CrayHPO_rpv.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Distributed hyper-parameter optimization with the Cray HPO library\n",
    "\n",
    "In this notebook we demonstrate distributed HPO with multi-node training using the Cray HPO library for the RPV classification problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)\n"
     ]
    }
   ],
   "source": [
    "!squeue -u $USER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# System\n",
    "import sys\n",
    "import glob\n",
    "\n",
    "# Externals\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Pick up Cray-HPO installation\n",
    "sys.path.append('/usr/common/software/cray-hpo/0.4.0')\n",
    "from crayai import hpo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configure the search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Job config\n",
    "n_nodes = 32\n",
    "nodes_per_eval = 4\n",
    "alloc_args = '-J cray-hpo-rpv -q interactive -C haswell -t 4:00:00 -c 64'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training config\n",
    "n_epochs = 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Optimizer config\n",
    "pop_size = 8\n",
    "num_demes = 4\n",
    "generations = 4\n",
    "mutation_rate = 0.05\n",
    "crossover_rate = 0.33\n",
    "results_file = 'hpo.log'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hyperparameters\n",
    "params = hpo.Params([\n",
    "    ['--h1', 16, (4, 64)],\n",
    "    ['--h2', 32, (4, 64)],\n",
    "    ['--h3', 64, (8, 128)],\n",
    "    ['--h4', 128, (32, 256)],\n",
    "    ['--dropout', 0.2, (0., 1.)],\n",
    "    ['--optimizer', 'Adam', ['Adam', 'Nadam', 'Adadelta']],\n",
    "    ['--lr', 1e-3, [1e-1, 1e-2, 1e-3, 1e-4, 1e-5]],\n",
    "])"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "# Hyperparameters\n",
    "params = hpo.Params([\n",
    "    ['--h1', 16, [4, 8, 16, 32, 64]],\n",
    "    ['--h2', 32, [4, 8, 16, 32, 64]],\n",
    "    ['--h3', 64, [8, 16, 32, 64, 128]],\n",
    "    ['--h4', 128, [32, 64, 128, 256]],\n",
    "    ['--lr', 0.001, (1e-6, 0.01)],\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Perform the search"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "!python train_rpv.py --help"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the evaluator\n",
    "cmd = 'python train_rpv.py --n-epochs %i --fom best' % (n_epochs)\n",
    "evaluator = hpo.Evaluator(cmd,\n",
    "                          nodes=n_nodes,\n",
    "                          nodes_per_eval=nodes_per_eval,\n",
    "                          alloc_args=alloc_args,\n",
    "                          launcher='wlm',\n",
    "                          verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the optimizer\n",
    "optimizer = hpo.GeneticOptimizer(evaluator,\n",
    "                                 pop_size=pop_size,\n",
    "                                 num_demes=num_demes,\n",
    "                                 generations=generations,\n",
    "                                 mutation_rate=mutation_rate,\n",
    "                                 crossover_rate=crossover_rate,\n",
    "                                 verbose=True,\n",
    "                                 log_fn=results_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2h 17min 6s, sys: 2h 4min 48s, total: 4h 21min 54s\n",
      "Wall time: 3h 6min 36s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# Run the optimizer\n",
    "optimizer.optimize(params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Destruct the evaluator to release the allocation\n",
    "del evaluator\n",
    "# Delete the optimizer to make it dump out the full Deme results\n",
    "del optimizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyze the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>generation</th>\n",
       "      <th>epoch</th>\n",
       "      <th>best_fom</th>\n",
       "      <th>avg_fom</th>\n",
       "      <th>checkpoint_in</th>\n",
       "      <th>checkpoint_out</th>\n",
       "      <th>--h1</th>\n",
       "      <th>--h2</th>\n",
       "      <th>--h3</th>\n",
       "      <th>--h4</th>\n",
       "      <th>--dropout</th>\n",
       "      <th>--optimizer</th>\n",
       "      <th>--lr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.050737</td>\n",
       "      <td>0.557403</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18</td>\n",
