
       "std                                   1.282792   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   2.000000   \n",
       "75%                                   3.000000   \n",
       "max                                   7.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-6_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  2.060000   \n",
       "std                                   1.269852   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   2.000000   \n",
       "75%                                   3.000000   \n",
       "max                                   7.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-7_epoch  \\\n",
       "count                               100.000000   \n",
       "mean                                  1.860000   \n",
       "std                                   1.146096   \n",
       "min                                   1.000000   \n",
       "25%                                   1.000000   \n",
       "50%                                   1.000000   \n",
       "75%                                   3.000000   \n",
       "max                                   5.000000   \n",
       "\n",
       "       init_complete_all_correct_digit-8_epoch   last_epoch  \n",
       "count                               100.000000   100.000000  \n",
       "mean                                  1.890000  4658.730000  \n",
       "std                                   1.262473   258.331783  \n",
       "min                                   1.000000  4030.000000  \n",
       "25%                                   1.000000  4466.750000  \n",
       "50%                                   1.000000  4654.000000  \n",
       "75%                                   2.250000  4841.500000  \n",
       "max                                   6.000000  5529.000000  \n",
       "\n",
       "[8 rows x 26 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_run_info_modulo[epoch_keys_modulo].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_run_info_modulo_plot = df_run_info_modulo[epoch_keys_modulo].rename(index=str, columns={'init_all_correct_epoch':'all_correct',\n",
    "                                                                                            '{}_carry-0_epoch'.format(all_correct_case):'carry-0',\n",
    "                                                                                            '{}_carry-1_epoch'.format(all_correct_case):'carry-1',\n",
    "                                                                                            '{}_carry-2_epoch'.format(all_correct_case):'carry-2',\n",
    "                                                                                            '{}_carry-3_epoch'.format(all_correct_case):'carry-3',\n",
    "                                                                                            '{}_digit-1_epoch'.format(all_correct_case):'digit-1',\n",
    "                                                                                            '{}_digit-2_epoch'.format(all_correct_case):'digit-2',\n",
    "                                                                                            '{}_digit-3_epoch'.format(all_correct_case):'digit-3',\n",
    "                                                                                            '{}_digit-4_epoch'.format(all_correct_case):'digit-4'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_modulo_plot.boxplot(column=['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3'])\n",
    "df_run_info_modulo_plot[['carry-0', 'carry-1', 'carry-2', 'carry-3']].to_csv('result_statistics/epoch_modulo_carry.csv', index=False)\n",
    "save_fig('epoch_modulo_carry_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "column_list =['carry-0', 'carry-1', 'carry-2', 'carry-3']\n",
    "\n",
    "x = ['0', '1', '2', '3']\n",
    "y = df_run_info_modulo_plot[column_list].mean()\n",
    "e = df_run_info_modulo_plot[column_list].std()\n",
    "\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.xlabel('Carries', fontsize=font_size['xlabel'])\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_modulo_carry')\n",
    "\n",
    "df_plot = df_run_info_modulo_plot[column_list]\n",
    "df_plot.columns = x\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_modulo_carry.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,4))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_run_info_modulo_plot.boxplot(column=['all_correct', 'digit-1', 'digit-2', 'digit-3', 'digit-4'])\n",
    "save_fig('epoch_modulo_digit')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Compare add, subtract, multiply, divide, and modulo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = df_run_info_add[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Add'})\n",
    "df2 = df_run_info_subtract[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Subtract'})\n",
    "df3 = df_run_info_multiply[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Multiply'})\n",
    "df4 = df_run_info_divide[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Divide'})\n",
    "df5 = df_run_info_modulo[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Modulo'})\n",
    "df6 = df_run_info_fixed_random[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Fixed_random'})\n",
    "df7 = df_run_info_random[['{}_epoch'.format(all_correct_case)]].rename(index=str, columns={'{}_epoch'.format(all_correct_case):'Random'})\n",
    "\n",
