rank_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math, glob, time, os, re, json\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from collections import defaultdict\n",
    "from matplotlib import pyplot as plt\n",
    "from tqdm.notebook import trange, tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalized_histogram(data, bins):\n",
    "    weights = np.ones_like(data) / len(data)\n",
    "    return np.histogram(data, bins, weights = weights)\n",
    "\n",
    "plt.rcParams.update({'font.size': 22})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 2431 games and 665119 moves.\n"
     ]
    }
   ],
   "source": [
    "moves_data = pd.read_csv('moves-20b.csv.gz',index_col=0)\n",
    "games_data = pd.read_csv('games-20b.csv.gz',index_col=0)\n",
    "print(f\"Loaded {games_data.shape[0]} games and {moves_data.shape[0]} moves.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "user_moves = moves_data[moves_data.move_color==moves_data.user_color].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_features(df):\n",
    "    pol_rank_frac, _bins = normalized_histogram(df.policy_rank.values,bins=[-0.5,0.5,3.5,9.5,19.5,39.5,79.5])\n",
    "    pol_loss_frac, _bins = normalized_histogram(df.top_value - df.policy_value,bins=np.arange(-0.025,0.975,0.05)) #??\n",
    "    score_loss, _bins = normalized_histogram(df.points_lost.clip(-6,50),bins=[-6,-1.5,-0.5,0.5,1.5,3,6,12,25,50])\n",
    "    \n",
    "    f =  {f\"{k}.{i}\":v for k,l in {'policy_rank_hist':pol_rank_frac,\n",
    "                       #'policy_loss_hist':pol_loss_frac,\n",
    "                       'score_loss_hist':score_loss}.items() for i,v in enumerate(l) }\n",
    "    f['n_moves'] = max(df.move)\n",
    "    return f"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>game_id</th>\n",
       "      <th>move</th>\n",
       "      <th>num_legal</th>\n",
       "      <th>policy_rank</th>\n",
       "      <th>policy_value</th>\n",
       "      <th>top_value</th>\n",
       "      <th>black_score</th>\n",
       "      <th>points_lost</th>\n",
       "      <th>move_color</th>\n",
       "      <th>user_color</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24614579</td>\n",
       "      <td>1</td>\n",
       "      <td>362</td>\n",
       "      <td>37</td>\n",
       "      <td>0.000729</td>\n",
       "      <td>0.083457</td>\n",
       "      <td>-1.340746</td>\n",
       "      <td>0.628943</td>\n",
       "      <td>black</td>\n",
       "      <td>white</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24614579</td>\n",
       "      <td>2</td>\n",
       "      <td>361</td>\n",
       "      <td>6</td>\n",
       "      <td>0.018248</td>\n",
       "      <td>0.416445</td>\n",
       "      <td>-1.148735</td>\n",
       "      <td>0.192010</td>\n",
       "      <td>white</td>\n",
       "      <td>white</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>24614579</td>\n",
       "      <td>3</td>\n",
       "      <td>360</td>\n",
       "      <td>21</td>\n",
       "      <td>0.000658</td>\n",
       "      <td>0.574577</td>\n",
       "      <td>-1.575831</td>\n",
       "      <td>0.427096</td>\n",
       "      <td>black</td>\n",
       "      <td>white</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24614579</td>\n",
       "      <td>4</td>\n",
       "      <td>359</td>\n",
       "      <td>5</td>\n",
       "      <td>0.016626</td>\n",
       "      <td>0.451228</td>\n",
       "      <td>-1.588290</td>\n",
       "      <td>-0.012458</td>\n",
       "      <td>white</td>\n",
       "      <td>white</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24614579</td>\n",
       "      <td>5</td>\n",
       "      <td>358</td>\n",
       "      <td>15</td>\n",
       "      <td>0.000753</td>\n",
       "      <td>0.564865</td>\n",
       "      <td>-1.899404</td>\n",
       "      <td>0.311114</td>\n",
       "      <td>black</td>\n",
       "      <td>white</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665114</th>\n",
       "      <td>24679322</td>\n",
       "      <td>264</td>\n",
       "      <td>107</td>\n",
       "      <td>4</td>\n",
       "      <td>0.046985</td>\n",
       "      <td>0.078753</td>\n",
       "      <td>-50.762959</td>\n",
       "      <td>-1.000729</td>\n",
       "      <td>white</td>\n",
       "      <td>black</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665115</th>\n",
       "      <td>24679322</td>\n",
       "      <td>265</td>\n",
       "      <td>114</td>\n",
       "      <td>0</td>\n",
       "      <td>0.070761</td>\n",
       "      <td>0.070761</td>\n",
       "      <td>-49.834446</td>\n",
       "      <td>-0.928513</td>\n",
       "      <td>black</td>\n",
       "      <td>black</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665116</th>\n",
       "      <td>24679322</td>\n",
       "      <td>266</td>\n",
       "      <td>106</td>\n",
       "      <td>0</td>\n",
       "      <td>0.080871</td>\n",
       "      <td>0.080871</td>\n",
       "      <td>-50.532585</td>\n",
       "      <td>-0.698139</td>\n",
       "      <td>white</td>\n",
       "      <td>black</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665117</th>\n",
