proofing_ground_demo_final.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import sys\n",
    "import json\n",
    "sys.path.append('code/')\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import os, binascii, hashlib, base58\n",
    "\n",
    "from parse_config import ConfigDict\n",
    "from balances import Balances\n",
    "from simulation import Simulation\n",
    "\n",
    "pd.set_option('display.max_rows', 20)\n",
    "\n",
    "config = ConfigDict().parse()\n",
    "\n",
    "def print_params(params):\n",
    "    for k,v in params().items():\n",
    "        print(\"{}: {}\".format(k,v))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## No initial issuance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_users: 10000\n",
      "n_miners_max: 100\n",
      "init_issue: False\n",
      "reward_dist: False\n",
      "reward: 100.0\n",
      "greed_factor: 0.0\n",
      "\n",
      "Gini Coefficient: 0.00\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>address</th>\n",
       "      <th>greedy</th>\n",
       "      <th>miner</th>\n",
       "      <th>values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5KMb3SC16x9Tamu6i1hQP1Baj4MVMBPB6rhUQktm4SQAqW...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5JYLnjWHrYorq7hweEnDhCcyiVr3v7ofAsSE4Tq7jQDoc8...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5Kk2FUZsNFzoLrQKnjHM37m7Sx5VkRpQpbP877DQruFqqx...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5KXnfujPsVMk5VARQTtyEg7eS9NRbmStv8udEbFm4sp9S7...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5KU2pJKgr7FZdQbmp5ofQJXnWwe1S4CcN5d96psp4F7Jru...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             address  greedy  miner  values\n",
       "0  5KMb3SC16x9Tamu6i1hQP1Baj4MVMBPB6rhUQktm4SQAqW...   False  False     0.0\n",
       "1  5JYLnjWHrYorq7hweEnDhCcyiVr3v7ofAsSE4Tq7jQDoc8...   False  False     0.0\n",
       "2  5Kk2FUZsNFzoLrQKnjHM37m7Sx5VkRpQpbP877DQruFqqx...   False  False     0.0\n",
       "3  5KXnfujPsVMk5VARQTtyEg7eS9NRbmStv8udEbFm4sp9S7...   False  False     0.0\n",
       "4  5KU2pJKgr7FZdQbmp5ofQJXnWwe1S4CcN5d96psp4F7Jru...   False  False     0.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = Balances(**config['default'])\n",
    "print_params(b.params)\n",
    "print(\"\\nGini Coefficient: {:0.2f}\".format(b.gini()))\n",
    "b.data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fair initial issuance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_users: 10000\n",
      "n_miners_max: 100\n",
      "init_issue: True\n",
      "reward_dist: False\n",
      "reward: 100.0\n",
      "greed_factor: 0.05\n",
      "init_issue_total: 1000000.0\n",
      "init_issue_max: 500.0\n",
      "init_issue_users: 9900\n",
      "\n",
      "Gini Coefficient: 0.50\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>address</th>\n",
       "      <th>greedy</th>\n",
       "      <th>miner</th>\n",
       "      <th>values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5JnwCb1X2VNP3Suu5dst2fFbUh67VJDPpNWXWobqZRXtqT...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>2.307097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5JBGGnW7LVtPPUybgaeJ1FAGRabupLhZtracb2zACaRp3g...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>93.585998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5JasuWxoESVkAf1SCQU6eWfwLySpwfwXKMiKKHMhkBfjHb...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>90.274064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5Kcstjsy6ySiRhgncopzh423YnkzMEmcgfe1QmRzNRp5RN...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>21.045219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5JQ1zW6EPXFp2mekKABKoqnxBko4cTUhuEGAuVVnZZtboL...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>156.698716</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             address  greedy  miner  \\\n",
       "0  5JnwCb1X2VNP3Suu5dst2fFbUh67VJDPpNWXWobqZRXtqT...   False  False   \n",
       "1  5JBGGnW7LVtPPUybgaeJ1FAGRabupLhZtracb2zACaRp3g...   False  False   \n",
       "2  5JasuWxoESVkAf1SCQU6eWfwLySpwfwXKMiKKHMhkBfjHb...   False  False   \n",
       "3  5Kcstjsy6ySiRhgncopzh423YnkzMEmcgfe1QmRzNRp5RN...   False  False   \n",
       "4  5JQ1zW6EPXFp2mekKABKoqnxBko4cTUhuEGAuVVnZZtboL...   False  False   \n",
       "\n",
       "       values  \n",
       "0    2.307097  \n",
       "1   93.585998  \n",
       "2   90.274064  \n",
       "3   21.045219  \n",
       "4  156.698716  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = Balances(**config['init_issue_fair_greedy'])\n",
    "print_params(b.params)\n",
    "print(\"\\nGini Coefficient: {:0.2f}\".format(b.gini()))\n",