    "df_init_all_correct = pd.concat([df1, df2, df3, df4, df5, df6, df7], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Add</th>\n",
       "      <th>Subtract</th>\n",
       "      <th>Multiply</th>\n",
       "      <th>Divide</th>\n",
       "      <th>Modulo</th>\n",
       "      <th>Fixed_random</th>\n",
       "      <th>Random</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3038.0</td>\n",
       "      <td>2854.0</td>\n",
       "      <td>5676.0</td>\n",
       "      <td>2439.0</td>\n",
       "      <td>4556.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>13640.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3329.0</td>\n",
       "      <td>2439.0</td>\n",
       "      <td>6332.0</td>\n",
       "      <td>2395.0</td>\n",
       "      <td>4734.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>15644.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3139.0</td>\n",
       "      <td>2998.0</td>\n",
       "      <td>6176.0</td>\n",
       "      <td>2515.0</td>\n",
       "      <td>4458.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>16985.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3246.0</td>\n",
       "      <td>2326.0</td>\n",
       "      <td>6160.0</td>\n",
       "      <td>2624.0</td>\n",
       "      <td>4690.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>18968.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2719.0</td>\n",
       "      <td>2600.0</td>\n",
       "      <td>6449.0</td>\n",
       "      <td>2393.0</td>\n",
       "      <td>4660.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>13834.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Add  Subtract  Multiply  Divide  Modulo  Fixed_random   Random\n",
       "0  3038.0    2854.0    5676.0  2439.0  4556.0           3.0  13640.0\n",
       "1  3329.0    2439.0    6332.0  2395.0  4734.0           1.0  15644.0\n",
       "2  3139.0    2998.0    6176.0  2515.0  4458.0           4.0  16985.0\n",
       "3  3246.0    2326.0    6160.0  2624.0  4690.0           4.0  18968.0\n",
       "4  2719.0    2600.0    6449.0  2393.0  4660.0           3.0  13834.0"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_init_all_correct.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,6))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_init_all_correct.boxplot(column=['Add', 'Subtract', 'Multiply', 'Divide', 'Modulo'])\n",
    "df_init_all_correct[['Add', 'Subtract', 'Multiply', 'Divide', 'Modulo']].to_csv('result_statistics/epoch_all_operators.csv', index=False)\n",
    "save_fig('epoch_all_operators_boxplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATAAAAEFCAYAAABkTLvtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XeYFFXW+PHvQYkGgsBIEMbAmgERBYwDoqKuYtZ3UcGwvLorpl1XzIjyqrv6MydYkuwYWNeAYgCFQV0FFlQQA4JIFAkSZUgj5/fHrV6aoWe6a6a6q6v7fJ6nn5muvlN1mhlOV92691xRVYwxJopqhB2AMcZUlSUwY0xkWQIzxkSWJTBjTGRZAjPGRJYlMGNMZFkCM8ZEliUwY0xkWQIzxkTWrmEHEJbGjRtrYWFh2GEYk5emT5++UlWbVHc/eZvACgsLmTZtWthhGJOXRGRBEPuxS0hjTGRZAjPGRJYlMGNMZFkCM8ZEliUwY0xkWQIzxkSWJTBjTGRZAjPGRFbeDmQ1ETRzAMy6Z+fth90NbQdkOhqTBSyBmehoO8A93i9yz7uXhBeLyQp2CWmMiSxLYMaYyLIEZoyJLEtgxpjIsgRmjIksS2DGmMiyBGaMiSxLYMaYyLIEZoyJLEtgxpjIsgRmjIksS2DGmMiyBGaMiSxLYMaYyLIEZoyJLEtgxpjIsgRmoqW4GHpNhlMmQWGhe27yllVkNdFRXAx9+0LpZvd8wQL3HKBXr/DiMqGxMzATHbffDqWlO24rLXXbTV7ydQYmIm2A64GjgYbALgmaqaruH0Bsxuxo4UJ/203OSzmBiUgX4H2gLlAGLPO+7tQ0mNCMKadVK3fZmGi7yUt+LiHvB2oDVwP1VHUfVd030cNPACLSQEReEZFvReQbEekiIo1EZLyIzPG+NvTaiog8LiJzRWSmiHSI209vr/0cEentJwYTEYMGQd26O26rV89tN3nJTwI7CnhFVQeraqIzr6p6DHhXVQ8C2gHfAP2BD1S1DfCB9xzgNKCN9+gLPAMgIo2Au4FOuMvbu2NJz+SQo46CjRthT+/CoX59GDzYOvDzmJ8+sC1AoJ0NIlIfOAHoA6CqW4AtItITKPKajQRKgFuAnsDzqqrAZO/srZnXdryqrvL2Ox7oAbwYZLwmZHvtBY89Bk1fhMa1bV1I4yuBfQIcEfDx9wVWAMNFpB0wHXeToEBVl3ptfgIKvO9bAIvifn6xt62i7TsQkb64MzcKCgooKSkJ7I2YDGnblvYrh8GajXxRUkKNTZvYVqdO2FGZkPhJYLcBn4jIpao6KsDjdwD6qeoUEXmM7ZeLgLulKSIaxMFUdTAwGKBjx45aVFQUxG5NJixdCtOnw8knw0cNACh65hm3/cMPQw7OhKXCBCYidyXYPAEYISJX4c6W1iRoo6p6b4rHXwwsVtUp3vNXcAlsmYg0U9Wl3iXicu/1JcA+cT/f0tu2hO2XnLHtJSnGYKJg9Gi44QaYM2f7tjPOgLVrw4vJhE5cd1KCF0S2VXGfqqqJxodVdJyPgKtUdbaIDAB28176WVUfEJH+QCNV/YuInAFcC5yO67B/XFWP9jrxp+PO5gA+A46M9Ykl0rFjR502bZrvN2dCsmULfPopnHgivF/ktlkfWGSJyHRV7Vjd/VR2Cdm1ujtPUT+gWERqAfOAy3F3R0eLyJXAAuBCr+3buOQ1Fyj12qKqq0TkXuA/XruBlSUvE0G1arnkVd6mTe6s7PDDMx+TCV2FCUxVJ2UiAFX9AkiUiU9K0FaBP1awn2HAsGCjM1lhzBiYOxeuuw52Lfcn+8c/wuuvw/LlsEvKJ/4mR9hcSJP93n4bnnsucYK65hoYNQoq6Aoxuc3PVKKTgF7AHar6Y4LXmwP34cZplQQWoTHPPus66yXBLLWO1e5GMRHm5wysH3BMouQF4G3v4rUzJlj161f82sKF7izM5B0/CawDbjBrZT4mcX+WMVVz1VUwcGDlbUaPhssugyVLMhOTyRp+ElhTIOHZV5xlXjtjqk8VNm92Qygq07u3uxPZYqfJFybH+RmJv5YdB5Emsg+woerhGBNHJLVLwyZN3MPkHT9nYFOBs0Vk70Qvep34Z3vtjKm+8tVXK/PVV26YxcaN6YvHZB0/CewJYA/gIxE5S0RqA4hIba96xIfA7sDjwYdp8s7GjdC8OTz6aGrtFy+Gv//dJTKTN1K+hFTVcd5o9zuB1wAVkdW40tLiPe5V1XfTEqnJL5s2uUGqnTql1r5bN/j5550LHpqcVuFcyAp/QOQU3FCJTkAD3ITuycATqjo+8AjTxOZCRtDMATDrnp23H3Y3tB2Q6WhMNQQ1F9J3AssVlsCy2JYt8OWX0KFD4sGrFfnqK7jxRvh//w8OOyx98ZlqCyqB2VQik30mTHAj7N/12RvRsKHrC1u+PHlbkxN8L2wrIq2Ay3DVWRvghld8BoxS1QRLxhjjU6dOMHw4dPVZEKV5c/j66/TEZLKSr0tIEfk97i5jLXZePm0LcL2qPhdceOljl5A5TBW2bbPqFFks45eQ3mTuZ4HNwCCgG3Cw9/U+YBPwlNfOmKr58kt48UV3F7IqFi6EwkI3vcjkPD99YDcD63GVTu9S1RJVne19vQs4EvjFa2dM1RQXwxVXQFkVV+5r2RJOOAH2Tjje2uSYlC8hRWQVbl3IvpW0GQKcp6qNAoovbewSMktt2wbffQcHHRR2JCaNwrgLWRdYmaTNCq+dMVVTo0YwyWvdOliZ7M/VRJ2fBLYA199Vma4EvPitySNPPAH331/9/WzcCM2awcMPV39fJqv5SWCvAUeJyNMi0iD+BRHZ01vT8Wjg1SADNHlkyhT46KPq76duXXjoITjvvOrvy2Q1P31gewKf4u48rgdmAEuBvYF2wJ7At0BnVV2XlmgDZH1gWaqsbOeFO0zOyXgfmJeUjgGGALsAxwEXAMfjBsQOAY6NQvIyWSj2QRpU8lKFmTPhs8+C2Z/JSr6mEqnqWlX9X1wFira45NUWaKiq/6uqq9MQo8kHxx8PDzwQ7D7PPRfuSrTAvMkVVfq4U9WtwKyAYzH5avNmOPDAYMduicA//gGtWwe3T5N1qpTAROR43FzI+ri5kJ+ragC9ryYv1a4NQ4cGv9/OnYPfp8kqvhKYiByLW/36gNgmQL3X5gBXqGqylYuM2dFPP6Vv5Pybb8Lq1W7VIpNz/MyFPBIYD7TBlY8eCFzjff0I+A0wXkQ6pCFOk6vmz3djtkaOTM/+hw519cFMTvJzBjbIa99TVd8s99o9Xl38V7x2pwUUn8l1u+3mBq+ecEJ69j9kiKsTZnKSnwR2DPBqguQFgKq+ISKvAacGEpnJD02aQP/+6d2/yVl+hlFsA+YmaTMHr0/MmKRWroSJE6teeSJVI0bAH/6Q3mOYUPhJYNNwI+4r0w5bF9Kk6p//dKsJzZ6d3uPMn+8GtG7dmt7jmIzzk8DuAE4WkWsSvSgifwROwi27Zkxyl10Gb70FhxyS3uPcfTdMngw1a6b3OCbj/PSBnQJMAJ4UkRtwdx6XAQW4aUVtgHeBU0Ukvh9MVfXegOI1uWS33eCMM9J/nNjKRqr+VjkyWc/PZO5tVTyGqmrWFSe3ydwhGzcOfvjBVV/NxJnRkCGuvM6sWTZZPAsENZnbz2/S5xIxxlTin/+E99+HvhUW+A1W8+Zuqba1a2GvvTJzTJN2trCtCYcqrFgBTZuGHYkJgS1sa6JNJJzk9fPPmT+mSRvfCUxE2orIAyLyhoi8H7e9UEQuFBFfw55FZL6IfCkiX4jING9bIxEZLyJzvK8Nve0iIo+LyFwRmRk/bUlEenvt54hIb7/vy2TQDTcEUzrar1Gj3MDWhVb1PFf4SmAiMhC3CvdfgDPZsV+sBvAicEkV4uiqqu3jTin7Ax+oahvgA+85uClKbbxHX+AZL65GwN1AJ1xZ67v9JlKTIaqwZAksX575Yx97LNxzj6t+YXKCn8ncF+PGgo0H2gM7fISq6jzcYNezAoirJxCb3TsSODtu+/PqTAYaiEgz3PSl8aq6yiuqOB7oEUAcJmgirgM/jAnW++0Hd94JBQWZP7ZJCz9nYNfhphL1VNWZwJYEbb7BnR35ocA4EZkuIrFbUgWqutT7/ifcWDOAFsCiuJ9d7G2raLvJNrHR8GGNx/r1V/j3v93KRSby/AyjOBwYoaqJElfMj2xPNqk6TlWXiEhTXDmeb+NfVFUVkUBulXoJsi9AQUEBJSUlQezWpEi2bqXLRRex4JJLWHLuuaHE0HD6dNr9+c98+X//x89duoQSgwmOnwQmuAndlSkANvkJQFWXeF+Xe9UsjgaWiUgzVV3qXSLGOkyWAPvE/XhLb9sSoKjc9pIExxoMDAY3jKKoqKh8E5NOXmHBNj170iasf/suXaBlSw7v0QPq1w8nBhMYP5eQc3AldRISkRq4KUVfpbpDEdlNRPaIfY+brjQLGAPE7iT2Bt7wvh8DXObdjewMrPUuNd8DThGRhl7n/SneNpNNGjaERx+FriGOia5dGy66yJJXjvCTwEYDHUTkTxW8fhuu1PQLPvZZAHwsIjNwVSzGquq7wAO4ieNzgO7ec4C3gXm4vrghwB8AVHUVcC/wH+8x0NtmskVZGXzzTdhROOvXw/Dh8P33YUdiqsnPXMi6wL9xJXOm4TrfjwIewS2v1hGYDJyoqmku8FR9NhI/wyZOdKVz3norMxO4K/Pjj9CihbsTeuON4cYSdTMHwKx7dt5+2N3QdkCFPxbUSHxfU4lEpD7wGNALt7htzDagGLhWVddXN6hMsASWYStWwMsvw+WXuyoUYfv2W7eUm1WnCMb7Re5r95KUmocxmRtVXQv0EZGbcGdfe+GWVZuqqiuqG4zJYU2awLXXhh3FdgcdFHYEJgBVXdh2FdZJblI1e7Z79OgBtWqFHY2zYQPcd59bTOQ0W4Mmqmwyt0m/kSPh/POhtDTsSLarW9et3D19etiRmGqwym4m/e65B847Dxo0CDuS7WrUcHchs+WM0FSJnYGZ9KtZE448MuwodmbJK/IsgZn0GjoUHnkk7CgSU4WLL4Z7bcmGqLIEZtJr3DgYMybsKBITcWdhu2Tdkg0mRdYHZtLr5Zezu/LD88+HHUH0FRfDTZNhxWZoVQiDBkGvXhk5tJ96YMNEpNJaXyLyWxEZVv2wTE6pWzfsCJLb5KsGgYkpLnYLsyzf7ObmLFjgnhcXZ+Twfi4h++AKGVamHdsnYZt8d/rp8NBDYUeR3Jlnurukxr/bb995eExpqdueAUFfQtYGfg14nyaKtmyBPfeEevXCjiS5M85wHfrGv4rWF8jQugN+E1iFv2URqQ2cgKugavJdrVrw0kthR5Gaq68OO4LoatXKXTYm2p4BlV5Cisi82MPbdGP8trjHAmA1rirFm+kO2kTA2rVhR+DPpk3ZU+4nSgYN2vksu149tz0DkvWB1cBVYhXc2ZdU8NgKfAk8CNycrmBNRPz4IzRu7KYQRUXv3nDKKXYp6df//A8MHgxNa7tM0Lq1e56hu5CVXkKqamHsexHZBjyiqgPTHZSJuF12gVtvdeWbo+L6610SU7USO36ceSYceigUd3bPUyynExQ/fWBdgflpisPkkoICGBixz7ljKqyWbipSVgYHHAD77IMrqJx5KQ+jUNVJqrpTb52I1BSRI0TkwGBDM5G0Zg1MmRLNS7EFC9zAW5OaXXeFxx6Dfv1CC8HPQNYLRWS0twp2bNv+uEU8pgFfi8irImKj+/PZq69C584wY0bYkfg3fDj87newypZTSOq77+DLL8OOwtdA1iuAg8otlvEwbiGPicBM3MrZlwcXnomcc85xwyfatQs7Ev/69nWFFxs1St423w0Y4IpBhjxNzM/Z0iHA+NgTEdkTOB0YraoXi0hN4AtcAhsSaJQmOho2dMuWRVHz5mFHEB1PPunOskOeJubnDKwJsDTueRdcAnwJQFW34hLc/oFFZ6Llww9dldOtW8OOpOpmzICbbnId1KZijRqFu76nx08CWw/ErwZ6Im5s2Mdx2zYBewQQl4mikSPhL3+JdnmauXPhuefcqkVmZ7G1DebMCTsSwN8l5BzgNG/KkAIXAjNVdWVcm9bA8gDjM1EyZAgsWuTKNUfVmWfCzz9DnTphR5Kdvv/eJbHYyubl14V8wRtDl2RdyKD4Wdi2NzAcWIwbeV8I3Kiqj8e1+Q6Yo6ohr1yanK0LaUwVbdtW7Q+poNaF9DMObCTwAFAPdyn5JPBEXEDHsP2OpMk3d9zhVrrOBZ9/7qYVzZ8fdiTZ5euv3fi+LDrD9hWJqt6mqo29x/W64+nbNKAh8GigEZpo+PLL3Ok32n13WLzYzek0zqJF0L49PPhg2JHsILBBp6q6BdgS1P5MxLzxhru0yAVt2rizDbNdQQE89RScemrYkezAdwITkbbA74CDgd1Utbu3vRA4GhivqqsDjNFku1ifSBZdWgRCNesumUJTqxb8/vdhR7ETX78ZERkIfAb8BTgTN8E7fl8vApcEFp3JfmVl7ozlqafCjiRYs2dDYSGMHRt2JOF7/HF45ZWwo0jIz1zIi4E7cINV2wP3x7+uqvNw/WCVLvxhcsz69dCtG+y3X9iRBGvffaFTJzezIJ9t2wajRmXt0nh+LiGvA+YCPVV1i4ick6DNN0BREIGZiGjY0I3/yjW1asHo0WFHEb4aNVx1kfXrw44kIT+XkIcD73md9RX5ESioXkgmMrZtS1wPPZesWwer87RLt7TUTQurUWP7wNUs4yeBCZDsNlMBbjqRyQeTJ7t+orfeCjuS9FizBpo0gWeeCTuScPz1r/Cb32Tt2Rf4n0pUYdlKEakBHIerD2bywX77uT/y448PO5L0aNDAjXs68cSwIwnHcce5mzR7ZO/0Zj8JbDRwn4j8SVUfTvD6bbiR+I8FEpnJfnvvDTfn+BouN9wQdgTh6d7dPbKYn0vIR4EZwF9FZApwGoCIPOQ9vweYDAwOPEqTfX74Ad5/P/fLzqjCzJlZUX00Y375xZWK3rAh7EiS8jMXciNu3NcooANu0KoANwFHAv8Aeqhqjv9FGwCGDXOjstesCTuS9FJ18yIztM5hVnjrLXfmGYGknXI1ih1+yNXFPwrYC1gLTFXVFVUOQmQX3BiyJar6WxHZF1cocS9gOnCpN3SjNvA8LmH+DFykqvO9fdwKXAn8Clynqu9VdkyrRlFNpaUwbZorK5zrPvrIrb7TrFnYkWTOrFlw2GFp233Gq1HEU9VVqvqeqr6gqmOrk7w81+PGkMU8iFuD8gDcit9XetuvBFZ72x/x2iEihwAXA4cCPYCnvaRo0qVevfxIXuBuUuRL8orNZ01j8gpS6JO8RKQlcAbwd++5AN2A2NyFkcDZ3vc9ved4r5/kte8JvKSqm1X1B9yA26Mz8w7y0IsvuqqlUVw6rapefz33B7Zu3uwWYxkxIuxIUpbyXUgRGZZiU1XVK5M3+69HcXMrY/dq9wLWxPWlLQZaeN+3ABZ5BykTkbVe+xa4Gwgk+Jn499AX6AtQUFBASUmJjzBNzKHPPEOtVav4/MD8WQq03b33UmPzZj5v2jTsUNKm5urV/KZRI35cuZLVEfm/4WcYRZ8kryuuU1/ZfslXKRH5LbBcVaeLSJGPWKpEVQfj3SXt2LGjFhWl/ZC5adIkWLeOoiwdnZ0WY8dC48YU7Zrjy56ecw5Nwo7BBz+/jX0r2N4A16F/J/AJ0N/HPo8FzhKR04E6wJ64cWQNRGRX7yysJbDEa78E2AdY7C2gWx/XmR/bHhP/MyZoIlk7tSRt9t477AjSa8oUaNUqcn19foZRLKjgMUNV/44bhd8DSHnkm6reqqotVbUQ1wk/QVV74cpSn+816w284X0/xnuO9/oEryrsGOBiEant3cFsA0xNNQ7jw4UXujFC+WjIEPjzn8OOIniqcPnlcN55YUfiW5AVWReJyJu4O4pDq7m7W4CXROQ+4PO4/Q0FRonIXGAVLumhql+JyGjga6AM+KOq/lrNGEx5W7fCpk3RXvexOr79Fv7zn0AWtcgqIvDmm5GctF6lcWAV7kzkb8C1qhrucr0psHFgxrdcS1whCnUcWCLeuKtuuIGtJhdt3Bh2BOHKxeT12mtwxRWwNpr/bf1UZD2hgkc3b83ID3CVWt9IsisTRStWQOPG8PzzYUcSriefhCOOyJ0xcAsWuGXkdt897EiqxE8fWAluiERFBPgQyPHyBHmqrAyuvho6dAg7knA1bepGqf/yS1aXmUnZDTdAv36wSzQnrvhZmXsAiRPYNtx0n6mqGpk7f5X2gZVfLj0mQ8ulG5N2qvDddxDSYOSg+sAC7cSPkpQ68d8vcl+7l6Q7nOy2fr271Dj0UHfHyrgqHA0ahB1F1b33HvToAe++G8paj1nXiW9y2JgxcPjhMH162JFkh8GDXX/giurWMAjRUUfBAw9A167J22axHJ8XYQJx8sluEGe+93/FHHss3HVXtM9GGzWCW24JO4pq8zOZe14Vj6Gqun8Vf9Zkg6ZN4aqrwo4iexx6qHtE1a23Qs+e0Llz2JFUm59LyBpAbaDQe+wD1PW+xrbVxt2NjH/YZWqUTZ3qxgrleulov379Ff797+jNSli2DIYOhU8+CTuSQPhJLm1xE6Qn40