selfplay_calibration_rank_elo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import math\n",
    "import pickle\n",
    "import sys\n",
    "import threading\n",
    "import time\n",
    "import traceback\n",
    "from collections import defaultdict\n",
    "from concurrent.futures.thread import ThreadPoolExecutor\n",
    "import numpy as np\n",
    "from elote import EloCompetitor\n",
    "from katrain.core.constants import (\n",
    "    AI_LOCAL,\n",
    "    AI_RANK,\n",
    "    AI_TENUKI,\n",
    "    AI_WEIGHTED,\n",
    "    AI_PICK,\n",
    "    AI_TERRITORY,\n",
    "    AI_POLICY,AI_INFLUENCE\n",
    ")\n",
    "\n",
    "from settings import Logger\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "DB_FILENAME = \"tournament_ai_performance.pickle\"\n",
    "class AI:\n",
    "    pass\n",
    "\n",
    "class FixedEloCompetitor(EloCompetitor):  # rating doesn't update on wins/losses\n",
    "    pass\n",
    "    \n",
    "with open(DB_FILENAME, \"rb\") as f:\n",
    "    ai_database,_ = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "ranked = [ai for ai in ai_database if ai.strategy==AI_RANK]\n",
    "calibrate = [ai for ai in ai_database if ai.strategy==AI_WEIGHTED]\n",
    "policy = [ai for ai in ai_database if ai.strategy==AI_POLICY and ai.ai_settings['opening_moves']==0][0]\n",
    "policy_elo = 1700\n",
    "kyu_elo = [(r.ai_settings['kyu_rank'],r.elo_comp.rating) for r in ranked]\n",
    "#kyu_elo.append( (-5,policy.elo_comp.rating) ) # policy ~5d\n",
    "kyu_elo = sorted(kyu_elo)\n",
    "kyu,elo = zip(*kyu_elo)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f5fd02e8f60>,\n",
       " <matplotlib.lines.Line2D at 0x7f5fd02f50b8>]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "kyu2 = np.array(kyu)[:,np.newaxis]\n",
    "reg = LinearRegression().fit(kyu2,elo)\n",
    "plt.plot(kyu,elo,kyu,reg.predict(kyu2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(-4, 1392.5226294653357),\n",
       " (-3, 1328.2407152976634),\n",
       " (-2, 1263.9588011299913),\n",
       " (-1, 1199.6768869623193),\n",
       " (0, 1135.3949727946472),\n",
       " (1, 1071.113058626975),\n",
       " (2, 1006.8311444593029),\n",
       " (3, 942.5492302916308),\n",
       " (4, 878.2673161239586),\n",
       " (5, 813.9854019562865),\n",
       " (6, 749.7034877886144),\n",
       " (7, 685.4215736209424),\n",
       " (8, 621.1396594532702),\n",
       " (9, 556.8577452855981),\n",
       " (10, 492.5758311179259),\n",
       " (11, 428.2939169502538),\n",
       " (12, 364.0120027825817),\n",
       " (13, 299.7300886149095),\n",
       " (14, 235.44817444723742),\n",
       " (15, 171.16626027956522),\n",
       " (16, 106.88434611189314),\n",
       " (17, 42.60243194422105),\n",
       " (18, -21.679482223451032)]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(zip(kyu,reg.predict(kyu2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "wt_elo =  [([r.ai_settings['weaken_fac']],r.elo_comp.rating) for r in calibrate]\n",
    "x,y = zip(*wt_elo)\n",
    "reg = LinearRegression().fit(x, y)\n",