       "      <td>24679322</td>\n",
       "      <td>267</td>\n",
       "      <td>113</td>\n",
       "      <td>0</td>\n",
       "      <td>0.079108</td>\n",
       "      <td>0.079108</td>\n",
       "      <td>-49.847683</td>\n",
       "      <td>-0.684902</td>\n",
       "      <td>black</td>\n",
       "      <td>black</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665118</th>\n",
       "      <td>24679322</td>\n",
       "      <td>268</td>\n",
       "      <td>106</td>\n",
       "      <td>100</td>\n",
       "      <td>0.001247</td>\n",
       "      <td>0.085504</td>\n",
       "      <td>-50.477768</td>\n",
       "      <td>-0.630085</td>\n",
       "      <td>white</td>\n",
       "      <td>black</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>665119 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         game_id  move  num_legal  policy_rank  policy_value  top_value  \\\n",
       "0       24614579     1        362           37      0.000729   0.083457   \n",
       "1       24614579     2        361            6      0.018248   0.416445   \n",
       "2       24614579     3        360           21      0.000658   0.574577   \n",
       "3       24614579     4        359            5      0.016626   0.451228   \n",
       "4       24614579     5        358           15      0.000753   0.564865   \n",
       "...          ...   ...        ...          ...           ...        ...   \n",
       "665114  24679322   264        107            4      0.046985   0.078753   \n",
       "665115  24679322   265        114            0      0.070761   0.070761   \n",
       "665116  24679322   266        106            0      0.080871   0.080871   \n",
       "665117  24679322   267        113            0      0.079108   0.079108   \n",
       "665118  24679322   268        106          100      0.001247   0.085504   \n",
       "\n",
       "        black_score  points_lost move_color user_color  \n",
       "0         -1.340746     0.628943      black      white  \n",
       "1         -1.148735     0.192010      white      white  \n",
       "2         -1.575831     0.427096      black      white  \n",
       "3         -1.588290    -0.012458      white      white  \n",
       "4         -1.899404     0.311114      black      white  \n",
       "...             ...          ...        ...        ...  \n",
       "665114   -50.762959    -1.000729      white      black  \n",
       "665115   -49.834446    -0.928513      black      black  \n",
       "665116   -50.532585    -0.698139      white      black  \n",
       "665117   -49.847683    -0.684902      black      black  \n",
       "665118   -50.477768    -0.630085      white      black  \n",
       "\n",
       "[665119 rows x 10 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "moves_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = []\n",
    "y = []\n",
    "x_game_id = []\n",
    "for game_id,df in user_moves.groupby('game_id'):\n",
    "    x_game_id.append(game_id)\n",
    "    x.append(make_features(df))\n",
    "    y.append( games_data.loc[game_id,'user_rank'] )"
   ]
  },
  {
   "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>policy_rank_hist.0</th>\n",
       "      <th>policy_rank_hist.1</th>\n",
       "      <th>policy_rank_hist.2</th>\n",
       "      <th>policy_rank_hist.3</th>\n",
       "      <th>policy_rank_hist.4</th>\n",
       "      <th>policy_rank_hist.5</th>\n",
       "      <th>score_loss_hist.0</th>\n",
       "      <th>score_loss_hist.1</th>\n",
       "      <th>score_loss_hist.2</th>\n",
       "      <th>score_loss_hist.3</th>\n",
       "      <th>score_loss_hist.4</th>\n",
       "      <th>score_loss_hist.5</th>\n",
       "      <th>score_loss_hist.6</th>\n",
       "      <th>score_loss_hist.7</th>\n",
       "      <th>score_loss_hist.8</th>\n",
       "      <th>n_moves</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.294118</td>\n",
       "      <td>0.264706</td>\n",
       "      <td>0.127451</td>\n",
       "      <td>0.098039</td>\n",
       "      <td>0.098039</td>\n",
       "      <td>0.098039</td>\n",
       "      <td>0.049020</td>\n",
       "      <td>0.088235</td>\n",
       "      <td>0.274510</td>\n",
       "      <td>0.196078</td>\n",
       "      <td>0.039216</td>\n",
       "      <td>0.147059</td>\n",
       "      <td>0.196078</td>\n",
       "      <td>0.009804</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.367347</td>\n",
       "      <td>0.272109</td>\n",
       "      <td>0.183673</td>\n",
       "      <td>0.081633</td>\n",
       "      <td>0.040816</td>\n",
       "      <td>0.034014</td>\n",
       "      <td>0.088435</td>\n",
       "      <td>0.170068</td>\n",
       "      <td>0.312925</td>\n",
       "      <td>0.102041</td>\n",
       "      <td>0.102041</td>\n",
       "      <td>0.088435</td>\n",
       "      <td>0.095238</td>\n",
       "      <td>0.040816</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.355030</td>\n",
       "      <td>0.213018</td>\n",
       "      <td>0.142012</td>\n",
       "      <td>0.100592</td>\n",
       "      <td>0.100592</td>\n",
       "      <td>0.071006</td>\n",