pL11HVZrhJ2N2AKbgyNm1Vdd/4R9BBmwx67jno2zfsKLLPmDFu1Z6oJYKCApg/H665JuxIAuEngQ3CVZ4oUtVJsbmGqvqrqpbgklojr53JFc8+60oq53oZGb+6d4eXXnKDWqNi/Xo3fKJePaib9bP9UuIngZ0DvFHRytyqugk3Cv/cIAIzWaJmTTjooLCjyD577AEXXQR77hl2JKm79FI47bTcmUWAv7uQewE1k7Sp6bUzueD++13drz/8IexIstPatfDqq+4ubcuWYUdTOVX47W9dn12U756W4+cM7HvgfBFJWMlORBrianRV9W6lyTYffgiffhp2FNlrxQo3Efrtt8OOJDkRdyc5R/q+YvycgT0LPA5MFZFBuHmPy4AC4ETgdmBvrA8sd7zzTvTusmXSAQe4tROzfUjF7NluwvYFF0R2zmNFUk5gqvqkiLQB+gHDEzQR4AlVfTqo4EyIVN2nds1kvQZ5LgrLj/397/D00+7GQ+PGYUcTKF9jtFT1elwd+2G4Sqnz2F4x9TjvdRN127a55bWefTbsSLLf2rXQvz9k8yo+Dz4IkyfnXPKCKkwlUtVPAesYyWVr17ozi4KCsCPJfnXrumlWjRtDNq5ypeoKMR5+eNiRpIUN7jE7a9gQXngh7CiioVYt+PFHqF077Eh2Nn8+nHIKDBvmBt3mIN8JzCsdfSDQEEjYI6iqH1YzLhMWVXd3LYcXcA1cNiYvcGfSzZpBYWHYkaSNrz4wEbkTWAF8ibsLObGCR/QVF0OvyXDKJPcHUFwcdkSZ8dln7o/+rbfCjiQ6ysrckmSPPBJ2JDtq184tQpztY9SqwU81ir8A9+AW7RgFLMItYZZ7iovd/L/Sze75ggXb5wP26hVeXJlQUAB33glduoQdSXTsuqs7c922LexItps4EY4+GnbbLexI0spPSek5uInbHVQ1wpXcnEorshYWuqRVXuvWrl/BmGy2ejW0aOEWq33qqbCjSSioiqx++sD2AYbkQvJKauFCf9tzxeLFsGgRdOqUm0uIZcLmzeH3iTVsCB98AHvvHW4cGeDnr3QZ+XLXslWrirevX5/ZWDJp+HA45hj46aewI4mmk0+Gyy4LOwqnSxfYN/crWflJYKOBk0UkS2+5BGjQIFdyJF69em75qZYt4fXXw4kr3fr1g7FjoXnzsCOJptNPh5NOCjeGv/7VlbvOoYoTlfFzRnU30Bl4RUSuU9Uf0hRT+GId9TddCSs2Q6vWLql17QpffbW9g3vdOldWJVdm9zdo4P4Tmqq58cawI3BLpa1blzt/k0n46cSfhyuXE/t4XgusSdA0EjXwA1lW7eyzobTULVEV9T+Y1193l8eXXBL99xKmTZtcP2KbNuHFUFaW9QUow+jEr4EbNhHfk53oLz0//vpVXQLbtGn7f/jVq10HahSNGOE68S+9NOxIou2CC2DuXPjmm8we95df3AdQs2ZZn7yC5KcaRWEa44geEejTZ/vziRPhzDPd2dixx4YWVpW99hosXx52FNF3002wceP2ah6Z8tRTMGAAzJmT0wNXy8ufVJ1urVu7y6/Y2ok//+zW3ovK5ZiITd4OQlgLxZ5/vhu+kUfJC6qx5JmI7CEi+4hIhIqCp9F++7nyM3XruiW3unff8Qwtm111lasXZYIxf37m71Tvvz/ccENmj5kF/M6F3FVE+ovIXFwH/nxgtYjM9bbbGV1M375wrre+ybZt7owsG5WVuVkHK1eGHUnueOIJt+DHhg3pP9amTXD99fBD7g4KqIyfu5C1gHdx5aMVtwbkUqAZ0BLXef8RcEpFKxdlk0DuQqZq2DDXNzJ1KvzmN9XbV7pkus8mly1cCFu2uJLT6fbxx24A7dix0K1b+o8XkKDuQvo5A7sJKALGAgeraqGqdvE69w8E3gSO99qZeF26uMu02K31ZcvCjSdebMVtS17BadUqM8kLXJ2vRYvC63sLmZ8E9jtgFnC2qs6Jf0FVv8etB/kVkOPlGqrg4IPhoYdckli/3i2G2r9/2FHBmjWu7tc//hF2JLnns8/gllvSOyL+l1/c18aN8/YDyE8COwB4R1UT1gzxtr8DZP0g1lDVrg1//jOcc457vmGDSyRhKC2FCy+EQw4J5/i5bOZM1xeWrr6psjJo3x5uuy09+48IPwlsC7B7kja7ASmvwyUidURkqojMEJGvROQeb/u+IjLFuznwstf/hojU9p7P9V4vjNvXrd722SJyqo/3lVm1arn+sE6d3PMHHnD9YqtWZT6W5s3dndPY0A8TnIsucjdu9tsvPfvfutUN2znhhPTsPyL8JLCZuIVtmyR6UUQa4xa2neFjn5uBbqraDmgP9BCRzsCDwCOqegCwGrjSa38lsNrb/ojXDhE5BLgYOBToATztlb7Ofuee6xJao0bu+aJFmTluaWne3rnKiLp13SOd+x8wAHr0SN8xIsBPAnsSaIJb2PZKEdlPROp6Z0uXA1O815/1IqnwAAALj0lEQVRMdYfqeBfy1PQeCnQDXvG2jwTO9r7v6T3He/0kERFv+0uqutmbZD4XONrHewtPfH/YwoVw4IHw6KPpP+7Yse7sYMqU9B8rX02Z4hbVCHqGw6RJ7u6j8TWVaLSItAf6A4MTNBHgr6o62k8A3pnSdFwf21PA98AaVY2Vq14MtPC+b4ErZY2qlonIWmAvb/vkuN3G/0x0NGni+jRi48dWroQ6dWD3ZFfuVXDMMS5Rdqz2nWxTkZo1txeJDHKRlIEDYelSmDUr7wtP+hp4qqq3icgY3KXcEUB9XFWKz4Fh3pqRvqjqr0B7EWkAvAYc5HcfqRKRvkBfgIKCAkqSLEba3utc/yKTi5YedxzMmwfz5nHwoEHs+fXXTB0xAq3iCtmF60ZQ+MvInbbP370389v1gY8+qmbAplJPP+3uPAf4N1Tj5pups3w5pR/a4l9VWdh2Mjue7QRCVdeIyESgC9BARHb1zsJaAku8Zktwpa0Xe6P+6wM/x22Pif+Z+GMMxjt77NixoxYlW4j0/QYAJG2XLnXqwKxZnHjyye75vHlV6BQuAkbsOCh3xgwKf/qJwuOOy6vKBaGJDaWo7lCHoPaTQ0I9/xSRJt6ZFyJSFzgZ+Aa3NNv5XrPewBve92O853ivT1A3lWAMcLF3l3JfoA0wNTPvIo06d3YDYMGN4j/gAHj55erv96mn3F2ybFpFJ1fNnOkWiZk0qfr7eucdd/c619dm8KHSj19v+MLHwHqgh6omHCLhtXsHN4zi+IraJdAMGOn1g9UARqvqWyLyNfCSiNyHuzwd6rUfCozy5mKuwt15RFW/EpHRwNe4mmV/9C5Nq2bmAJh1z/bnL3ifeIfdDW0HVHm31XLwwe6uU6xi6qJFrs+sTh3/+3rsMbj6ajekw6TXvvvCkUdW7fdU3q+/uv7QZs2qv68cUelcSBG5AhgCnKmqb1e6I5EewNvAFao6Isgg0yGluZDZShWOP979QX/ySWqXFEHN6zQmAJmaC3kuMC9Z8gJQ1XeBOcAF1Q3KJCEC997rhl+IuIQ2e3ZqP/v44zB0aPJ2Jljr1rlHVY0b5z6wzA6SJbAjgBIf+/sQNyDVpFvXrtCzp/v+lVfcdKBU7kq9/jq8+256YzM7WrrUzVccMaJqP//pp3DqqW7ZO7ODZLegGuPWg0zVMty4LJNJ3bu7VZNipay/+851HCfq45owwY3CN5nTrJk7Yz7xxKr9fKdO7kPqtNOCjSsHJEtgG0k+/zHe7sCmqodjqqRhw+2j+bdscZ/WHTrAv/6VuH35NS9N+t1yS9V/tkYNOO+84GLJIckuIRcBfjraOrLjqkUm02rWhCefdPMrwS11/9BD0GsynDwJ9toLiovDjTEfqbohFan2VcZccw2MGpWemHJAsgRWAnQRkaRJTESOBI7BjeEyYRGBM87YfjnZpw/cfDMs3+yer1rlyl1bEsussjJ35/ihh1L/mY0bYcYMG/dViWTDKA7EFSlcBJyuqgkXuxORg3BDKPYBDlNVnx8zmRfpYRR+7LOPm49XXuvWbvEJkzkTJrjxfH7Gcam6u485NmMiIwvbqupsERkIDAA+F5FXgAm4ydLgJkyfBJwH1AbuikLyyitLdppR5dineub5qVn/00+wxx6w2245l7yClPRfRlUHikgZcDeurPT/lGsiuCKGt6vq/cGHaKqlVSu36lCi7Sbz/vUv10951lmVt7vxRleO57vvLIFVIqV/GVX9PxEpBq4AjsVNAQK3KtHHwHBVTfC/xIRu0CDX5xU/dKJePbfdZN7f/uamAyVLYP36uWKFlrwqlfKyarkmb/rAwHXY33QlrNgMrVq75NXL1l4JxZIlbgX0PE9MGekDMzmiVy8oGOK+t7mQ4WqRpM7mvHluHdE//cmN7zOVyu9yjsaE4emn4a67Er82bhw8/LAbQmGSsgRmTKbNmOHmNybqvrn6anfTpXnzzMcVQXYJaUymPf007JJg0awNG9ywiSDr5+c4OwMzJtMSJa+lS6FlS3jxxczHE2GWwIwJw8MPu6lF8S64AI6OxmqA2cIuIY0Jw4YPQT+GEQKxqkdFwIbmuIkvJhWWwHJdNtb3N3DXG3AXrtT3RyvgktehTZuwo4ocG8hqTFiKi+GGK2HlZjc7YvDgvBlgbANZjYmy4mK4/HLY6i3gVVrqpnxB3iSxIFgnvjFhuP327ckrprTUbTcpswRmTBgqKmdkZY58sQRmTBgqKmdkZY58sQRmTBgGDdp5cRUrc+SbJTBjwtCrl7vr2LS2KwnaunVe3YUMit2FNCYsVuao2iyBGRMGG2AcCEtgxoSh7QBLVAGwPjBjTGRZAjPGRJYlMGNMZFkCM8ZEliUwY0xkWQIzxkSWJTBjTGRZAjPGRFbeVmQVkRXAghSaNgZWpjmcTLH3kp1y5b34eR+tVbVJdQ+YtwksVSIyLYjSt9nA3kt2ypX3Esb7sEtIY0xkWQIzxkSWJbDkBocdQIDsvWSnXHkvGX8f1gdmjIksOwMzxkSWJTBjTNYRERWRkmTtLIEZYyLLEpgxJrIsgVVCRPp4p7JFYceSz0TkbBFpm6TNRSJycKZiMtnBEpjJaiLSABgOTBCRIypocznwAjA0k7FVl4iUeB+QI0I6fmHs+CKyv4i8IiI/i8h6ERknIod57ZqIyGARWSoim0TkPyLSNcH+6ovI/SIy22u3WkTeE5HuFRy/lojcKSLfi8hmEflBRO4TkdqpvgdLYCarqeoa4CygNvCBiOwwVUVEfo9LXPOACzMfYU4oBKYABcAIYBzQHSgRkTbAZOAo4GVgNNAOeEdE/ruMuPdB8wnQH1gLPAr8C+gCjBOR/40/oIiIt6+BgAJPAm8BV3jbU6Oq9qjgAfTx/nGLwo4l3x/AccB6YA3Q2dv2B2AbMBtoHnaMVXhPrYCDgGYhHb/Q+/tW4PZyr93pbV8FPAvUiHvtUu+1R+K2Pedtew5vfKm3vY2X0DYDhXHbf+e1/xSoE7e9EfC991pJ0vcQ9i8xmx+WwLLrARwDrPMej3u/m2/CSgBRf8QlsB+AXcq91sp7bQOwR7nXdgG2AhO