       "      <td>28</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.047372</td>\n",
       "      <td>0.442378</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16</td>\n",
       "      <td>32</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0.046571</td>\n",
       "      <td>0.614770</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>30</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.046571</td>\n",
       "      <td>1.279397</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>30</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   generation  epoch  best_fom   avg_fom  checkpoint_in  checkpoint_out  --h1  \\\n",
       "0           0      1  0.050737  0.557403            NaN             NaN    18   \n",
       "1           1      2  0.047372  0.442378            NaN             NaN    16   \n",
       "2           2      3  0.046571  0.614770            NaN             NaN    22   \n",
       "3           3      4  0.046571  1.279397            NaN             NaN    22   \n",
       "\n",
       "   --h2  --h3  --h4  --dropout --optimizer   --lr  \n",
       "0    28    61   153   0.213413       Nadam  0.001  \n",
       "1    32    61   153   0.213413        Adam  0.001  \n",
       "2    30    61   153   0.200000        Adam  0.001  \n",
       "3    30    61   153   0.200000        Adam  0.001  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Summary table for each generation\n",
    "summary = pd.read_csv(results_file, delim_whitespace=True)\n",
    "summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "# Plot the figure of merit as a function of generation\n",
    "plt.figure()\n",
    "plt.plot(summary.generation, summary.avg_fom, label='average loss')\n",
    "plt.plot(summary.generation, summary.best_fom, label='best loss')\n",
    "plt.xlabel('generation')\n",
    "plt.ylabel('validation loss')\n",
    "plt.legend(loc=0);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load all of the HPO search results\n",
    "deme_results = [pd.read_csv('Deme%i_%s' % (deme, results_file), delim_whitespace=True)\n",
    "                for deme in range(1, num_demes+1)]\n",
    "results = pd.concat(deme_results, ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>generation</th>\n",
       "      <th>tag</th>\n",
       "      <th>fitness</th>\n",
       "      <th>FoM</th>\n",
       "      <th>--h1</th>\n",
       "      <th>--h2</th>\n",
       "      <th>--h3</th>\n",
       "      <th>--h4</th>\n",
       "      <th>--dropout</th>\n",
       "      <th>--optimizer</th>\n",
       "      <th>--lr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.051227</td>\n",
       "      <td>16</td>\n",
       "      <td>32</td>\n",
       "      <td>64</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind1</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.693713</td>\n",
       "      <td>13</td>\n",
       "      <td>34</td>\n",
       "      <td>68</td>\n",
       "      <td>151</td>\n",
       "      <td>0.222666</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind2</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.692677</td>\n",
       "      <td>17</td>\n",
       "      <td>30</td>\n",
       "      <td>53</td>\n",
       "      <td>135</td>\n",
       "      <td>0.238800</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind3</td>\n",
       "      <td>0.150166</td>\n",
       "      <td>0.330986</td>\n",
       "      <td>13</td>\n",
       "      <td>34</td>\n",
       "      <td>60</td>\n",
       "      <td>151</td>\n",
       "      <td>0.199867</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind4</td>\n",
       "      <td>0.000052</td>\n",
       "      <td>0.689294</td>\n",
       "      <td>13</td>\n",
       "      <td>30</td>\n",
       "      <td>53</td>\n",
       "      <td>130</td>\n",
       "      <td>0.189532</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind5</td>\n",
       "      <td>0.009691</td>\n",
       "      <td>0.488709</td>\n",
       "      <td>15</td>\n",
       "      <td>28</td>\n",
       "      <td>67</td>\n",
       "      <td>122</td>\n",
       "      <td>0.198813</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind6</td>\n",
       "      <td>0.185194</td>\n",
       "      <td>0.315066</td>\n",
       "      <td>18</td>\n",