    "b.data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Real initial issuance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_users: 10000\n",
      "n_miners_max: 100\n",
      "init_issue: True\n",
      "reward_dist: False\n",
      "reward: 100.0\n",
      "greed_factor: 0.05\n",
      "init_issue_total: 1000000.0\n",
      "init_issue_max: 50000.0\n",
      "init_issue_users: 1000\n",
      "\n",
      "Gini Coefficient: 0.95\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>address</th>\n",
       "      <th>greedy</th>\n",
       "      <th>miner</th>\n",
       "      <th>values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5K8k76F2wgLuZCPkL2eZh7MGyw3x7owAoJX16nihqcEfey...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5JsXpK2QG7UhAHV6QyTiHh4e8kbHVj4L6fL6hxapHRAzxX...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5KYTGky4RBG3uzTFMKozMTDQ6RtSjUYD8nw1iY7uQH1uYT...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5KAoiWjzEgSbMaGMUerqiqUGNxujNqnNUk8XviBmjH9rKW...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5JfJuU2WM74iuZCYmmDpaJkDJiiFP13HysNF1kbaFH15sf...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             address  greedy  miner  values\n",
       "0  5K8k76F2wgLuZCPkL2eZh7MGyw3x7owAoJX16nihqcEfey...   False  False     0.0\n",
       "1  5JsXpK2QG7UhAHV6QyTiHh4e8kbHVj4L6fL6hxapHRAzxX...   False  False     0.0\n",
       "2  5KYTGky4RBG3uzTFMKozMTDQ6RtSjUYD8nw1iY7uQH1uYT...   False  False     0.0\n",
       "3  5KAoiWjzEgSbMaGMUerqiqUGNxujNqnNUk8XviBmjH9rKW...   False  False     0.0\n",
       "4  5JfJuU2WM74iuZCYmmDpaJkDJiiFP13HysNF1kbaFH15sf...   False  False     0.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = Balances(**config['init_issue_real_greedy'])\n",
    "print_params(b.params)\n",
    "print(\"\\nGini Coefficient: {:0.2f}\".format(b.gini()))\n",
    "b.data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simulation experiments on greed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No initial issuance\t\t0.00000\t\t0.54520\t\t0.14151\n",
      "Distributed initial issuance\t\t0.49028\t\t0.56347\t\t0.14730\n",
      "Concentrated initial issuance\t\t0.94809\t\t0.56658\t\t0.15260\n"
     ]
    }
   ],
   "source": [
    "config = ConfigDict().parse()\n",
    "init_states = { 'default': \"No initial issuance\",\n",
    "                'init_issue_dist': \"Distributed initial issuance\",\n",
    "                'init_issue_conc': \"Concentrated initial issuance\" }\n",
    "greed_factor = { 'greedy': 0.05,\n",
    "                 'no_greed': 0, }\n",
    "\n",
    "g_all = {}\n",
    "g_init = []\n",
    "data = {}\n",
    "exps = []\n",
    "\n",
    "for init_key in init_states.keys():\n",
    "    s = Simulation(balances = Balances(**config[init_key]), **config[init_key])\n",
    "    g_init.append(s.balances.gini(initial=True) \n",
    "    \n",
    "    for greed_key in greed_factor.keys():\n",
    "        exp_key = '{}_{}'.format(init_key, greed_key)\n",
    "        exps.append(exp_key)\n",
    "        s.balances.greed_factor = greed_factor[greed_key]\n",
    "        s.balances.reinit()\n",
    "\n",
    "        # print_params(s.balances.params)\n",
    "        # print_params(s.params)\n",
    "        \n",
    "        (tx_data, g) = s.simulate()\n",
    "        \n",
    "        data[exp_key] = tx_data\n",
    "        g_all[exp_key] = g.copy()\n",
    "    \n",
    "    print(\"{}\\t\\t{:.5f}\\t\\t{:.5f}\\t\\t{:.5f}\".format(init_states[init_key], \n",
    "                                                    s.balances.gini(initial=True), \n",
    "                                                    np.mean(g_all['{}_greedy'.format(init_key)]), \n",
    "                                                    np.mean(g_all['{}_no_greed'.format(init_key)])))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEYCAYAAAD8hukFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1105f4a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# histograms of simulation results\n",
    "plt.close()\n",
    "fig, axs = plt.subplots(3,2)\n",
    "g_init = [ 0.0, 0.49028, 0.94809]\n",
    "\n",
    "for ax_row, g_i, init_key in zip(axs, g_init, init_states.keys()):\n",
    "    ax_row[0].set_ylabel(r'$N_{experiments}$')\n",
    "    for ax, greed_key in zip(ax_row, greed_factor.keys()):\n",
    "        exp = '{}_{}'.format(init_key, greed_key)\n",
    "        ax.hist(g_all[exp], bins=np.arange(0.,1.,0.05))\n",
    "        ax.set_title(exp)\n",
    "        ax.axvline(x=g_i, color='r')\n",
    "        \n",
    "# xlabels\n",
    "for ax in axs[2]:\n",
    "    ax.set_xlabel('Gini Coeff')\n",
    "    \n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}