957W8duuBRgmOc6+3r7vito33tnVN0D72/64k2XuwS0iPiMz3+gP++8D1vQBMLP9aWP0WqUj0XpI8/hF2zKlQ1U+AU3DrmfYD5uA+XJaGGlgKqvA7GZHB8L5Q1V/LbfvR+/qdqq6Pf8Fruwxo6W06EKgHzFDVVQn2P8H7Gt+H2QF39vxxgvYlqQZuC9tu9yjQoNy29kBPYCQwv9xrX2Qgpqr6Htjko/2PyZtkjWOBut73zYFDcP+Zsl2iv6/KZPLva235Dapa5rqpdn7NUwbU9L6v732t6IMktj3+/dcHVqnq1gTtf6o02jiWwDyq+mj5bSLSB5fARqhqSaZjqipVPSnsGNJBRPoD9wPTgFtwnb1jRaSnqo4PNbgkEv195ZBYktu7gteblWsX+76RiNRMkMQq2s9O7BLSRIKI3IFLXlOA7qo6AegG/AKMEZHTwoyvKsIeRhGg2UAp0M67G1lebMjFZ3HbPsPln+MStC9K9cCWwEzWE5EBuI7gfwMnq+paAFWdifvPsRZ4XUTOCi3IPKaqW4BiYA/c7+m/RGR/4Dpcp/+ouJdi/cuDRKROXPtGwB2pHtsuIU1WE5H6wGXAJOC3qvpL/Ouq+pWInIjrKO4LjMl8lFV2Ga7zu6J+pijpDxwPXCsiRwETcTXyL8QltmtV9Ye49i8CF+HG+M0SkTdwfWrnA/8B9k/loJbATFZT1bUicgKuw7e0gjazReQYotGZ/1+qujDsGIKiqqtEpAtwK3AucBOwEZgK/E1Vx5VrryJyAS7x9QGuxXX2D8cNbk3pJpQVNDTGRJb1gRljIssSmDEmsiyBGWMiyxKYMSayLIEZYyLLEpgxJrIsgRljIssSmDEmsiyBGWMiyxKYMSay/j/zieFK5iwC5wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#column_list = ['all_correct', 'Add', 'Subtract', 'Multiply', 'Divide', 'Modulo']\n",
    "column_list =['Add', 'Subtract', 'Multiply', 'Divide', 'Modulo']\n",
    "\n",
    "#x = column_list\n",
    "x = ['+', '−', '×', '÷', 'mod']\n",
    "y = df_init_all_correct[column_list].mean()\n",
    "e = df_init_all_correct[column_list].std()\n",
    "\n",
    "plt.figure(figsize=(len(x)-1,4))\n",
    "plt.ylabel('Conquest epoch', fontsize=font_size['ylabel'])\n",
    "plt.tick_params(axis='x', labelsize=font_size['xtick'])\n",
    "plt.tick_params(axis='y', labelsize=font_size['ytick'])\n",
    "plt.grid(axis='y')\n",
    "plt.errorbar(x, y, e, fmt='r:o', ecolor='orange', capsize=3)\n",
    "\n",
    "save_fig('epoch_all_operators')\n",
    "\n",
    "df_init_all_correct[column_list].to_csv('result_statistics/epoch_all_operators.csv', index=False)\n",
    "\n",
    "df_plot = df_init_all_correct[column_list]\n",
    "df_plot.columns = column_list\n",
    "df_plot = df_plot.stack().to_frame().reset_index(level=1)\n",
    "df_plot.columns = ['carries', 'conquest_epoch']\n",
    "df_plot.to_csv('result_statistics/epoch_all_operators.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.ylabel('Conquest epoch')\n",
    "df_init_all_correct.boxplot(column=['Add', 'Subtract', 'Multiply', 'Divide', 'Modulo', 'Fixed_random', 'Random'])\n",
    "df_init_all_correct[['Add', 'Subtract', 'Multiply', 'Divide', 'Modulo', 'Fixed_random', 'Random']].to_csv('result_statistics/epoch_all_operators_with_random.csv', index=False)\n",
    "save_fig('epoch_all_operators_boxplot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Carry: all-correct order "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "ordered_list = epoch_order_count(all_run_info_add)['{}_carry_epoch'.format(all_correct_case)].most_common()\n",
    "x_count = 0\n",
    "for _, x in ordered_list:\n",
    "    x_count = x_count + x\n",
    "    \n",
    "y_list = list()\n",
    "x_list = list()\n",
    "for y, x in ordered_list:\n",
    "    x_list.append(x / x_count * 100)\n",
    "    y_list.append(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6,6))\n",
    "y_pos = np.arange(len(y_list))\n",
    "\n",
    "ax.barh(y_pos, x_list)\n",
    "ax.set_yticks(y_pos)\n",
    "ax.set_yticklabels(y_list)\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.grid(axis='x')\n",
    "ax.set_ylabel('Conquest order')\n",
    "ax.set_xlabel('Percentage of cases (%)')\n",
    "ax.set_title('The conquest order of carry datasets: addition')\n",
    "\n",
    "save_fig('conquest_order_of_carry_datasets_add')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "ordered_list = epoch_order_count(all_run_info_subtract)['{}_carry_epoch'.format(all_correct_case)].most_common()\n",
    "x_count = 0\n",
    "for _, x in ordered_list:\n",
    "    x_count = x_count + x\n",
    "    \n",
    "y_list = list()\n",
    "x_list = list()\n",
    "for y, x in ordered_list:\n",
    "    x_list.append(x / x_count * 100)\n",
    "    y_list.append(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6,6))\n",
    "y_pos = np.arange(len(y_list))\n",
    "\n",
    "ax.barh(y_pos, x_list)\n",
    "ax.set_yticks(y_pos)\n",
    "ax.set_yticklabels(y_list)\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.grid(axis='x')\n",
    "ax.set_ylabel('Conquest order')\n",
    "ax.set_xlabel('Percentage of cases (%)')\n",
    "ax.set_title('The conquest order of carry datasets: subtraction')\n",
    "\n",
    "save_fig('conquest_order_of_carry_datasets_subtract')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "ordered_list = epoch_order_count(all_run_info_multiply)['{}_carry_epoch'.format(all_correct_case)].most_common()\n",
    "x_count = 0\n",
    "for _, x in ordered_list:\n",
    "    x_count = x_count + x\n",
    "    \n",
    "y_list = list()\n",
    "x_list = list()\n",
    "for y, x in ordered_list:\n",
    "    x_list.append(x / x_count * 100)\n",
    "    y_list.append(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x1296 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6,18))\n",
    "y_pos = np.arange(len(y_list))\n",
    "\n",
    "ax.barh(y_pos, x_list)\n",
    "ax.set_yticks(y_pos)\n",
    "ax.set_yticklabels(y_list)\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.grid(axis='x')\n",
    "ax.set_ylabel('Conquest order')\n",
    "ax.set_xlabel('Percentage of cases (%)')\n",
    "ax.set_title('The conquest order of carry datasets: multiplication')\n",
    "\n",
    "save_fig('conquest_order_of_carry_datasets_multiply')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "ordered_list = epoch_order_count(all_run_info_divide)['{}_carry_epoch'.format(all_correct_case)].most_common()\n",