       "      <td>30</td>\n",
       "      <td>74</td>\n",
       "      <td>129</td>\n",
       "      <td>0.237594</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>deme1_ind7</td>\n",
       "      <td>0.001328</td>\n",
       "      <td>0.574136</td>\n",
       "      <td>15</td>\n",
       "      <td>27</td>\n",
       "      <td>60</td>\n",
       "      <td>129</td>\n",
       "      <td>0.192997</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind9</td>\n",
       "      <td>0.179592</td>\n",
       "      <td>0.312939</td>\n",
       "      <td>18</td>\n",
       "      <td>30</td>\n",
       "      <td>74</td>\n",
       "      <td>153</td>\n",
       "      <td>0.237594</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind10</td>\n",
       "      <td>0.999551</td>\n",
       "      <td>0.052488</td>\n",
       "      <td>18</td>\n",
       "      <td>28</td>\n",
       "      <td>60</td>\n",
       "      <td>151</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind11</td>\n",
       "      <td>0.998695</td>\n",
       "      <td>0.055508</td>\n",
       "      <td>18</td>\n",
       "      <td>28</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind12</td>\n",
       "      <td>0.334028</td>\n",
       "      <td>0.259762</td>\n",
       "      <td>17</td>\n",
       "      <td>33</td>\n",
       "      <td>68</td>\n",
       "      <td>116</td>\n",
       "      <td>0.202904</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind13</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.687076</td>\n",
       "      <td>14</td>\n",
       "      <td>30</td>\n",
       "      <td>61</td>\n",
       "      <td>130</td>\n",
       "      <td>0.169310</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind14</td>\n",
       "      <td>0.084188</td>\n",
       "      <td>0.366021</td>\n",
       "      <td>18</td>\n",
       "      <td>30</td>\n",
       "      <td>64</td>\n",
       "      <td>129</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind15</td>\n",
       "      <td>0.249488</td>\n",
       "      <td>0.286253</td>\n",
       "      <td>16</td>\n",
       "      <td>30</td>\n",
       "      <td>68</td>\n",
       "      <td>152</td>\n",
       "      <td>0.119696</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1</td>\n",
       "      <td>deme1_ind16</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.048207</td>\n",
       "      <td>18</td>\n",
       "      <td>34</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.051514</td>\n",
       "      <td>18</td>\n",
       "      <td>34</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind18</td>\n",
       "      <td>0.999997</td>\n",
       "      <td>0.055129</td>\n",
       "      <td>17</td>\n",
       "      <td>33</td>\n",
       "      <td>68</td>\n",
       "      <td>116</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind19</td>\n",
       "      <td>0.936671</td>\n",
       "      <td>0.616224</td>\n",
       "      <td>19</td>\n",
       "      <td>27</td>\n",
       "      <td>65</td>\n",
       "      <td>155</td>\n",
       "      <td>0.213683</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind20</td>\n",
       "      <td>0.920411</td>\n",
       "      <td>0.687369</td>\n",
       "      <td>15</td>\n",
       "      <td>32</td>\n",
       "      <td>68</td>\n",
       "      <td>151</td>\n",
       "      <td>0.184057</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind21</td>\n",
       "      <td>0.977300</td>\n",
       "      <td>0.385921</td>\n",
       "      <td>16</td>\n",
       "      <td>27</td>\n",
       "      <td>68</td>\n",
       "      <td>148</td>\n",
       "      <td>0.221547</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind22</td>\n",
       "      <td>0.999987</td>\n",
       "      <td>0.059223</td>\n",
       "      <td>17</td>\n",
       "      <td>33</td>\n",
       "      <td>60</td>\n",
       "      <td>151</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind23</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.051172</td>\n",
       "      <td>15</td>\n",
       "      <td>34</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2</td>\n",
       "      <td>deme1_ind24</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>7.037060</td>\n",
       "      <td>16</td>\n",
       "      <td>29</td>\n",