       "      <td>0.153846</td>\n",
       "      <td>0.153846</td>\n",
       "      <td>0.272189</td>\n",
       "      <td>0.142012</td>\n",
       "      <td>0.106509</td>\n",
       "      <td>0.088757</td>\n",
       "      <td>0.041420</td>\n",
       "      <td>0.035503</td>\n",
       "      <td>0.005917</td>\n",
       "      <td>337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.232759</td>\n",
       "      <td>0.232759</td>\n",
       "      <td>0.155172</td>\n",
       "      <td>0.103448</td>\n",
       "      <td>0.086207</td>\n",
       "      <td>0.146552</td>\n",
       "      <td>0.137931</td>\n",
       "      <td>0.068966</td>\n",
       "      <td>0.215517</td>\n",
       "      <td>0.129310</td>\n",
       "      <td>0.086207</td>\n",
       "      <td>0.112069</td>\n",
       "      <td>0.120690</td>\n",
       "      <td>0.112069</td>\n",
       "      <td>0.017241</td>\n",
       "      <td>231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.107383</td>\n",
       "      <td>0.281879</td>\n",
       "      <td>0.194631</td>\n",
       "      <td>0.120805</td>\n",
       "      <td>0.100671</td>\n",
       "      <td>0.114094</td>\n",
       "      <td>0.040268</td>\n",
       "      <td>0.046980</td>\n",
       "      <td>0.288591</td>\n",
       "      <td>0.228188</td>\n",
       "      <td>0.140940</td>\n",
       "      <td>0.107383</td>\n",
       "      <td>0.087248</td>\n",
       "      <td>0.040268</td>\n",
       "      <td>0.020134</td>\n",
       "      <td>298</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2426</th>\n",
       "      <td>0.289720</td>\n",
       "      <td>0.233645</td>\n",
       "      <td>0.196262</td>\n",
       "      <td>0.112150</td>\n",
       "      <td>0.084112</td>\n",
       "      <td>0.065421</td>\n",
       "      <td>0.046729</td>\n",
       "      <td>0.056075</td>\n",
       "      <td>0.233645</td>\n",
       "      <td>0.065421</td>\n",
       "      <td>0.186916</td>\n",
       "      <td>0.149533</td>\n",
       "      <td>0.168224</td>\n",
       "      <td>0.056075</td>\n",
       "      <td>0.037383</td>\n",
       "      <td>214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2427</th>\n",
       "      <td>0.323529</td>\n",
       "      <td>0.227941</td>\n",
       "      <td>0.147059</td>\n",
       "      <td>0.095588</td>\n",
       "      <td>0.102941</td>\n",
       "      <td>0.066176</td>\n",
       "      <td>0.051471</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>0.257353</td>\n",
       "      <td>0.257353</td>\n",
       "      <td>0.147059</td>\n",
       "      <td>0.139706</td>\n",
       "      <td>0.022059</td>\n",
       "      <td>0.007353</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2428</th>\n",
       "      <td>0.270677</td>\n",
       "      <td>0.263158</td>\n",
       "      <td>0.067669</td>\n",
       "      <td>0.210526</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.067669</td>\n",
       "      <td>0.045113</td>\n",
       "      <td>0.075188</td>\n",
       "      <td>0.278195</td>\n",
       "      <td>0.255639</td>\n",
       "      <td>0.157895</td>\n",
       "      <td>0.127820</td>\n",
       "      <td>0.045113</td>\n",
       "      <td>0.015038</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2429</th>\n",
       "      <td>0.338346</td>\n",
       "      <td>0.218045</td>\n",
       "      <td>0.165414</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.030075</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.180451</td>\n",
       "      <td>0.187970</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.135338</td>\n",
       "      <td>0.120301</td>\n",
       "      <td>0.082707</td>\n",
       "      <td>0.037594</td>\n",
       "      <td>265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2430</th>\n",
       "      <td>0.404110</td>\n",
       "      <td>0.273973</td>\n",
       "      <td>0.130137</td>\n",
       "      <td>0.075342</td>\n",
       "      <td>0.068493</td>\n",
       "      <td>0.027397</td>\n",
       "      <td>0.157534</td>\n",
       "      <td>0.068493</td>\n",
       "      <td>0.157534</td>\n",
       "      <td>0.157534</td>\n",
       "      <td>0.164384</td>\n",
       "      <td>0.205479</td>\n",
       "      <td>0.089041</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2431 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      policy_rank_hist.0  policy_rank_hist.1  policy_rank_hist.2  \\\n",
       "0               0.294118            0.264706            0.127451   \n",
       "1               0.367347            0.272109            0.183673   \n",
       "2               0.355030            0.213018            0.142012   \n",
       "3               0.232759            0.232759            0.155172   \n",
       "4               0.107383            0.281879            0.194631   \n",
       "...                  ...                 ...                 ...   \n",
       "2426            0.289720            0.233645            0.196262   \n",
       "2427            0.323529            0.227941            0.147059   \n",
       "2428            0.270677            0.263158            0.067669   \n",