    "x_count = 0\n",
    "for _, x in ordered_list:\n",
    "    x_count = x_count + x\n",
    "    \n",
    "y_list = list()\n",
    "x_list = list()\n",
    "for y, x in ordered_list:\n",
    "    x_list.append(x / x_count * 100)\n",
    "    y_list.append(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x144 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6,2))\n",
    "y_pos = np.arange(len(y_list))\n",
    "\n",
    "ax.barh(y_pos, x_list)\n",
    "ax.set_yticks(y_pos)\n",
    "ax.set_yticklabels(y_list)\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.grid(axis='x')\n",
    "ax.set_ylabel('Conquest order')\n",
    "ax.set_xlabel('Percentage of cases (%)')\n",
    "ax.set_title('The conquest order of carry datasets: division')\n",
    "\n",
    "save_fig('conquest_order_of_carry_datasets_divide')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "ordered_list = epoch_order_count(all_run_info_modulo)['{}_carry_epoch'.format(all_correct_case)].most_common()\n",
    "x_count = 0\n",
    "for _, x in ordered_list:\n",
    "    x_count = x_count + x\n",
    "    \n",
    "y_list = list()\n",
    "x_list = list()\n",
    "for y, x in ordered_list:\n",
    "    x_list.append(x / x_count * 100)\n",
    "    y_list.append(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6,6))\n",
    "y_pos = np.arange(len(y_list))\n",
    "\n",
    "ax.barh(y_pos, x_list)\n",
    "ax.set_yticks(y_pos)\n",
    "ax.set_yticklabels(y_list)\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.grid(axis='x')\n",
    "ax.set_ylabel('Conquest order')\n",
    "ax.set_xlabel('Percentage of cases (%)')\n",
    "ax.set_title('The conquest order of carry datasets: modulo')\n",
    "\n",
    "save_fig('conquest_order_of_carry_datasets_modulo')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Digit: all-correct order"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Add"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 7, 6, 5, 1, 2, 3, 4]', 20),\n",
       " ('[8, 6, 7, 5, 1, 2, 3, 4]', 17),\n",
       " ('[7, 6, 8, 5, 1, 2, 3, 4]', 16),\n",
       " ('[7, 8, 6, 5, 1, 2, 3, 4]', 14),\n",
       " ('[6, 8, 7, 5, 1, 2, 3, 4]', 10),\n",
       " ('[6, 7, 8, 5, 1, 2, 3, 4]', 9),\n",
       " ('[8, 7, 6, 1, 5, 2, 3, 4]', 6),\n",
       " ('[8, 7, 6, 5, 1, 2, 4, 3]', 2),\n",
       " ('[6, 8, 7, 1, 5, 2, 3, 4]', 2),\n",
       " ('[8, 6, 7, 5, 1, 2, 4, 3]', 2),\n",
       " ('[7, 6, 8, 5, 1, 2, 4, 3]', 1),\n",
       " ('[7, 8, 6, 5, 1, 2, 4, 3]', 1)]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_add)['init_complete_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 7, 6, 5, 1, 2, 3, 4]', 25),\n",
       " ('[8, 6, 7, 5, 1, 2, 3, 4]', 19),\n",
       " ('[6, 8, 7, 5, 1, 2, 3, 4]', 16),\n",
       " ('[6, 7, 8, 5, 1, 2, 3, 4]', 13),\n",
       " ('[7, 6, 8, 5, 1, 2, 3, 4]', 10),\n",
       " ('[7, 8, 6, 5, 1, 2, 3, 4]', 7),\n",
       " ('[8, 7, 6, 5, 1, 2, 4, 3]', 2),\n",
       " ('[6, 8, 7, 5, 1, 2, 4, 3]', 2),\n",
       " ('[6, 8, 7, 1, 5, 2, 3, 4]', 2),\n",
       " ('[8, 7, 6, 1, 5, 2, 3, 4]', 2),\n",
       " ('[7, 6, 8, 5, 1, 2, 4, 3]', 1),\n",
       " ('[6, 7, 8, 5, 1, 2, 4, 3]', 1)]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_add)['init_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Subtract"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[5, 8, 7, 6, 4, 1, 2, 3]', 6),\n",
       " ('[7, 5, 8, 6, 4, 1, 2, 3]', 5),\n",
       " ('[7, 6, 5, 8, 4, 1, 2, 3]', 4),\n",
       " ('[7, 6, 8, 5, 4, 1, 2, 3]', 3),\n",
       " ('[5, 6, 7, 8, 4, 1, 3, 2]', 3),\n",
       " ('[6, 5, 8, 7, 4, 1, 3, 2]', 3),\n",
       " ('[7, 5, 8, 6, 4, 1, 3, 2]', 3),\n",
       " ('[8, 5, 7, 6, 4, 1, 3, 2]', 3),\n",
       " ('[7, 8, 6, 5, 4, 1, 2, 3]', 3),\n",
       " ('[8, 7, 5, 6, 4, 1, 3, 2]', 3),\n",
       " ('[6, 8, 7, 5, 4, 1, 3, 2]', 3),\n",
       " ('[5, 8, 6, 7, 4, 1, 3, 2]', 3),\n",
       " ('[6, 5, 7, 8, 4, 1, 2, 3]', 3),\n",
       " ('[8, 7, 6, 5, 4, 1, 3, 2]', 3),\n",
       " ('[5, 8, 7, 6, 4, 1, 3, 2]', 3),\n",
       " ('[5, 7, 6, 8, 4, 1, 2, 3]', 3),\n",
       " ('[8, 5, 7, 6, 4, 1, 2, 3]', 3),\n",
       " ('[5, 8, 6, 7, 4, 1, 2, 3]', 2),\n",
       " ('[6, 7, 5, 8, 4, 1, 2, 3]', 2),\n",
       " ('[8, 6, 5, 7, 4, 1, 2, 3]', 2),\n",
       " ('[8, 6, 7, 5, 4, 1, 2, 3]', 2),\n",
       " ('[8, 5, 6, 7, 4, 1, 3, 2]', 2),\n",
       " ('[5, 6, 8, 7, 4, 1, 3, 2]', 2),\n",
       " ('[6, 5, 7, 8, 4, 1, 3, 2]', 2),\n",
       " ('[6, 8, 5, 7, 4, 1, 2, 3]', 2),\n",
       " ('[8, 6, 5, 7, 4, 1, 3, 2]', 2),\n",
       " ('[7, 6, 5, 8, 4, 1, 3, 2]', 2),\n",
       " ('[6, 7, 5, 8, 4, 1, 3, 2]', 2),\n",
       " ('[7, 8, 5, 6, 4, 1, 2, 3]', 2),\n",
       " ('[5, 7, 6, 8, 4, 1, 3, 2]', 2),\n",
       " ('[8, 5, 6, 7, 4, 1, 2, 3]', 1),\n",
       " ('[7, 5, 6, 8, 4, 2, 1, 3]', 1),\n",
       " ('[6, 7, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[8, 7, 5, 6, 4, 1, 2, 3]', 1),\n",
       " ('[5, 7, 8, 6, 4, 1, 2, 3]', 1),\n",
       " ('[8, 6, 7, 5, 4, 1, 3, 2]', 1),\n",
       " ('[5, 6, 8, 7, 4, 1, 2, 3]', 1),\n",
       " ('[5, 6, 7, 8, 4, 1, 2, 3]', 1),\n",
       " ('[6, 8, 7, 5, 4, 1, 2, 3]', 1),\n",
       " ('[6, 7, 8, 5, 4, 1, 2, 3]', 1),\n",
       " ('[7, 5, 6, 8, 4, 1, 2, 3]', 1),\n",
       " ('[7, 8, 6, 5, 4, 1, 3, 2]', 1),\n",
       " ('[5, 7, 8, 6, 4, 1, 3, 2]', 1),\n",
       " ('[7, 8, 5, 6, 4, 1, 3, 2]', 1),\n",
       " ('[7, 5, 6, 8, 4, 1, 3, 2]', 1),\n",