       "      <td>58</td>\n",
       "      <td>149</td>\n",
       "      <td>0.212392</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind25</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.534809</td>\n",
       "      <td>15</td>\n",
       "      <td>32</td>\n",
       "      <td>67</td>\n",
       "      <td>118</td>\n",
       "      <td>0.220147</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind26</td>\n",
       "      <td>0.999435</td>\n",
       "      <td>0.053793</td>\n",
       "      <td>15</td>\n",
       "      <td>34</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.235769</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind27</td>\n",
       "      <td>0.984464</td>\n",
       "      <td>0.069327</td>\n",
       "      <td>14</td>\n",
       "      <td>31</td>\n",
       "      <td>78</td>\n",
       "      <td>150</td>\n",
       "      <td>0.169537</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind28</td>\n",
       "      <td>0.959383</td>\n",
       "      <td>0.081358</td>\n",
       "      <td>19</td>\n",
       "      <td>33</td>\n",
       "      <td>60</td>\n",
       "      <td>151</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind29</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.050149</td>\n",
       "      <td>17</td>\n",
       "      <td>33</td>\n",
       "      <td>68</td>\n",
       "      <td>116</td>\n",
       "      <td>0.221547</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>3</td>\n",
       "      <td>deme1_ind30</td>\n",
       "      <td>0.999551</td>\n",
       "      <td>0.053397</td>\n",
       "      <td>19</td>\n",
       "      <td>33</td>\n",
       "      <td>64</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00100</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind2</td>\n",
       "      <td>0.999829</td>\n",
       "      <td>0.053398</td>\n",
       "      <td>19</td>\n",
       "      <td>31</td>\n",
       "      <td>62</td>\n",
       "      <td>121</td>\n",
       "      <td>0.157999</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind3</td>\n",
       "      <td>0.242353</td>\n",
       "      <td>0.293045</td>\n",
       "      <td>17</td>\n",
       "      <td>31</td>\n",
       "      <td>65</td>\n",
       "      <td>130</td>\n",
       "      <td>0.186276</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind4</td>\n",
       "      <td>0.337259</td>\n",
       "      <td>0.262926</td>\n",
       "      <td>17</td>\n",
       "      <td>34</td>\n",
       "      <td>68</td>\n",
       "      <td>113</td>\n",
       "      <td>0.198825</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind5</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.694354</td>\n",
       "      <td>18</td>\n",
       "      <td>34</td>\n",
       "      <td>62</td>\n",
       "      <td>109</td>\n",
       "      <td>0.268746</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind6</td>\n",
       "      <td>0.333326</td>\n",
       "      <td>0.264068</td>\n",
       "      <td>13</td>\n",
       "      <td>27</td>\n",
       "      <td>60</td>\n",
       "      <td>129</td>\n",
       "      <td>0.105537</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>0</td>\n",
       "      <td>deme4_ind7</td>\n",
       "      <td>0.283187</td>\n",
       "      <td>0.279348</td>\n",
       "      <td>18</td>\n",
       "      <td>30</td>\n",
       "      <td>65</td>\n",
       "      <td>149</td>\n",
       "      <td>0.200502</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>104</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind9</td>\n",
       "      <td>0.989595</td>\n",
       "      <td>0.055267</td>\n",
       "      <td>18</td>\n",
       "      <td>28</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind10</td>\n",
       "      <td>0.000456</td>\n",
       "      <td>0.261487</td>\n",
       "      <td>14</td>\n",
       "      <td>38</td>\n",
       "      <td>60</td>\n",
       "      <td>124</td>\n",
       "      <td>0.198341</td>\n",
       "      <td>Nadam</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind11</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.291491</td>\n",
       "      <td>17</td>\n",
       "      <td>31</td>\n",
       "      <td>61</td>\n",
       "      <td>130</td>\n",
