wordle.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting emoji\n",
      "  Downloading emoji-1.6.3.tar.gz (174 kB)\n",
      "\u001b[K     |████████████████████████████████| 174 kB 12.7 MB/s eta 0:00:01\n",
      "\u001b[?25hBuilding wheels for collected packages: emoji\n",
      "  Building wheel for emoji (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25h  Created wheel for emoji: filename=emoji-1.6.3-py3-none-any.whl size=170281 sha256=c4e72636fdb6cfed9dce4c5249910685788509f1d24b82674af0bcee034eebe3\n",
      "  Stored in directory: /home/jevans96/.cache/pip/wheels/7d/09/45/1ad33b3ac1e05d07355a3c81fe56c89837774fdb6665574765\n",
      "Successfully built emoji\n",
      "Installing collected packages: emoji\n",
      "Successfully installed emoji-1.6.3\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install emoji"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import emoji\n",
    "import numpy as np\n",
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('./words.json', 'r') as f:\n",
    "    words = json.load(f)\n",
    "    \n",
    "dictionary = words['solutions']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def word_possible(guessed_word, results, word_to_check):\n",
    "    for i, (letter, result) in enumerate(zip(guessed_word, results)):\n",
    "        if result == 0 and letter in word_to_check:\n",
    "            return 0\n",
    "        if result == 1 and (word_to_check[i] == letter or letter not in word_to_check):\n",
    "            return 0\n",
    "        if result == 2 and not word_to_check[i] == letter:\n",
    "            return 0\n",
    "    return 1\n",
    "\n",
    "def get_results_set(guessed_word, dictionary):\n",
    "    results_list = []\n",
    "    for word in dictionary:\n",
    "        if word == guessed_word:\n",
    "            continue\n",
    "        results = [0, 0, 0, 0, 0]\n",
    "        for i, letter in enumerate(guessed_word):\n",
    "            if letter == word[i]:\n",
    "                results[i] = 2\n",
    "            elif letter in word:\n",
    "                results[i] = 1\n",
    "        results_list.append(tuple(results))\n",
    "    return Counter(results_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "results_counters = {word: get_results_set(word, dictionary) for word in dictionary}\n",
    "results_sets = {word: tuple(counter) for word, counter in results_counters.items()}\n",
    "results_weights = {\n",
    "    word: [val/len(dictionary) for _, val in counter.items()]\n",
    "    for word, counter in results_counters.items()\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores = {\n",
    "    word: [\n",
    "        sum([word_possible(word, results, dict_word) for dict_word in dictionary])\n",
    "        for results in results_set\n",
    "    ] for word, results_set in results_sets.items()\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " word  weighted mean possible words after guess\n",
