evaluate_engines.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluate Hex Chess Engines"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pygame 2.5.2 (SDL 2.28.3, Python 3.11.5)\n",
      "Hello from the pygame community. https://www.pygame.org/contribute.html\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import pickle\n",
    "from tqdm import tqdm\n",
    "from hexchess import HexChessBoard, Player, RandomPlayer, GreedyPlayer\n",
    "from engines import QNetworkPlayer, A2CPlayer\n",
    "from utils import make_safe_filename\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "gpus = tf.config.list_physical_devices('GPU') \n",
    "#tf.config.experimental.set_memory_growth(gpus[0], True)\n",
    "#print(gpus[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Constants\n",
    "results_dir = \"./assets/results\"\n",
    "os.makedirs(results_dir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Game Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def play_game(white_player, black_player, board, max_moves = 50):\n",
    "    # Reset board\n",
    "    board.reset()\n",
    "\n",
    "    # Play game\n",
    "    n_moves = 0\n",
    "    turn_is_white = True\n",
    "    finished = False\n",
    "    for _ in range(max_moves * 2):\n",
    "        # Select active player\n",
    "        active_player = white_player if turn_is_white else black_player\n",
    "\n",
    "        # Get move\n",
    "        position_from, position_to = active_player.get_move()\n",
    "        success, finished = board.move(position_from, position_to, turn_is_white)\n",
    "\n",
    "        # Update turn\n",
    "        turn_is_white = not turn_is_white\n",
    "        n_moves += 1 if turn_is_white else 0\n",
    "        if finished:\n",
    "            break\n",
    "\n",
    "    # Determine winner (only valid if game is finished)\n",
    "    white_won = not turn_is_white\n",
    "    return finished, white_won, n_moves"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def play_games(\n",
    "    white_player_class,\n",
    "    black_player_class,\n",
    "    max_moves=50,\n",
    "    n_games=10,\n",
    "    progress_bar=True,\n",
    "):\n",
    "    # Parse player\n",
    "    if isinstance(white_player_class, dict):\n",
    "        white_player_kwargs = white_player_class['kwargs']\n",
    "        white_player_class = white_player_class['class']\n",
    "    else:\n",
    "        white_player_kwargs = {}\n",
    "    if isinstance(black_player_class, dict):\n",
    "        black_player_kwargs = black_player_class['kwargs']\n",
    "        black_player_class = black_player_class['class']\n",
    "    else:\n",
    "        black_player_kwargs = {}\n",
    "    \n",
    "    # Initialize board\n",
    "    board = HexChessBoard(save_prob=0.)\n",
    "    \n",
    "    # Initialize players\n",
    "    white_player = white_player_class(board, True, **white_player_kwargs)\n",
    "    black_player = black_player_class(board, False, **black_player_kwargs)\n",
    "    \n",
    "    \n",
    "    # Play games\n",
    "    fn = tqdm if progress_bar else lambda x: x\n",
    "    results = [None] * n_games\n",
    "    for game_index in fn(range(n_games)):\n",
    "        results[game_index] = play_game(\n",
    "            white_player=white_player,\n",
    "            black_player=black_player,\n",
    "            board=board,\n",
    "            max_moves=max_moves,\n",
    "        )\n",
    "\n",
    "    return results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate Games"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Configuration\n",
    "n_games = 100\n",
    "max_moves = 100\n",
    "white_players = [\n",
    "    {\"class\": RandomPlayer, \"label\": \"Random (baseline)\", \"kwargs\": {}}, \n",
    "    {\"class\": GreedyPlayer, \"label\": \"Greedy (baseline)\", \"kwargs\": {}},\n",
    "    {\"class\": QNetworkPlayer, \"label\": \"Deep Q Learning\", \"kwargs\": {}},\n",
    "    {\"class\": A2CPlayer, \"label\": \"Simple Actor Critic\", \"kwargs\": {\"model_name\": \"simple_a2c_random\"}},\n",
    "    {\"class\": A2CPlayer, \"label\": \"Advanced Actor Critic\", \"kwargs\": {\"model_name\": \"advanced_a2c_random\", \"residual_blocks\": True, \"shared_feature_extraction\": False, \"double_critic\": True, \"delayed_critic\": True, \"critic_delay_freq\": 5}}\n",
    "]\n",
    "n_players = len(white_players)\n",
    "black_players = white_players"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random (baseline) vs Random (baseline)\n",
      "Random (baseline) vs Greedy (baseline)\n",
      "Random (baseline) vs Deep Q Learning\n",
      "Random (baseline) vs Simple Actor Critic\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:30<00:00,  3.31it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random (baseline) vs Advanced Actor Critic\n",
      "Greedy (baseline) vs Random (baseline)\n",
      "Greedy (baseline) vs Greedy (baseline)\n",
      "Greedy (baseline) vs Deep Q Learning\n",
      "Greedy (baseline) vs Simple Actor Critic\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:05<00:00, 19.81it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Greedy (baseline) vs Advanced Actor Critic\n",
      "Deep Q Learning vs Random (baseline)\n",
      "Deep Q Learning vs Greedy (baseline)\n",
      "Deep Q Learning vs Deep Q Learning\n",
      "Deep Q Learning vs Simple Actor Critic\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.RestoredOptimizer` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.RestoredOptimizer`.\n",