       "      <td>0.186276</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind12</td>\n",
       "      <td>0.000381</td>\n",
       "      <td>0.263993</td>\n",
       "      <td>14</td>\n",
       "      <td>32</td>\n",
       "      <td>64</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind13</td>\n",
       "      <td>0.000332</td>\n",
       "      <td>0.265880</td>\n",
       "      <td>17</td>\n",
       "      <td>34</td>\n",
       "      <td>68</td>\n",
       "      <td>113</td>\n",
       "      <td>0.198825</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind14</td>\n",
       "      <td>0.991109</td>\n",
       "      <td>0.054667</td>\n",
       "      <td>19</td>\n",
       "      <td>31</td>\n",
       "      <td>62</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200623</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind15</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.047372</td>\n",
       "      <td>16</td>\n",
       "      <td>32</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>1</td>\n",
       "      <td>deme4_ind16</td>\n",
       "      <td>0.997512</td>\n",
       "      <td>0.051225</td>\n",
       "      <td>13</td>\n",
       "      <td>32</td>\n",
       "      <td>64</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind17</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>7.037060</td>\n",
       "      <td>16</td>\n",
       "      <td>31</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind18</td>\n",
       "      <td>0.999992</td>\n",
       "      <td>0.052791</td>\n",
       "      <td>13</td>\n",
       "      <td>32</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind19</td>\n",
       "      <td>0.919434</td>\n",
       "      <td>0.687249</td>\n",
       "      <td>21</td>\n",
       "      <td>33</td>\n",
       "      <td>71</td>\n",
       "      <td>135</td>\n",
       "      <td>0.234681</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind20</td>\n",
       "      <td>0.999995</td>\n",
       "      <td>0.051358</td>\n",
       "      <td>18</td>\n",
       "      <td>28</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.225205</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind21</td>\n",
       "      <td>0.999998</td>\n",
       "      <td>0.049667</td>\n",
       "      <td>13</td>\n",
       "      <td>31</td>\n",
       "      <td>62</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200623</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>117</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind22</td>\n",
       "      <td>0.999994</td>\n",
       "      <td>0.051891</td>\n",
       "      <td>19</td>\n",
       "      <td>31</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200623</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind23</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.046571</td>\n",
       "      <td>22</td>\n",
       "      <td>30</td>\n",
       "      <td>61</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>2</td>\n",
       "      <td>deme4_ind24</td>\n",
       "      <td>0.999986</td>\n",
       "      <td>0.054966</td>\n",
       "      <td>16</td>\n",
       "      <td>34</td>\n",
       "      <td>61</td>\n",
       "      <td>155</td>\n",
       "      <td>0.213413</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind25</td>\n",
       "      <td>0.919701</td>\n",
       "      <td>0.688768</td>\n",
       "      <td>13</td>\n",
       "      <td>31</td>\n",
       "      <td>71</td>\n",
       "      <td>128</td>\n",
       "      <td>0.234681</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind26</td>\n",
       "      <td>0.999988</td>\n",
       "      <td>0.057203</td>\n",
       "      <td>22</td>\n",
       "      <td>33</td>\n",
       "      <td>61</td>\n",
       "      <td>135</td>\n",
       "      <td>0.234681</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind27</td>\n",
       "      <td>0.999998</td>\n",
       "      <td>0.052355</td>\n",
       "      <td>13</td>\n",
       "      <td>31</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200623</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind28</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>7.037060</td>\n",
       "      <td>19</td>\n",
       "      <td>31</td>\n",
       "      <td>61</td>\n",
       "      <td>128</td>\n",