      "raise: 61.03\n",
      "arise: 63.75\n",
      "irate: 63.81\n",
      "arose: 66.05\n",
      "alter: 70.02\n",
      "saner: 70.16\n",
      "later: 70.25\n",
      "snare: 71.13\n",
      "stare: 71.32\n",
      "slate: 71.60\n",
      ".\n",
      ".\n",
      ".\n",
      "woozy: 688.59\n",
      "jazzy: 702.12\n",
      "fluff: 723.86\n",
      "fizzy: 728.34\n",
      "jiffy: 735.07\n",
      "civic: 737.23\n",
      "puppy: 778.19\n",
      "mamma: 791.62\n",
      "vivid: 816.97\n",
      "mummy: 821.09\n"
     ]
    }
   ],
   "source": [
    "means = {\n",
    "    word: np.average(score, weights=results_weights[word])\n",
    "    for word, score in scores.items()\n",
    "}\n",
    "print(' word  weighted mean possible words after guess')\n",
    "for mean, word in sorted((value,key) for (key,value) in means.items())[:10]:\n",
    "    print(f'{word}: {mean:.2f}')\n",
    "print('.')\n",
    "print('.')\n",
    "print('.')\n",
    "for mean, word in sorted((value,key) for (key,value) in means.items())[-11:-1]:\n",
    "    print(f'{word}: {mean:.2f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def emoji_results(results):\n",
    "    emoji_map = {\n",
    "        0: ':black_large_square:',\n",
    "        1: ':yellow_square:',\n",
    "        2: ':green_square:',\n",
    "    }\n",
    "    return emoji.emojize(''.join([emoji_map[i] for i in results]))\n",
    "\n",
    "raise_dictionaries = {\n",
    "    results: [word for word in dictionary if word_possible('raise', results, word)]\n",
    "    for results in results_sets['raise']\n",
    "}\n",
    "\n",
    "strs = []\n",
    "lens = []\n",
    "for results, raise_dict in raise_dictionaries.items():\n",
    "    if len(raise_dict) == 1:\n",
    "        best_choice = (0, raise_dict)\n",
    "    else:\n",
    "        raise_results_counters = {\n",
    "            word: get_results_set(word, raise_dict) for word in raise_dict\n",
    "        }\n",
    "        raise_results_sets = {\n",
    "            word: tuple(counter) for word, counter in raise_results_counters.items()\n",
    "        }\n",
    "        raise_results_weights = {\n",
    "            word: [val/len(raise_dict) for _, val in counter.items()]\n",
    "            for word, counter in raise_results_counters.items()\n",
    "        }\n",
    "        raise_scores = {\n",
    "            word: [\n",
    "                sum([word_possible(word, results, dict_word) for dict_word in raise_dict])\n",
    "                for results in results_set\n",
    "            ] for word, results_set in raise_results_sets.items()\n",
    "        }\n",
    "        raise_means = {\n",
    "            word: np.average(score, weights=raise_results_weights[word])\n",
    "            for word, score in raise_scores.items()\n",
    "        }\n",
    "        best = sorted((value,key) for (key,value) in raise_means.items())\n",