       "2429            0.338346            0.218045            0.165414   \n",
       "2430            0.404110            0.273973            0.130137   \n",
       "\n",
       "      policy_rank_hist.3  policy_rank_hist.4  policy_rank_hist.5  \\\n",
       "0               0.098039            0.098039            0.098039   \n",
       "1               0.081633            0.040816            0.034014   \n",
       "2               0.100592            0.100592            0.071006   \n",
       "3               0.103448            0.086207            0.146552   \n",
       "4               0.120805            0.100671            0.114094   \n",
       "...                  ...                 ...                 ...   \n",
       "2426            0.112150            0.084112            0.065421   \n",
       "2427            0.095588            0.102941            0.066176   \n",
       "2428            0.210526            0.082707            0.067669   \n",
       "2429            0.082707            0.082707            0.082707   \n",
       "2430            0.075342            0.068493            0.027397   \n",
       "\n",
       "      score_loss_hist.0  score_loss_hist.1  score_loss_hist.2  \\\n",
       "0              0.049020           0.088235           0.274510   \n",
       "1              0.088435           0.170068           0.312925   \n",
       "2              0.153846           0.153846           0.272189   \n",
       "3              0.137931           0.068966           0.215517   \n",
       "4              0.040268           0.046980           0.288591   \n",
       "...                 ...                ...                ...   \n",
       "2426           0.046729           0.056075           0.233645   \n",
       "2427           0.051471           0.117647           0.257353   \n",
       "2428           0.045113           0.075188           0.278195   \n",
       "2429           0.030075           0.082707           0.180451   \n",
       "2430           0.157534           0.068493           0.157534   \n",
       "\n",
       "      score_loss_hist.3  score_loss_hist.4  score_loss_hist.5  \\\n",
       "0              0.196078           0.039216           0.147059   \n",
       "1              0.102041           0.102041           0.088435   \n",
       "2              0.142012           0.106509           0.088757   \n",
       "3              0.129310           0.086207           0.112069   \n",
       "4              0.228188           0.140940           0.107383   \n",
       "...                 ...                ...                ...   \n",
       "2426           0.065421           0.186916           0.149533   \n",
       "2427           0.257353           0.147059           0.139706   \n",
       "2428           0.255639           0.157895           0.127820   \n",
       "2429           0.187970           0.142857           0.135338   \n",
       "2430           0.157534           0.164384           0.205479   \n",
       "\n",
       "      score_loss_hist.6  score_loss_hist.7  score_loss_hist.8  n_moves  \n",
       "0              0.196078           0.009804           0.000000      204  \n",
       "1              0.095238           0.040816           0.000000      294  \n",
       "2              0.041420           0.035503           0.005917      337  \n",
       "3              0.120690           0.112069           0.017241      231  \n",
       "4              0.087248           0.040268           0.020134      298  \n",
       "...                 ...                ...                ...      ...  \n",
       "2426           0.168224           0.056075           0.037383      214  \n",
       "2427           0.022059           0.007353           0.000000      271  \n",
       "2428           0.045113           0.015038           0.000000      266  \n",
       "2429           0.120301           0.082707           0.037594      265  \n",
       "2430           0.089041           0.000000           0.000000      291  \n",
       "\n",
       "[2431 rows x 16 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "ss = StandardScaler()\n",
    "\n",
    "df = pd.DataFrame(x)\n",
    "y_all=np.array(y)\n",
    "sel = y_all<25\n",
    "y_sel = y_all[sel]\n",
    "x=df.values\n",
    "xN_sel = ss.fit_transform(x[sel,:])\n",
    "xN_all = ss.transform(x)\n",
    "x_game_id=np.array(x_game_id)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import keras\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.callbacks import EarlyStopping\n",
    "from keras_tqdm import TQDMNotebookCallback\n",
    "model = Sequential()\n",
    "model.add(Dense(20, input_dim=16, activation='tanh', kernel_regularizer='l2'))\n",
    "model.add(Dense(10, activation='tanh', kernel_regularizer='l2'))\n",
    "model.add(Dense(1, activation=None))\n",