       " ('[7, 6, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[6, 5, 8, 7, 4, 1, 2, 3]', 1)]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_subtract)['init_complete_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 5, 7, 6, 4, 1, 2, 3]', 7),\n",
       " ('[6, 8, 5, 7, 4, 1, 2, 3]', 6),\n",
       " ('[6, 8, 7, 5, 4, 1, 3, 2]', 5),\n",
       " ('[6, 5, 7, 8, 4, 1, 3, 2]', 5),\n",
       " ('[5, 8, 7, 6, 4, 1, 2, 3]', 4),\n",
       " ('[6, 5, 7, 8, 4, 1, 2, 3]', 4),\n",
       " ('[5, 7, 8, 6, 4, 1, 2, 3]', 4),\n",
       " ('[8, 5, 6, 7, 4, 1, 3, 2]', 4),\n",
       " ('[6, 5, 8, 7, 4, 1, 3, 2]', 4),\n",
       " ('[5, 6, 8, 7, 4, 1, 3, 2]', 4),\n",
       " ('[8, 5, 7, 6, 4, 1, 3, 2]', 4),\n",
       " ('[6, 7, 5, 8, 4, 1, 2, 3]', 3),\n",
       " ('[7, 6, 8, 5, 4, 1, 2, 3]', 3),\n",
       " ('[8, 7, 6, 5, 4, 1, 3, 2]', 3),\n",
       " ('[5, 6, 7, 8, 4, 1, 3, 2]', 3),\n",
       " ('[5, 8, 7, 6, 4, 1, 3, 2]', 3),\n",
       " ('[5, 7, 6, 8, 4, 1, 3, 2]', 3),\n",
       " ('[8, 5, 6, 7, 4, 1, 2, 3]', 2),\n",
       " ('[7, 6, 5, 8, 4, 1, 2, 3]', 2),\n",
       " ('[8, 7, 5, 6, 4, 1, 3, 2]', 2),\n",
       " ('[7, 5, 8, 6, 4, 1, 3, 2]', 2),\n",
       " ('[6, 8, 7, 5, 4, 1, 2, 3]', 2),\n",
       " ('[6, 8, 5, 7, 4, 1, 3, 2]', 2),\n",
       " ('[5, 8, 6, 7, 4, 1, 3, 2]', 1),\n",
       " ('[6, 7, 5, 8, 4, 1, 3, 2]', 1),\n",
       " ('[7, 8, 6, 5, 4, 1, 2, 3]', 1),\n",
       " ('[7, 5, 6, 8, 4, 1, 2, 3]', 1),\n",
       " ('[8, 6, 7, 5, 4, 1, 2, 3]', 1),\n",
       " ('[7, 8, 6, 5, 4, 1, 3, 2]', 1),\n",
       " ('[6, 5, 8, 7, 4, 1, 2, 3]', 1),\n",
       " ('[6, 7, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[5, 7, 6, 8, 4, 1, 2, 3]', 1),\n",
       " ('[6, 7, 8, 5, 4, 1, 2, 3]', 1),\n",
       " ('[7, 8, 5, 6, 4, 1, 2, 3]', 1),\n",
       " ('[7, 6, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[8, 7, 6, 5, 4, 1, 2, 3]', 1),\n",
       " ('[5, 6, 8, 7, 4, 1, 2, 3]', 1),\n",
       " ('[8, 6, 5, 7, 4, 1, 2, 3]', 1),\n",
       " ('[5, 6, 7, 8, 4, 1, 2, 3]', 1),\n",
       " ('[5, 7, 8, 6, 4, 1, 3, 2]', 1),\n",
       " ('[5, 8, 6, 7, 4, 1, 2, 3]', 1),\n",
       " ('[8, 7, 5, 6, 4, 1, 2, 3]', 1)]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_subtract)['init_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Multiply"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[1, 8, 2, 7, 3, 6, 5, 4]', 24),\n",
       " ('[1, 8, 2, 3, 7, 6, 5, 4]', 22),\n",
       " ('[1, 2, 8, 3, 7, 6, 5, 4]', 21),\n",
       " ('[1, 2, 8, 7, 3, 6, 5, 4]', 14),\n",
       " ('[1, 8, 2, 3, 7, 6, 4, 5]', 6),\n",
       " ('[1, 8, 2, 7, 3, 6, 4, 5]', 5),\n",
       " ('[1, 2, 8, 3, 7, 6, 4, 5]', 2),\n",
       " ('[1, 2, 8, 3, 7, 5, 6, 4]', 2),\n",
       " ('[1, 2, 8, 7, 3, 6, 4, 5]', 2),\n",
       " ('[1, 8, 2, 3, 7, 5, 6, 4]', 1),\n",
       " ('[1, 8, 2, 7, 3, 5, 6, 4]', 1)]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_multiply)['init_complete_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[1, 8, 2, 7, 3, 6, 5, 4]', 29),\n",
       " ('[1, 8, 2, 3, 7, 6, 5, 4]', 27),\n",
       " ('[1, 2, 8, 3, 7, 6, 5, 4]', 13),\n",
       " ('[1, 2, 8, 7, 3, 6, 5, 4]', 13),\n",
       " ('[1, 8, 2, 7, 3, 6, 4, 5]', 7),\n",
       " ('[1, 8, 2, 3, 7, 6, 4, 5]', 5),\n",
       " ('[1, 8, 2, 7, 3, 5, 6, 4]', 2),\n",
       " ('[1, 2, 8, 3, 7, 6, 4, 5]', 1),\n",
       " ('[1, 2, 8, 7, 3, 6, 4, 5]', 1),\n",
       " ('[1, 2, 8, 3, 7, 5, 6, 4]', 1),\n",
       " ('[1, 8, 2, 3, 7, 5, 6, 4]', 1)]"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_multiply)['init_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Divide"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[5, 7, 6, 8, 4, 3, 2, 1]', 9),\n",
       " ('[6, 5, 8, 7, 4, 3, 2, 1]', 9),\n",
       " ('[7, 5, 8, 6, 4, 3, 2, 1]', 8),\n",
       " ('[8, 5, 6, 7, 4, 3, 2, 1]', 7),\n",
       " ('[8, 7, 6, 5, 4, 3, 2, 1]', 7),\n",
       " ('[7, 8, 6, 5, 4, 3, 2, 1]', 7),\n",
       " ('[5, 8, 7, 6, 4, 3, 2, 1]', 7),\n",
       " ('[6, 5, 7, 8, 4, 3, 2, 1]', 6),\n",
       " ('[6, 8, 7, 5, 4, 3, 2, 1]', 5),\n",
       " ('[8, 5, 7, 6, 4, 3, 2, 1]', 5),\n",
       " ('[5, 7, 8, 6, 4, 3, 2, 1]', 5),\n",
       " ('[8, 7, 5, 6, 4, 3, 2, 1]', 4),\n",
       " ('[7, 6, 8, 5, 4, 3, 2, 1]', 4),\n",
       " ('[8, 6, 7, 5, 4, 3, 2, 1]', 3),\n",
       " ('[7, 5, 6, 8, 4, 3, 2, 1]', 3),\n",
       " ('[7, 6, 5, 8, 4, 3, 2, 1]', 3),\n",
       " ('[5, 8, 6, 7, 4, 3, 2, 1]', 2),\n",
       " ('[6, 7, 5, 8, 4, 3, 2, 1]', 1),\n",
       " ('[8, 6, 5, 7, 4, 3, 2, 1]', 1),\n",
       " ('[5, 6, 7, 8, 4, 3, 2, 1]', 1),\n",
       " ('[7, 8, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[6, 8, 5, 7, 4, 3, 2, 1]', 1),\n",
       " ('[5, 6, 8, 7, 4, 3, 2, 1]', 1)]"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_divide)['init_complete_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 5, 6, 7, 4, 3, 2, 1]', 11),\n",
       " ('[5, 7, 6, 8, 4, 3, 2, 1]', 11),\n",
       " ('[5, 8, 7, 6, 4, 3, 2, 1]', 9),\n",
       " ('[6, 8, 7, 5, 4, 3, 2, 1]', 8),\n",
       " ('[5, 8, 6, 7, 4, 3, 2, 1]', 7),\n",
       " ('[8, 7, 6, 5, 4, 3, 2, 1]', 7),\n",
       " ('[5, 7, 8, 6, 4, 3, 2, 1]', 7),\n",
       " ('[6, 5, 7, 8, 4, 3, 2, 1]', 6),\n",
       " ('[5, 6, 8, 7, 4, 3, 2, 1]', 6),\n",
       " ('[6, 8, 5, 7, 4, 3, 2, 1]', 4),\n",
       " ('[8, 5, 7, 6, 4, 3, 2, 1]', 4),\n",
       " ('[6, 5, 8, 7, 4, 3, 2, 1]', 3),\n",
       " ('[7, 5, 8, 6, 4, 3, 2, 1]', 3),\n",
       " ('[6, 7, 8, 5, 4, 3, 2, 1]', 3),\n",
       " ('[8, 6, 7, 5, 4, 3, 2, 1]', 2),\n",
       " ('[8, 6, 5, 7, 4, 3, 2, 1]', 2),\n",
       " ('[7, 8, 6, 5, 4, 3, 2, 1]', 2),\n",