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [02:19<00:00,  1.39s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Deep Q Learning vs Advanced Actor Critic\n",
      "Simple Actor Critic vs Random (baseline)\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:31<00:00,  3.21it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simple Actor Critic vs Greedy (baseline)\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:07<00:00, 13.27it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simple Actor Critic vs Deep Q Learning\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.RestoredOptimizer` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.RestoredOptimizer`.\n",
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n",
      "100%|██████████| 100/100 [02:42<00:00,  1.63s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simple Actor Critic vs Simple Actor Critic\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:26<00:00,  1.16it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Simple Actor Critic vs Advanced Actor Critic\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/advanced_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [24:21<00:00, 14.62s/it]  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Advanced Actor Critic vs Random (baseline)\n",
      "Advanced Actor Critic vs Greedy (baseline)\n",
      "Advanced Actor Critic vs Deep Q Learning\n",
      "Advanced Actor Critic vs Simple Actor Critic\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/advanced_a2c_random.pth\n",
      "Model loaded from:  /Users/tawab/Desktop/columbia/Courses/Spring2024/Reinforcement Learning/project/github/can_chess_with_hexagons_rl/assets/a2c/simple_a2c_random.pth\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [01:58<00:00,  1.18s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Advanced Actor Critic vs Advanced Actor Critic\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Gather results\n",
    "rows = [player[\"label\"] if isinstance(player, dict) else player.name for player in white_players]\n",
    "cols = [player[\"label\"] if isinstance(player, dict) else player.name for player in black_players]\n",
    "games_won = np.zeros((len(white_players), len(black_players)), dtype=int)\n",
    "games_draw = np.zeros((len(white_players), len(black_players)), dtype=int)\n",
    "games_lost = np.zeros((len(white_players), len(black_players)), dtype=int)\n",
    "games_moves = np.zeros((len(white_players), len(black_players)), dtype=float)\n",
    "\n",
    "# All white players\n",
    "for i, white_player in enumerate(white_players):\n",
    "    # Against all black players\n",
    "    for j, black_player in enumerate(black_players):\n",
    "        # Battle start\n",
    "        print(f\"{white_player['label']} vs {black_player['label']}\")\n",
    "        \n",
    "        # Make sure we don't already have the results\n",
    "        results_filename = f\"{make_safe_filename(white_player['label'])}_vs_{make_safe_filename(black_player['label'])}_{n_games}.pkl\"\n",
    "        results_filepath = os.path.join(results_dir, results_filename)\n",
    "        if os.path.exists(results_filepath):\n",
    "            # Results path exists - load from file\n",
    "            with open(results_filepath, \"rb\") as results_file:\n",
    "                results = pickle.load(results_file)\n",
    "        else:\n",
    "            # Results path does not exist - play games\n",
    "            results = play_games(\n",
    "                white_player,\n",
    "                black_player,\n",
    "                max_moves=max_moves,\n",
    "                n_games=n_games,\n",
    "                progress_bar=True,\n",
    "            )\n",
    "            # Save to file\n",
    "            with open(results_filepath, \"wb\") as results_file:\n",
    "                pickle.dump(results, results_file)\n",
    "        \n",
    "        # Save statistics\n",
    "        for game_index, result in enumerate(results):\n",
    "            # Parse result\n",
    "            game_finished, white_won, n_moves = result\n",
    "            \n",
    "            # Win / Loss / Draw\n",
    "            if game_finished:\n",
    "                if white_won:\n",
    "                    games_won[i, j] += 1\n",
    "                else:\n",
    "                    games_lost[i, j] += 1\n",
    "            else:\n",
    "                games_draw[i, j] += 1\n",
    "\n",
    "            # Number of moves\n",
    "            games_moves[i, j] += n_moves / n_games\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Collect results in Dataframes\n",
    "games_won_df = pd.DataFrame(data = games_won, index = rows, columns = cols)\n",
    "games_lost_df = pd.DataFrame(data = games_lost, index = rows, columns = cols)\n",
    "games_draw_df = pd.DataFrame(data = games_draw, index = rows, columns = cols)\n",
    "games_moves_df = pd.DataFrame(data = games_moves, index = rows, columns = cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Save results\n",
    "games_df_names = [\"won\", \"lost\", \"draw\", \"moves\"]\n",
    "games_dfs = [games_won_df, games_lost_df, games_draw_df, games_moves_df]\n",
    "for name, game_df in zip(games_df_names, games_dfs):\n",
    "    game_df.to_csv(os.path.join(results_dir, f\"games_{name}.csv\"), index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Load results\n",
    "models = \"qnetworks\"\n",
    "games_df_names = [\"won\", \"lost\", \"draw\", \"moves\"]\n",
    "games_dfs = [pd.read_csv(os.path.join(results_dir, f\"games_{name}.csv\"), index_col=0) for name in games_df_names]\n",
