View Results.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path = './results/1538450275/'\n",
    "\n",
    "lc = pd.read_csv(model_path + '/learning_curve.tsv', sep='\\t')\n",
    "mi = pd.read_csv(model_path + '/model_info.tsv', sep='\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Unnamed: 0</th>\n",
       "      <th>batch_size</th>\n",
       "      <th>lr</th>\n",
       "      <th>name</th>\n",
       "      <th>window_size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>4096</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>conv_nn_2</td>\n",
       "      <td>1024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  batch_size      lr       name  window_size\n",
       "0           0        4096  0.0001  conv_nn_2         1024"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Unnamed: 0</th>\n",
       "      <th>Batch</th>\n",
       "      <th>Epoch</th>\n",
       "      <th>Mean Loss (epoch)</th>\n",
       "      <th>Mean Loss (last batches)</th>\n",
       "      <th>Valid Mean AUC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.161631</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>488.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.143228</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>732.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.137514</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>976.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.108300</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>1220.0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.134482</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0   Batch  Epoch  Mean Loss (epoch)  Mean Loss (last batches)  \\\n",
       "0           0   244.0      0                NaN                  0.161631   \n",
       "1           1   488.0      0                NaN                  0.143228   \n",
       "2           2   732.0      0                NaN                  0.137514   \n",
       "3           3   976.0      0                NaN                  0.108300   \n",
       "4           4  1220.0      0                NaN                  0.134482   \n",
       "\n",
       "   Valid Mean AUC  \n",
       "0             NaN  \n",
       "1             NaN  \n",
       "2             NaN  \n",
       "3             NaN  \n",
       "4             NaN  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lc.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Unnamed: 0</th>\n",
       "      <th>Batch</th>\n",
       "      <th>Epoch</th>\n",
       "      <th>Mean Loss (epoch)</th>\n",
       "      <th>Mean Loss (last batches)</th>\n",
       "      <th>Valid Mean AUC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>15</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0.114410</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.768198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>31</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0.100170</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.834621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>47</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>0.095574</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.861285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>63</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0.091160</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.868326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>79</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>0.088264</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.887178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>95</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "      <td>0.085112</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.891162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>111</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "      <td>0.082924</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.898618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>127</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7</td>\n",
       "      <td>0.081321</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.897038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>143</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "      <td>0.079686</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.897995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>159</th>\n",
       "      <td>159</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9</td>\n",
       "      <td>0.078212</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.872972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>175</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "      <td>0.076901</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.904485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>191</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>0.075860</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.896877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>207</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12</td>\n",
       "      <td>0.075045</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.897501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>223</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13</td>\n",
       "      <td>0.073793</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.884624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>239</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14</td>\n",
       "      <td>0.073055</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.890622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>255</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15</td>\n",
       "      <td>0.072058</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.907375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0  Batch  Epoch  Mean Loss (epoch)  Mean Loss (last batches)  \\\n",
       "15           15    NaN      0           0.114410                       NaN   \n",
       "31           31    NaN      1           0.100170                       NaN   \n",
       "47           47    NaN      2           0.095574                       NaN   \n",
       "63           63    NaN      3           0.091160                       NaN   \n",
       "79           79    NaN      4           0.088264                       NaN   \n",
       "95           95    NaN      5           0.085112                       NaN   \n",
       "111         111    NaN      6           0.082924                       NaN   \n",
       "127         127    NaN      7           0.081321                       NaN   \n",
       "143         143    NaN      8           0.079686                       NaN   \n",
       "159         159    NaN      9           0.078212                       NaN   \n",
       "175         175    NaN     10           0.076901                       NaN   \n",
       "191         191    NaN     11           0.075860                       NaN   \n",
       "207         207    NaN     12           0.075045                       NaN   \n",
       "223         223    NaN     13           0.073793                       NaN   \n",
       "239         239    NaN     14           0.073055                       NaN   \n",
       "255         255    NaN     15           0.072058                       NaN   \n",
       "\n",
       "     Valid Mean AUC  \n",
       "15         0.768198  \n",
       "31         0.834621  \n",
       "47         0.861285  \n",
       "63         0.868326  \n",
       "79         0.887178  \n",
       "95         0.891162  \n",
       "111        0.898618  \n",
       "127        0.897038  \n",
       "143        0.897995  \n",
       "159        0.872972  \n",
       "175        0.904485  \n",
       "191        0.896877  \n",
       "207        0.897501  \n",
       "223        0.884624  \n",
       "239        0.890622  \n",
       "255        0.907375  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter learning curve only for points with Mean Loss for the entire Epoch\n",
    "epoch_filt_lc = lc.dropna(axis=0, subset=['Mean Loss (epoch)'])\n",
    "epoch_filt_lc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ltd/anaconda3/envs/openke/lib/python2.7/site-packages/matplotlib/figure.py:402: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwIAAAHjCAYAAACQO6i2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9319bdbad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax1 = plt.subplots()\n",
    "fig.set_size_inches(12, 8)\n",
    "ax1.plot(epoch_filt_lc['Epoch'], epoch_filt_lc['Mean Loss (epoch)'], 'b-')\n",
    "ax1.set_ylabel('Mean Loss (epoch)', color='b')\n",
    "ax1.tick_params('y', colors='b')\n",
    "ax1.set_xlabel('Epoch')\n",
    "ax2 = ax1.twinx()\n",
    "ax2.plot(epoch_filt_lc['Epoch'], epoch_filt_lc['Valid Mean AUC'], 'r-', label='Valid Mean AUC')\n",
    "ax2.set_ylabel('Valid Mean AUC', color='r')\n",
    "ax2.tick_params('y', colors='r')\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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
}