    "py = reg.predict(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.0, 539),\n",
       " (0.05, 625),\n",
       " (0.1, 859),\n",
       " (0.2, 1035),\n",
       " (0.3, 1201),\n",
       " (0.4, 1299),\n",
       " (0.5, 1346),\n",
       " (0.75, 1374),\n",
       " (1.0, 1386),\n",
       " (1.5, 1376),\n",
       " (2.0, 1370),\n",
       " (3.0, 1383)]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DB_FILENAME = \"calibrated_ai_performance.scoreloss.pickle\"\n",
    "with open(DB_FILENAME, \"rb\") as f:\n",
    "    ai_database_sl,_ = pickle.load(f)\n",
    "str_elo =  [([r.ai_settings['strength']],r.elo_comp.rating) for r in ai_database_sl]\n",
    "[(a[0],round(b)) for a,b in str_elo]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "DB_FILENAME = \"calibrated_ai_performance.jank.pickle\"\n",
    "with open(DB_FILENAME, \"rb\") as f:\n",
    "    ai_database_jank,_ = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AI_LOCAL_ELO_GRID = [[0.0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.75, 1.0],[0, 5, 10, 15, 25, 50],[[-204.0, 791.0, 1154.0, 1372.0, 1480.0, 1473.0, 1700.0, 1700.0], [174.0, 1094.0, 1191.0, 1427.0, 1435.0, 1522.0, 1700.0, 1700.0], [619.0, 1155.0, 1323.0, 1380.0, 1450.0, 1558.0, 1700.0, 1700.0], [975.0, 1289.0, 1425.0, 1363.0, 1436.0, 1616.0, 1700.0, 1700.0], [1348.0, 1344.0, 1358.0, 1467.0, 1477.0, 1552.0, 1700.0, 1700.0], [1425.0, 1474.0, 1489.0, 1524.0, 1571.0, 1700.0, 1700.0, 1700.0]]]\n",
      "AI_TENUKI_ELO_GRID = [[0.0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.75, 1.0],[0, 5, 10, 15, 25, 50],[[241.0, 335.0, 530.0, 678.0, 830.0, 1075.0, 1414.0, 1700.0], [99.0, 469.0, 546.0, 707.0, 902.0, 1051.0, 1413.0, 1700.0], [327.0, 513.0, 645.0, 745.0, 900.0, 1134.0, 1376.0, 1700.0], [429.0, 519.0, 581.0, 754.0, 855.0, 1091.0, 1435.0, 1700.0], [492.0, 607.0, 682.0, 797.0, 873.0, 1208.0, 1454.0, 1700.0], [778.0, 830.0, 909.0, 949.0, 1169.0, 1461.0, 1483.0, 1700.0]]]\n",
      "AI_TERRITORY_ELO_GRID = [[0.0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.75, 1.0],[0, 5, 10, 15, 25, 50],[[34.0, 383.0, 566.0, 748.0, 992.0, 1264.0, 1547.0, 1700.0], [131.0, 517.0, 627.0, 826.0, 994.0, 1262.0, 1527.0, 1700.0], [291.0, 536.0, 586.0, 878.0, 937.0, 1342.0, 1547.0, 1700.0], [454.0, 526.0, 696.0, 850.0, 1038.0, 1275.0, 1529.0, 1700.0], [491.0, 603.0, 747.0, 792.0, 1050.0, 1390.0, 1635.0, 1700.0], [718.0, 841.0, 1039.0, 1076.0, 1332.0, 1523.0, 1700.0, 1700.0]]]\n",
      "AI_INFLUENCE_ELO_GRID = [[0.0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.75, 1.0],[0, 5, 10, 15, 25, 50],[[217.0, 439.0, 592.0, 768.0, 1028.0, 1227.0, 1449.0, 1591.0], [302.0, 571.0, 580.0, 840.0, 961.0, 1164.0, 1499.0, 1529.0], [388.0, 592.0, 619.0, 776.0, 1097.0, 1305.0, 1490.0, 1561.0], [467.0, 581.0, 764.0, 878.0, 1018.0, 1426.0, 1531.0, 1673.0], [539.0, 622.0, 815.0, 953.0, 1120.0, 1420.0, 1560.0, 1521.0], [772.0, 912.0, 958.0, 1145.0, 1318.0, 1511.0, 1577.0, 1519.0]]]\n",