    "model.compile(loss='mse', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e951a1e193984d3a9f31960b09150b94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Training', style=ProgressStyle(description_width='initial…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 0', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 1', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 2', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 3', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 4', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 5', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 6', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 7', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 8', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 9', max=2101.0, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 10', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 11', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 12', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 13', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 14', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 15', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 16', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 17', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 18', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 19', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 20', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 21', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 22', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 23', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 24', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 25', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 26', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 27', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 28', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 29', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 30', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 31', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 32', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 33', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 34', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 35', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 36', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 37', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 38', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 39', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 40', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 41', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 42', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 43', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 44', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 45', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 46', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 47', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 48', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 49', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 50', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 51', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 52', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 53', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 54', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 55', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 56', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 57', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 58', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 59', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 60', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, description='Epoch 61', max=2101.0, style=ProgressStyle(description_wi…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00062: early stopping\n"
     ]
    }
   ],
   "source": [
    "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=10)\n",
    "history = model.fit(xN_sel, y_sel, epochs=100, batch_size=16,verbose=0,\n",
    "                    validation_split=0.1,callbacks=[TQDMNotebookCallback(),es])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f2afc205048>"
      ]
     },
     "execution_count": 32,
     "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": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss']); plt.legend(['loss','validation loss'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Predicted Rank')"
      ]
     },
     "execution_count": 33,
     "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": [
    "pY = model.predict(xN_all)[:,0]\n",
    "pY_sel = model.predict(xN_sel)[:,0]\n",
    "plt.scatter(y,pY)\n",
    "plt.scatter(y_sel,pY_sel)\n",
    "plt.xlabel('OGS Rank')\n",
    "plt.ylabel('Predicted Rank')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Predicted Rank')"
      ]
     },
     "execution_count": 34,
     "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": [
    "from scipy.stats import kde\n",
    "nbins=300\n",
    "k = kde.gaussian_kde(np.array([y,pY]))\n",
    "xi, yi = np.mgrid[-10:30:nbins*1j, -10:30:nbins*1j]\n",
    "zi = k(np.vstack([xi.flatten(), yi.flatten()]))\n",
    "plt.pcolormesh(xi, yi, zi.reshape(xi.shape), cmap=plt.cm.magma)\n",
    "plt.plot([-10, 30], [-10, 30], linewidth=0.5, c='m')\n",
    "plt.xlabel('OGS Rank')\n",
    "plt.ylabel('Predicted Rank')"
   ]
  },
  {
   "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
}