    "        best_words = [word for mean, word in best if mean == best[0][0]]\n",
    "        best_choice = (best[0][0], best_words)\n",
    "    \n",
    "    strs.append(\n",
    "        f'{emoji_results(results)}\\t{len(raise_dict)}\\t{best_choice[0]:.2f}\\t\\t{\" \".join(best_choice[1])}')\n",
    "    lens.append(len(raise_dict))\n",
    "\n",
    "idxs = np.argsort(lens)[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "first result\tleft\texpected\tbest words\n",
      "⬛⬛⬛⬛⬛\t168\t6.31\t\tcould\n",
      "⬛⬛⬛⬛🟨\t121\t5.15\t\tolden\n",
      "⬛⬛🟨⬛⬛\t107\t6.43\t\tpilot\n",
      "🟨⬛⬛⬛⬛\t103\t6.20\t\ttruly\n",
      "🟨⬛⬛⬛🟨\t102\t8.76\t\touter\n",
      "⬛🟨⬛⬛⬛\t92\t4.21\t\tcloak\n",
      "⬛🟩⬛⬛⬛\t91\t7.56\t\ttangy\n",
      "⬛⬛⬛🟨⬛\t80\t4.90\t\tstunk\n",
      "🟨🟨⬛⬛⬛\t78\t5.81\t\tacorn\n",
      "⬛🟨⬛⬛🟨\t69\t3.38\t\tcleat\n",
      "⬛⬛⬛⬛🟩\t61\t4.23\t\tlunge\n",
      "⬛⬛🟩⬛⬛\t51\t3.64\t\tcling\n",
      "⬛🟨⬛🟨⬛\t43\t3.00\t\tstalk\n",
      "⬛🟨⬛⬛🟩\t41\t4.85\t\tplate\n",
      "⬛⬛⬛🟨🟨\t41\t2.25\t\tspelt\n",
      "🟨⬛⬛⬛🟩\t40\t4.18\t\tprove\n",
      "⬛⬛🟨⬛🟨\t35\t2.65\t\twiden\n",
      "🟨🟨⬛⬛🟨\t34\t3.55\t\ttread\n",
      "⬛🟨🟨⬛⬛\t34\t2.39\t\ttidal\n",
      "⬛⬛🟩🟨⬛\t29\t3.36\t\tstink\n",
      "🟨🟩⬛⬛🟨\t28\t7.00\t\ttaper\n",
      "🟨⬛🟩⬛⬛\t28\t4.33\t\tprint\n",
      "🟨🟩⬛⬛⬛\t26\t3.96\t\tparty\n",
      "🟨🟨⬛⬛🟩\t26\t3.88\t\tgrace\n",
      "🟨⬛🟨⬛🟨\t26\t4.36\t\tdiner\n",
      "⬛🟩⬛⬛🟩\t26\t4.68\t\tcable\n",
      "⬛⬛🟨⬛🟩\t25\t3.75\t\tbinge\n",
      "🟨⬛⬛🟨⬛\t24\t2.30\t\tshort\n",
      "🟨⬛🟨⬛⬛\t23\t1.91\t\tdroit\n",
      "⬛⬛🟩⬛🟩\t23\t4.27\t\tutile\n",
      "⬛🟨⬛🟩⬛\t22\t3.29\t\tslash\n",
      "🟨🟨⬛🟨⬛\t21\t3.00\t\tstrap\n",
      "⬛⬛🟨🟨⬛\t21\t1.90\t\tstoic\n",
      "⬛🟨⬛🟨🟩\t20\t5.32\t\tslate\n",
      "⬛🟩⬛⬛🟨\t20\t3.74\t\tnavel\n",
      "🟩⬛⬛⬛🟨\t20\t3.00\t\truler\n",
      "⬛🟩⬛🟨⬛\t20\t2.37\t\tsalon\n",
      "⬛⬛⬛🟩🟩\t20\t3.32\t\tloose\n",
      "🟨⬛⬛🟨🟨\t18\t2.76\t\tshrew\n",
      "🟨🟨🟨⬛⬛\t18\t2.29\t\ttrail\n",
      "⬛⬛⬛🟨🟩\t17\t2.75\t\tstone\n",
      "🟨⬛🟩⬛🟩\t17\t4.75\t\ttripe\n",
      "⬛⬛⬛🟩⬛\t17\t1.75\t\tblush\n",
      "⬛⬛🟩⬛🟨\t15\t2.14\t\tedict\n",
      "⬛⬛🟩🟨🟩\t15\t4.57\t\tsnipe spine\n",
      "⬛🟩🟨⬛⬛\t14\t2.08\t\tpanic\n",
      "🟩🟨⬛⬛🟨\t13\t2.17\t\trelay\n",
      "🟨⬛⬛🟩⬛\t13\t1.33\t\tcrust\n",
      "⬛🟨⬛🟨🟨\t12\t1.73\t\tsteak\n",
      "🟨⬛🟩⬛🟨\t12\t2.27\t\tfried\n",
      "⬛🟨🟩⬛⬛\t12\t1.55\t\talign axion\n",
      "🟨⬛⬛🟨🟩\t10\t3.44\t\tscree spree\n",
      "🟩⬛⬛⬛⬛\t10\t1.44\t\truddy\n",
      "🟩🟩⬛⬛⬛\t9\t2.00\t\trandy rayon\n",
      "⬛⬛🟩🟩⬛\t9\t2.00\t\thoist\n",
      "⬛⬛⬛🟩🟨\t9\t2.00\t\tguess\n",
      "⬛🟩⬛🟨🟩\t9\t3.50\t\tcaste sauce\n",
      "🟩⬛🟨⬛🟨\t8\t2.14\t\triper river rivet\n",
      "⬛🟨⬛🟩🟩\t8\t1.86\t\tcease\n",
      "🟨⬛⬛🟩🟩\t8\t3.00\t\tcurse nurse prose purse\n",
      "🟨🟨⬛🟩⬛\t8\t2.14\t\tbrash brass crash crass grass\n",
      "⬛🟩⬛🟩⬛\t7\t1.67\t\tpalsy\n",
      "⬛🟨🟨🟨⬛\t7\t1.00\t\tslain snail stain\n",
      "🟨⬛⬛🟩🟨\t7\t1.67\t\tcress\n",
      "🟩⬛⬛⬛🟩\t6\t1.00\t\trogue\n",