    "games_won_df, games_lost_df, games_draw_df, games_moves_df = games_dfs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "___GAMES WON___\n"
     ]
    },
    {
     "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>Random (baseline)</th>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Random (baseline)</th>\n",
       "      <td>38</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>35</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <td>100</td>\n",
       "      <td>41</td>\n",
       "      <td>96</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Random (baseline)  Greedy (baseline)  Deep Q Learning  \\\n",
       "Random (baseline)                     38                  1               17   \n",
       "Greedy (baseline)                    100                 41               96   \n",
       "Deep Q Learning                       33                  1                0   \n",
       "Simple Actor Critic                   24                  0                0   \n",
       "Advanced Actor Critic                 37                  0                0   \n",
       "\n",
       "                       Simple Actor Critic  Advanced Actor Critic  \n",
       "Random (baseline)                       35                     21  \n",
       "Greedy (baseline)                      100                    100  \n",
       "Deep Q Learning                          0                      0  \n",
       "Simple Actor Critic                      0                      0  \n",
       "Advanced Actor Critic                    0                      0  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"___GAMES WON___\")\n",
    "games_won_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "___GAMES LOST___\n"
     ]
    },
    {
     "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>Random (baseline)</th>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Random (baseline)</th>\n",
       "      <td>37</td>\n",
       "      <td>99</td>\n",
       "      <td>65</td>\n",
       "      <td>1</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <td>0</td>\n",
       "      <td>59</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <td>32</td>\n",
       "      <td>99</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <td>32</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "      <td>34</td>\n",
       "      <td>100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Random (baseline)  Greedy (baseline)  Deep Q Learning  \\\n",
       "Random (baseline)                     37                 99               65   \n",
       "Greedy (baseline)                      0                 59                4   \n",
       "Deep Q Learning                       32                 99              100   \n",
       "Simple Actor Critic                   32                100                0   \n",
       "Advanced Actor Critic                 34                100                0   \n",
       "\n",
       "                       Simple Actor Critic  Advanced Actor Critic  \n",
       "Random (baseline)                        1                     29  \n",
       "Greedy (baseline)                        0                      0  \n",
       "Deep Q Learning                          0                      0  \n",
       "Simple Actor Critic                      0                      0  \n",
       "Advanced Actor Critic                    0                      0  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"___GAMES LOST___\")\n",
    "games_lost_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "___GAMES DRAW___\n"
     ]
    },
    {
     "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>Random (baseline)</th>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Random (baseline)</th>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>64</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <td>44</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Random (baseline)  Greedy (baseline)  Deep Q Learning  \\\n",
       "Random (baseline)                     25                  0               18   \n",
       "Greedy (baseline)                      0                  0                0   \n",
       "Deep Q Learning                       35                  0                0   \n",
       "Simple Actor Critic                   44                  0              100   \n",
       "Advanced Actor Critic                 29                  0              100   \n",
       "\n",
       "                       Simple Actor Critic  Advanced Actor Critic  \n",
       "Random (baseline)                       64                     50  \n",
       "Greedy (baseline)                        0                      0  \n",
       "Deep Q Learning                        100                    100  \n",
       "Simple Actor Critic                    100                    100  \n",
       "Advanced Actor Critic                  100                    100  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"___GAMES DRAW___\")\n",
    "games_draw_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "___NET WINS___\n"
     ]
    },
    {
     "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>Random (baseline)</th>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Random (baseline)</th>\n",
       "      <td>1</td>\n",
       "      <td>-98</td>\n",
       "      <td>-48</td>\n",
       "      <td>34</td>\n",
       "      <td>-8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <td>100</td>\n",
       "      <td>-18</td>\n",
       "      <td>92</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <td>1</td>\n",
       "      <td>-98</td>\n",