      "AI_PICK_ELO_GRID = [[0.0, 0.05, 0.1, 0.2, 0.3, 0.5, 0.75, 1.0],[0, 5, 10, 15, 25, 50],[[-533.0, -515.0, -355.0, 234.0, 650.0, 1147.0, 1546.0, 1700.0], [-531.0, -517.0, -69.0, 347.0, 647.0, 1182.0, 1550.0, 1700.0], [-535.0, -311.0, 140.0, 459.0, 693.0, 1252.0, 1548.0, 1700.0], [-365.0, -82.0, 265.0, 508.0, 864.0, 1301.0, 1619.0, 1700.0], [-113.0, 273.0, 363.0, 641.0, 983.0, 1486.0, 1700.0, 1700.0], [514.0, 670.0, 870.0, 1128.0, 1305.0, 1550.0, 1700.0, 1700.0]]]\n"
     ]
    }
   ],
   "source": [
    "for stratname in ['AI_LOCAL','AI_TENUKI','AI_TERRITORY','AI_INFLUENCE','AI_PICK']:\n",
    "    strat = eval(stratname)\n",
    "    pick_ais = [ai for ai in ai_database_jank if ai.strategy==strat]\n",
    "    settings_elo =  [([r.ai_settings['pick_frac'],r.ai_settings['pick_n']],min(policy_elo,r.elo_comp.rating)) for r in pick_ais]\n",
    "    settings,elo = zip(*settings_elo)\n",
    "    frac, n = zip(*settings)\n",
    "    \n",
    "    frac_x = np.unique(frac)\n",
    "    n_y = np.unique(n)\n",
    "    matrix = np.zeros( (len(n_y),len(frac_x)) )    \n",
    "    \n",
    "    for (f,n),elo in settings_elo:\n",
    "        xi, = np.where(frac_x==f)\n",
    "        yi, = np.where(n_y==n)\n",
    "        matrix[yi[0]][xi[0]] = round(elo)\n",
    "        \n",
    "    print(f\"{stratname}_ELO_GRID = [{repr(list(frac_x))},{repr(list(n_y))},{repr(list(list(x) for x in matrix))}]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_monotonicity(a):\n",
    "    for row in a:\n",
    "        print(row,all(a >= b for a,b in zip(row[1:],row[:-1])))\n",
    "    a = np.transpose(a)\n",
    "    for row in a:\n",
    "        print(row,all(a >= b for a,b in zip(row[1:],row[:-1])))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-204.0, 791.0, 1154.0, 1372.0, 1402.0, 1473.0, 1700.0, 1700.0] True\n",
      "[174.0, 1094.0, 1191.0, 1384.0, 1435.0, 1522.0, 1700.0, 1700.0] True\n",
      "[619.0, 1155.0, 1323.0, 1390.0, 1450.0, 1558.0, 1700.0, 1700.0] True\n",
      "[975.0, 1289.0, 1332.0, 1401.0, 1461.0, 1575.0, 1700.0, 1700.0] True\n",
      "[1344.0, 1348.0, 1358.0, 1467.0, 1477.0, 1616.0, 1700.0, 1700.0] True\n",
      "[1425.0, 1474.0, 1489.0, 1524.0, 1571.0, 1700.0, 1700.0, 1700.0] True\n",
      "[-204.  174.  619.  975. 1344. 1425.] True\n",
      "[ 791. 1094. 1155. 1289. 1348. 1474.] True\n",
      "[1154. 1191. 1323. 1332. 1358. 1489.] True\n",
      "[1372. 1384. 1390. 1401. 1467. 1524.] True\n",
      "[1402. 1435. 1450. 1461. 1477. 1571.] True\n",
      "[1473. 1522. 1558. 1575. 1616. 1700.] True\n",
      "[1700. 1700. 1700. 1700. 1700. 1700.] True\n",
      "[1700. 1700. 1700. 1700. 1700. 1700.] True\n"
     ]
    }
   ],
   "source": [
    "check_monotonicity(      [\n",
    "        [-204.0, 791.0, 1154.0, 1372.0, 1402.0, 1473.0, 1700.0, 1700.0],\n",
    "        [174.0, 1094.0, 1191.0, 1384.0, 1435.0, 1522.0, 1700.0, 1700.0],\n",
    "        [619.0, 1155.0, 1323.0, 1390.0, 1450.0, 1558.0, 1700.0, 1700.0],\n",
    "        [975.0, 1289.0, 1332.0, 1401.0, 1461.0, 1575.0, 1700.0, 1700.0],\n",
    "        [1344.0, 1348.0, 1358.0, 1467.0, 1477.0, 1616.0, 1700.0, 1700.0],\n",
    "        [1425.0, 1474.0, 1489.0, 1524.0, 1571.0, 1700.0, 1700.0, 1700.0],\n",
    "    ] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}