       "      <td>0.225205</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind29</td>\n",
       "      <td>0.918648</td>\n",
       "      <td>0.693126</td>\n",
       "      <td>16</td>\n",
       "      <td>33</td>\n",
       "      <td>63</td>\n",
       "      <td>121</td>\n",
       "      <td>0.180320</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind30</td>\n",
       "      <td>0.919795</td>\n",
       "      <td>0.688378</td>\n",
       "      <td>13</td>\n",
       "      <td>31</td>\n",
       "      <td>62</td>\n",
       "      <td>128</td>\n",
       "      <td>0.200623</td>\n",
       "      <td>Adadelta</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind31</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.049462</td>\n",
       "      <td>21</td>\n",
       "      <td>30</td>\n",
       "      <td>71</td>\n",
       "      <td>153</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>3</td>\n",
       "      <td>deme4_ind32</td>\n",
       "      <td>0.999998</td>\n",
       "      <td>0.052847</td>\n",
       "      <td>21</td>\n",
       "      <td>32</td>\n",
       "      <td>71</td>\n",
       "      <td>128</td>\n",
       "      <td>0.234681</td>\n",
       "      <td>Adam</td>\n",
       "      <td>0.00100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>128 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     generation          tag   fitness       FoM  --h1  --h2  --h3  --h4  \\\n",
       "0             0   deme1_ind0  1.000000  0.051227    16    32    64   128   \n",
       "1             0   deme1_ind1  0.000045  0.693713    13    34    68   151   \n",
       "2             0   deme1_ind2  0.000047  0.692677    17    30    53   135   \n",
       "3             0   deme1_ind3  0.150166  0.330986    13    34    60   151   \n",
       "4             0   deme1_ind4  0.000052  0.689294    13    30    53   130   \n",
       "..          ...          ...       ...       ...   ...   ...   ...   ...   \n",
       "123           3  deme4_ind28  0.000045  7.037060    19    31    61   128   \n",
       "124           3  deme4_ind29  0.918648  0.693126    16    33    63   121   \n",
       "125           3  deme4_ind30  0.919795  0.688378    13    31    62   128   \n",
       "126           3  deme4_ind31  1.000000  0.049462    21    30    71   153   \n",
       "127           3  deme4_ind32  0.999998  0.052847    21    32    71   128   \n",
       "\n",
       "     --dropout --optimizer     --lr  \n",
       "0     0.200000        Adam  0.00100  \n",
       "1     0.222666    Adadelta  0.00001  \n",
       "2     0.238800    Adadelta  0.00010  \n",
       "3     0.199867        Adam  0.00001  \n",
       "4     0.189532    Adadelta  0.00100  \n",
       "..         ...         ...      ...  \n",
       "123   0.225205        Adam  0.10000  \n",
       "124   0.180320    Adadelta  0.00001  \n",
       "125   0.200623    Adadelta  0.00100  \n",
       "126   0.200000        Adam  0.00100  \n",
       "127   0.234681        Adam  0.00100  \n",
       "\n",
       "[128 rows x 11 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x2520 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Make scatter plots of the parameters\n",
    "hps = ['h1', 'h2', 'h3', 'h4', 'dropout', 'optimizer', 'lr']\n",
    "fig, axs = plt.subplots(nrows=len(hps), figsize=(6, 5 * len(hps)))\n",
    "\n",
    "gen_groups = results.groupby('generation')\n",
    "    \n",
    "# Loop over HPs\n",
    "for i, hp in enumerate(hps):\n",
    "    \n",
    "    # Loop over generations\n",
    "    for gen in range(generations):\n",
    "        gen_results = gen_groups.get_group(gen)\n",
    "        axs[i].scatter(gen_results['--'+hp], gen_results.FoM, label=gen)\n",
    "    \n",
    "    # Finish the plot\n",
    "    axs[i].set_title(hp)\n",
    "    axs[i].set_xlabel('Parameter value')\n",
    "    axs[i].set_ylabel('Loss')\n",
    "    axs[i].set_ylim(bottom=0, top=1)\n",
    "    axs[i].legend(title='generation', loc=0)\n",
    "    \n",
    "    if hp == 'lr':\n",
    "        axs[i].set_xlim((1e-6, 1))\n",
    "        axs[i].set_xscale('log')\n",
    "\n",
    "plt.tight_layout()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tensorflow-intel(cpu)/1.13.1-py36",
   "language": "python",
   "name": "tensorflow_intel_1.13.1_py36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}