Hyperparam_search_results.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import glob\n",
    "import os\n",
    "import re\n",
    "import yaml\n",
    "import pprint\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "RESULTS_DIR = '/home/sprasanna/Projects/training-datasets/'\n",
    "logfiles = glob.glob(os.path.join(RESULTS_DIR,'results*'))\n",
    "logfiles.sort(key=lambda f: int(filter(str.isdigit, f)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = pd.DataFrame(columns=['Hyp_param_set_number', 'Accuracy','Total_time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "for nfiles in range(len(logfiles)):\n",
    "    _,filename = os.path.split(logfiles[nfiles])\n",
    "    df = pd.read_csv(logfiles[nfiles],header=None,names=['Results','Values'],index_col=0)\n",
    "    results.loc[nfiles] = [filename,df.loc['test acc','Values'],df.loc['total time','Values']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_accuracy = results['Accuracy'].idxmax()\n",
    "accuracy_value = results.loc[max_accuracy, 'Accuracy']\n",
    "max_accuracy_experiment = results.loc[max_accuracy, 'Hyp_param_set_number']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "regex = re.compile(r'\\d+')\n",
    "n_experiment_set = int(regex.findall(max_accuracy_experiment)[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results of hyperparameter sweep experiment:\n",
      "-------------------------------------------\n",
      "Maximum accuracy: 0.9426\n",
      "Experiment set with max accuracy: results_job_id_2.log\n",
      "\n",
      "{'batch_norm': 'on',\n",
      " 'batch_size': 512,\n",
      " 'data_aug_cutout_size': 5,\n",
      " 'max_learning_rate': 0.4,\n",
      " 'momentum': 0.9}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fada2272a50>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Results of hyperparameter sweep experiment:')\n",
    "print('-------------------------------------------')\n",
    "print('Maximum accuracy: {}'.format(accuracy_value))\n",
    "print('Experiment set with max accuracy: {}\\n'.format(max_accuracy_experiment))\n",
    "\n",
    "with open(\"hyperparams.yml\", 'r') as stream:\n",
    "    hyper_param_set = yaml.load(stream)\n",
    "pprint.pprint(hyper_param_set[n_experiment_set-1][\"hyperparam_set\"])\n",
    "\n",
    "results.plot(x='Hyp_param_set_number',y='Accuracy',kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Hyp_param_set_number</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Total_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>results_job_id_1.log</td>\n",
       "      <td>0.9347</td>\n",
       "      <td>117.764999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>results_job_id_2.log</td>\n",
       "      <td>0.9426</td>\n",
       "      <td>75.399140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>results_job_id_3.log</td>\n",
       "      <td>0.9301</td>\n",
       "      <td>119.592690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>results_job_id_4.log</td>\n",
       "      <td>0.9397</td>\n",
       "      <td>75.619865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>results_job_id_5.log</td>\n",
       "      <td>0.8285</td>\n",
       "      <td>117.680233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>results_job_id_6.log</td>\n",
       "      <td>0.8883</td>\n",
       "      <td>74.556600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>results_job_id_7.log</td>\n",
       "      <td>0.8129</td>\n",
       "      <td>114.646887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>results_job_id_8.log</td>\n",
       "      <td>0.8737</td>\n",
       "      <td>74.983109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>results_job_id_9.log</td>\n",
       "      <td>0.9274</td>\n",
       "      <td>116.857387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>results_job_id_10.log</td>\n",
       "      <td>0.9419</td>\n",
       "      <td>75.062058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>results_job_id_11.log</td>\n",
       "      <td>0.9213</td>\n",
       "      <td>116.772228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>results_job_id_12.log</td>\n",
       "      <td>0.9376</td>\n",
       "      <td>75.540528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>results_job_id_13.log</td>\n",
       "      <td>0.8182</td>\n",
       "      <td>116.791409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>results_job_id_14.log</td>\n",
       "      <td>0.8741</td>\n",
       "      <td>74.523308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>results_job_id_15.log</td>\n",
       "      <td>0.8042</td>\n",
       "      <td>117.292228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>results_job_id_16.log</td>\n",
       "      <td>0.8634</td>\n",
       "      <td>74.602888</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Hyp_param_set_number  Accuracy  Total_time\n",
       "0    results_job_id_1.log    0.9347  117.764999\n",
       "1    results_job_id_2.log    0.9426   75.399140\n",
       "2    results_job_id_3.log    0.9301  119.592690\n",
       "3    results_job_id_4.log    0.9397   75.619865\n",
       "4    results_job_id_5.log    0.8285  117.680233\n",
       "5    results_job_id_6.log    0.8883   74.556600\n",
       "6    results_job_id_7.log    0.8129  114.646887\n",
       "7    results_job_id_8.log    0.8737   74.983109\n",
       "8    results_job_id_9.log    0.9274  116.857387\n",
       "9   results_job_id_10.log    0.9419   75.062058\n",
       "10  results_job_id_11.log    0.9213  116.772228\n",
       "11  results_job_id_12.log    0.9376   75.540528\n",
       "12  results_job_id_13.log    0.8182  116.791409\n",
       "13  results_job_id_14.log    0.8741   74.523308\n",
       "14  results_job_id_15.log    0.8042  117.292228\n",
       "15  results_job_id_16.log    0.8634   74.602888"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}