       " ('[7, 5, 6, 8, 4, 3, 2, 1]', 1),\n",
       " ('[8, 7, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[5, 6, 7, 8, 4, 3, 2, 1]', 1),\n",
       " ('[7, 8, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[7, 6, 8, 5, 4, 3, 2, 1]', 1)]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_divide)['init_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modulo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 7, 6, 5, 4, 3, 2, 1]', 6),\n",
       " ('[6, 5, 8, 7, 4, 3, 1, 2]', 5),\n",
       " ('[8, 5, 7, 6, 4, 3, 1, 2]', 4),\n",
       " ('[8, 6, 7, 5, 4, 3, 1, 2]', 4),\n",
       " ('[7, 8, 6, 5, 4, 3, 1, 2]', 4),\n",
       " ('[5, 7, 8, 6, 4, 3, 1, 2]', 3),\n",
       " ('[7, 8, 5, 6, 4, 3, 1, 2]', 3),\n",
       " ('[7, 5, 8, 6, 4, 3, 2, 1]', 3),\n",
       " ('[5, 8, 7, 6, 4, 3, 2, 1]', 3),\n",
       " ('[8, 6, 5, 7, 4, 3, 1, 2]', 3),\n",
       " ('[5, 6, 7, 8, 4, 3, 1, 2]', 3),\n",
       " ('[7, 6, 5, 8, 4, 3, 1, 2]', 3),\n",
       " ('[7, 5, 8, 6, 4, 3, 1, 2]', 3),\n",
       " ('[8, 5, 7, 6, 4, 3, 2, 1]', 3),\n",
       " ('[6, 5, 7, 8, 4, 2, 3, 1]', 2),\n",
       " ('[6, 7, 5, 8, 4, 3, 1, 2]', 2),\n",
       " ('[8, 6, 5, 7, 4, 3, 2, 1]', 2),\n",
       " ('[7, 8, 6, 5, 4, 3, 2, 1]', 2),\n",
       " ('[8, 7, 5, 6, 4, 3, 1, 2]', 2),\n",
       " ('[5, 6, 8, 7, 4, 3, 1, 2]', 2),\n",
       " ('[5, 8, 7, 6, 4, 3, 1, 2]', 2),\n",
       " ('[5, 7, 6, 8, 4, 3, 1, 2]', 2),\n",
       " ('[7, 6, 5, 8, 4, 3, 2, 1]', 2),\n",
       " ('[6, 8, 5, 7, 4, 3, 1, 2]', 2),\n",
       " ('[6, 5, 8, 7, 4, 3, 2, 1]', 2),\n",
       " ('[5, 7, 6, 8, 4, 1, 3, 2]', 2),\n",
       " ('[8, 7, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[7, 8, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[7, 5, 6, 8, 4, 3, 2, 1]', 1),\n",
       " ('[6, 7, 5, 8, 4, 2, 3, 1]', 1),\n",
       " ('[8, 5, 6, 7, 4, 1, 3, 2]', 1),\n",
       " ('[6, 5, 8, 7, 4, 1, 3, 2]', 1),\n",
       " ('[6, 8, 5, 7, 4, 2, 3, 1]', 1),\n",
       " ('[6, 5, 7, 8, 4, 3, 2, 1]', 1),\n",
       " ('[5, 8, 7, 6, 4, 1, 2, 3]', 1),\n",
       " ('[5, 7, 6, 8, 4, 3, 2, 1]', 1),\n",
       " ('[6, 8, 7, 5, 4, 3, 2, 1]', 1),\n",
       " ('[5, 8, 6, 7, 4, 3, 1, 2]', 1),\n",
       " ('[8, 7, 6, 5, 4, 3, 1, 2]', 1),\n",
       " ('[8, 5, 6, 7, 4, 3, 1, 2]', 1),\n",
       " ('[8, 7, 5, 6, 4, 1, 3, 2]', 1),\n",
       " ('[6, 7, 8, 5, 4, 3, 2, 1]', 1),\n",
       " ('[6, 8, 7, 5, 4, 3, 1, 2]', 1),\n",
       " ('[7, 5, 6, 8, 4, 1, 3, 2]', 1),\n",
       " ('[8, 5, 6, 7, 4, 3, 2, 1]', 1),\n",
       " ('[7, 5, 6, 8, 4, 3, 1, 2]', 1),\n",
       " ('[7, 6, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[8, 7, 6, 5, 4, 1, 2, 3]', 1),\n",
       " ('[8, 6, 5, 7, 4, 1, 3, 2]', 1),\n",
       " ('[5, 6, 8, 7, 4, 3, 2, 1]', 1),\n",
       " ('[5, 6, 7, 8, 4, 2, 1, 3]', 1),\n",
       " ('[5, 7, 8, 6, 4, 3, 2, 1]', 1)]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_modulo)['init_complete_all_correct_digit_epoch'].most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('[8, 5, 7, 6, 4, 3, 1, 2]', 7),\n",
       " ('[8, 5, 7, 6, 4, 3, 2, 1]', 6),\n",
       " ('[8, 7, 6, 5, 4, 3, 2, 1]', 5),\n",
       " ('[8, 5, 6, 7, 4, 3, 1, 2]', 5),\n",
       " ('[8, 7, 6, 5, 4, 3, 1, 2]', 4),\n",
       " ('[8, 7, 5, 6, 4, 3, 1, 2]', 4),\n",
       " ('[5, 7, 6, 8, 4, 1, 3, 2]', 4),\n",
       " ('[6, 5, 7, 8, 4, 3, 2, 1]', 3),\n",
       " ('[8, 6, 5, 7, 4, 3, 1, 2]', 3),\n",
       " ('[6, 8, 5, 7, 4, 3, 1, 2]', 3),\n",
       " ('[5, 7, 8, 6, 4, 3, 1, 2]', 3),\n",
       " ('[8, 7, 5, 6, 4, 3, 2, 1]', 2),\n",
       " ('[6, 7, 8, 5, 4, 3, 1, 2]', 2),\n",
       " ('[5, 6, 8, 7, 4, 3, 1, 2]', 2),\n",
       " ('[6, 5, 8, 7, 4, 3, 1, 2]', 2),\n",
       " ('[5, 8, 7, 6, 4, 3, 1, 2]', 2),\n",
       " ('[5, 6, 7, 8, 4, 3, 1, 2]', 2),\n",
       " ('[5, 7, 6, 8, 4, 3, 1, 2]', 2),\n",
       " ('[6, 5, 7, 8, 4, 3, 1, 2]', 2),\n",
       " ('[5, 6, 8, 7, 4, 3, 2, 1]', 2),\n",
       " ('[6, 5, 8, 7, 4, 3, 2, 1]', 2),\n",
       " ('[6, 8, 5, 7, 4, 3, 2, 1]', 2),\n",
       " ('[8, 5, 6, 7, 4, 3, 2, 1]', 2),\n",
       " ('[6, 8, 7, 5, 4, 3, 2, 1]', 2),\n",
       " ('[6, 5, 7, 8, 4, 2, 3, 1]', 1),\n",
       " ('[6, 7, 5, 8, 4, 3, 1, 2]', 1),\n",
       " ('[7, 8, 5, 6, 4, 3, 1, 2]', 1),\n",
       " ('[8, 6, 5, 7, 4, 2, 3, 1]', 1),\n",
       " ('[7, 8, 5, 6, 4, 3, 2, 1]', 1),\n",
       " ('[6, 7, 5, 8, 4, 2, 3, 1]', 1),\n",
       " ('[8, 5, 6, 7, 4, 1, 3, 2]', 1),\n",
       " ('[6, 7, 8, 5, 4, 1, 3, 2]', 1),\n",
       " ('[7, 6, 8, 5, 4, 3, 1, 2]', 1),\n",
       " ('[5, 8, 7, 6, 4, 3, 2, 1]', 1),\n",
       " ('[5, 8, 6, 7, 4, 3, 1, 2]', 1),\n",
       " ('[6, 5, 8, 7, 4, 2, 1, 3]', 1),\n",
       " ('[7, 6, 5, 8, 4, 3, 1, 2]', 1),\n",
       " ('[8, 7, 6, 5, 4, 2, 1, 3]', 1),\n",
       " ('[6, 8, 7, 5, 4, 2, 3, 1]', 1),\n",
       " ('[8, 6, 7, 5, 4, 3, 1, 2]', 1),\n",
       " ('[8, 7, 5, 6, 4, 1, 3, 2]', 1),\n",
       " ('[6, 7, 8, 5, 4, 3, 2, 1]', 1),\n",
       " ('[6, 8, 7, 5, 4, 3, 1, 2]', 1),\n",
       " ('[5, 8, 7, 6, 4, 1, 3, 2]', 1),\n",
       " ('[7, 5, 6, 8, 4, 3, 1, 2]', 1),\n",
       " ('[6, 8, 5, 7, 4, 1, 3, 2]', 1),\n",
       " ('[5, 6, 8, 7, 4, 1, 3, 2]', 1),\n",
       " ('[8, 6, 5, 7, 4, 1, 3, 2]', 1),\n",
       " ('[5, 7, 6, 8, 4, 2, 3, 1]', 1),\n",
       " ('[5, 7, 8, 6, 4, 3, 2, 1]', 1),\n",
       " ('[8, 5, 7, 6, 4, 1, 2, 3]', 1)]"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epoch_order_count(all_run_info_modulo)['init_all_correct_digit_epoch'].most_common()"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}