       "      <td>-100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <td>-8</td>\n",
       "      <td>-100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "      <td>3</td>\n",
       "      <td>-100</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Random (baseline)  Greedy (baseline)  Deep Q Learning  \\\n",
       "Random (baseline)                      1                -98              -48   \n",
       "Greedy (baseline)                    100                -18               92   \n",
       "Deep Q Learning                        1                -98             -100   \n",
       "Simple Actor Critic                   -8               -100                0   \n",
       "Advanced Actor Critic                  3               -100                0   \n",
       "\n",
       "                       Simple Actor Critic  Advanced Actor Critic  \n",
       "Random (baseline)                       34                     -8  \n",
       "Greedy (baseline)                      100                    100  \n",
       "Deep Q Learning                          0                      0  \n",
       "Simple Actor Critic                      0                      0  \n",
       "Advanced Actor Critic                    0                      0  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"___NET WINS___\")\n",
    "net_wins_df = games_won_df - games_lost_df\n",
    "net_wins_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__NR OF MOVES__\n"
     ]
    },
    {
     "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>Random (baseline)</th>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Random (baseline)</th>\n",
       "      <td>65.55</td>\n",
       "      <td>12.07</td>\n",
       "      <td>41.28</td>\n",
       "      <td>87.60</td>\n",
       "      <td>80.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greedy (baseline)</th>\n",
       "      <td>11.49</td>\n",
       "      <td>19.68</td>\n",
       "      <td>13.20</td>\n",
       "      <td>16.33</td>\n",
       "      <td>11.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Deep Q Learning</th>\n",
       "      <td>70.76</td>\n",
       "      <td>10.14</td>\n",
       "      <td>5.00</td>\n",
       "      <td>100.00</td>\n",
       "      <td>100.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Simple Actor Critic</th>\n",
       "      <td>69.39</td>\n",
       "      <td>17.99</td>\n",
       "      <td>100.00</td>\n",
       "      <td>100.00</td>\n",
       "      <td>100.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Advanced Actor Critic</th>\n",
       "      <td>62.03</td>\n",
       "      <td>6.00</td>\n",
       "      <td>100.00</td>\n",
       "      <td>100.00</td>\n",
       "      <td>100.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Random (baseline)  Greedy (baseline)  Deep Q Learning  \\\n",
       "Random (baseline)                  65.55              12.07            41.28   \n",
       "Greedy (baseline)                  11.49              19.68            13.20   \n",
       "Deep Q Learning                    70.76              10.14             5.00   \n",
       "Simple Actor Critic                69.39              17.99           100.00   \n",
       "Advanced Actor Critic              62.03               6.00           100.00   \n",
       "\n",
       "                       Simple Actor Critic  Advanced Actor Critic  \n",
       "Random (baseline)                    87.60                  80.53  \n",
       "Greedy (baseline)                    16.33                  11.24  \n",
       "Deep Q Learning                     100.00                 100.00  \n",
       "Simple Actor Critic                 100.00                 100.00  \n",
       "Advanced Actor Critic               100.00                 100.00  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"__NR OF MOVES__\")\n",
    "games_moves_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "net_wins = net_wins_df.to_numpy()\n",
    "white_model_names = net_wins_df.columns.to_numpy()\n",
    "black_model_names = net_wins_df.index.to_numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8, 8))\n",
    "img = ax.imshow(net_wins, cmap=\"seismic\", interpolation=\"none\", origin=\"lower\", vmin=-100, vmax=100)\n",
    "fig.colorbar(img, ax=ax)\n",
    "ax.set_xticks(np.arange(len(white_model_names)))\n",
    "ax.set_xticklabels(white_model_names, rotation=45, horizontalalignment=\"right\")\n",
    "ax.set_yticks(np.arange(len(black_model_names)))\n",
    "ax.set_yticklabels(black_model_names, rotation=0, horizontalalignment=\"right\")\n",
    "ax.set_ylabel(\"White player (win = +)\")\n",
    "ax.set_xlabel(\"Black player (win = -)\")\n",
    "fig.tight_layout()\n",
    "fig.savefig(os.path.join(\".\", \"net_wins.png\"), dpi=300, facecolor=\"white\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8, 8))\n",
    "img = ax.imshow(games_moves_df.to_numpy(), cmap=\"Reds\", interpolation=\"none\", origin=\"lower\", vmin=0, vmax=100)\n",
    "fig.colorbar(img, ax=ax)\n",
    "ax.set_xticks(np.arange(len(white_model_names)))\n",
    "ax.set_xticklabels(white_model_names, rotation=45, horizontalalignment=\"right\")\n",
    "ax.set_yticks(np.arange(len(black_model_names)))\n",
    "ax.set_yticklabels(black_model_names, rotation=0, horizontalalignment=\"right\")\n",
    "ax.set_ylabel(\"White player\")\n",
    "ax.set_xlabel(\"Black player\")\n",
    "fig.tight_layout()\n",
    "fig.savefig(os.path.join(\".\", \"nr_moves.png\"), dpi=300, facecolor=\"white\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (Local)",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}