      "🟩🟩🟨⬛⬛\t6\t1.40\t\trabid radii radio rapid\n",
      "⬛🟩🟩⬛⬛\t6\t1.80\t\tfaint faith paint taint\n",
      "⬛⬛🟨🟩⬛\t6\t1.40\t\ttipsy\n",
      "⬛🟨⬛🟩🟨\t5\t2.50\t\tbeast feast leash least yeast\n",
      "🟨⬛🟩🟨⬛\t5\t1.00\t\tshirk\n",
      "⬛🟩🟨🟨⬛\t5\t2.50\t\tbasic basil basin basis satin\n",
      "⬛⬛🟨🟨🟩\t5\t1.00\t\tsiege singe\n",
      "⬛🟩⬛🟩🟩\t5\t1.50\t\tlapse pause\n",
      "🟨🟨⬛🟨🟩\t5\t4.00\t\tscare share snare spare stare\n",
      "🟨⬛🟩🟩⬛\t5\t2.50\t\tbrisk crisp frisk prism\n",
      "⬛🟨🟩⬛🟩\t5\t1.50\t\talike alive\n",
      "🟨🟨🟩⬛⬛\t4\t1.67\t\tbriar friar triad trial\n",
      "🟩🟨⬛⬛⬛\t4\t1.00\t\troyal rural\n",
      "🟩🟩⬛⬛🟨\t4\t1.67\t\tracer ramen rarer raven\n",
      "🟨⬛🟨🟨🟨\t4\t1.00\t\tmiser wiser\n",
      "⬛⬛🟨🟨🟨\t4\t1.00\t\tislet sheik sinew stein\n",
      "🟨🟩🟩⬛⬛\t4\t1.67\t\tdairy fairy hairy\n",
      "🟨🟩⬛⬛🟩\t4\t1.67\t\tbarge carve farce large\n",
      "🟨⬛🟨🟨⬛\t4\t1.00\t\tsprig strip\n",
      "🟨🟨⬛🟨🟨\t4\t3.00\t\tshear smear spear swear\n",
      "🟩⬛🟨⬛⬛\t4\t1.00\t\trobin\n",
      "⬛🟨🟨⬛🟨\t3\t1.00\t\temail ideal media\n",
      "🟨⬛🟨⬛🟩\t3\t1.00\t\tdirge eerie fibre\n",
      "⬛🟩🟩⬛🟩\t3\t1.00\t\tnaive waive\n",
      "🟩⬛🟨🟨🟨\t3\t1.00\t\tresin risen riser\n",
      "⬛⬛🟩🟨🟨\t3\t1.00\t\tshied spied spiel\n",
      "⬛⬛🟩🟩🟩\t3\t1.00\t\tnoise poise\n",
      "🟨⬛🟩🟨🟩\t2\t1.00\t\tshire spire\n",
      "🟨🟩⬛🟩⬛\t2\t1.00\t\tharsh marsh\n",
      "🟨🟩⬛🟨⬛\t2\t1.00\t\tsatyr savor\n",
      "🟨🟨⬛🟩🟩\t2\t1.00\t\tarose erase\n",
      "🟨🟩⬛🟨🟨\t2\t1.00\t\tsafer saner\n",
      "🟩⬛🟨⬛🟩\t2\t1.00\t\tridge rifle\n",
      "⬛🟨🟨⬛🟩\t2\t1.00\t\timage inane\n",
      "⬛🟩🟩🟩⬛\t2\t1.00\t\tdaisy waist\n",
      "🟩⬛⬛🟩🟩\t2\t1.00\t\treuse rouse\n",
      "⬛⬛🟩🟩🟨\t2\t1.00\t\texist heist\n",
      "🟩⬛⬛🟨🟨\t2\t1.00\t\trebus reset\n",
      "🟨🟩🟨⬛⬛\t2\t1.00\t\tnadir tapir\n",
      "⬛🟨🟩🟩⬛\t1\t0.00\t\tamiss\n",
      "⬛🟨🟨🟨🟩\t1\t0.00\t\taisle\n",
      "🟩🟩⬛🟨⬛\t1\t0.00\t\traspy\n",
      "🟨⬛🟩🟨🟨\t1\t0.00\t\tskier\n",
      "🟨🟩⬛🟩🟩\t1\t0.00\t\tparse\n",
      "🟨🟨🟩⬛🟩\t1\t0.00\t\tafire\n",
      "⬛🟩⬛🟨🟨\t1\t0.00\t\teasel\n",
      "🟩⬛🟨🟩🟩\t1\t0.00\t\trinse\n",
      "🟩🟨⬛🟩⬛\t1\t0.00\t\troast\n",
      "🟩⬛🟩⬛🟨\t1\t0.00\t\treign\n",
      "⬛🟨🟨🟨🟨\t1\t0.00\t\tsepia\n",
      "🟩⬛🟩⬛⬛\t1\t0.00\t\trhino\n",
      "🟩🟩⬛⬛🟩\t1\t0.00\t\trange\n",
      "🟨🟨🟨⬛🟩\t1\t0.00\t\tirate\n",
      "🟨🟨🟨⬛🟨\t1\t0.00\t\taider\n",
      "⬛🟨🟩🟨🟩\t1\t0.00\t\taside\n",
      "🟩⬛🟨🟨⬛\t1\t0.00\t\trisky\n",
      "🟨🟨🟩🟩🟩\t1\t0.00\t\tarise\n",
      "⬛🟨🟨🟩⬛\t1\t0.00\t\tquasi\n",
      "🟩🟩🟩⬛⬛\t1\t0.00\t\trainy\n",
      "⬛🟩🟩🟨⬛\t1\t0.00\t\tsaint\n",
      "⬛🟨🟩⬛🟨\t1\t0.00\t\talien\n",
      "🟩⬛⬛🟨⬛\t1\t0.00\t\trusty\n",
      "🟨⬛🟨🟩⬛\t1\t0.00\t\tfirst\n",
      "🟨🟨🟨🟨⬛\t1\t0.00\t\tstair\n",
      "🟩⬛⬛🟩⬛\t1\t0.00\t\troost\n",
      "🟩🟨🟨⬛⬛\t1\t0.00\t\trival\n"
     ]
    }
   ],
   "source": [
    "print('first result\\tleft\\texpected\\tbest words')\n",
    "for i in idxs:\n",
    "    print(strs[i])"
   ]
  },
  {
   "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}