Visualization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>epoch</th>\n",
       "      <th>loss</th>\n",
       "      <th>acc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3.518767</td>\n",
       "      <td>0.141574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>3.167011</td>\n",
       "      <td>0.163365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2.948693</td>\n",
       "      <td>0.208929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2.617472</td>\n",
       "      <td>0.279911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2.348704</td>\n",
       "      <td>0.340681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>2.138005</td>\n",
       "      <td>0.389816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>2.003399</td>\n",
       "      <td>0.423410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>1.909276</td>\n",
       "      <td>0.447824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1.841778</td>\n",
       "      <td>0.464118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1.764172</td>\n",
       "      <td>0.485296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>1.700561</td>\n",
       "      <td>0.502651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>1.640164</td>\n",
       "      <td>0.513867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>13</td>\n",
       "      <td>1.582906</td>\n",
       "      <td>0.523661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "      <td>1.529423</td>\n",
       "      <td>0.535128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>1.494608</td>\n",
       "      <td>0.542076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>16</td>\n",
       "      <td>1.457570</td>\n",
       "      <td>0.546680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>17</td>\n",
       "      <td>1.425620</td>\n",
       "      <td>0.553990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>18</td>\n",
       "      <td>1.397976</td>\n",
       "      <td>0.559626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>19</td>\n",
       "      <td>1.367130</td>\n",
       "      <td>0.566797</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>20</td>\n",
       "      <td>1.344677</td>\n",
       "      <td>0.573800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>21</td>\n",
       "      <td>1.313548</td>\n",
       "      <td>0.582366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>22</td>\n",
       "      <td>1.293113</td>\n",
       "      <td>0.584794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>23</td>\n",
       "      <td>1.264488</td>\n",
       "      <td>0.589872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>24</td>\n",
       "      <td>1.238285</td>\n",
       "      <td>0.598242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>25</td>\n",
       "      <td>1.220307</td>\n",
       "      <td>0.602790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>26</td>\n",
       "      <td>1.197404</td>\n",
       "      <td>0.608064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>27</td>\n",
       "      <td>1.172865</td>\n",
       "      <td>0.617997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>28</td>\n",
       "      <td>1.157766</td>\n",
       "      <td>0.621122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>29</td>\n",
       "      <td>1.142872</td>\n",
       "      <td>0.626702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>30</td>\n",
       "      <td>1.115034</td>\n",
       "      <td>0.635575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>31</td>\n",
       "      <td>1.099478</td>\n",
       "      <td>0.638756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>32</td>\n",
       "      <td>1.075674</td>\n",
       "      <td>0.648493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>33</td>\n",
       "      <td>1.055952</td>\n",
       "      <td>0.651758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>34</td>\n",
       "      <td>1.028288</td>\n",
       "      <td>0.664927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>35</td>\n",
       "      <td>1.012214</td>\n",
       "      <td>0.667690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>36</td>\n",
       "      <td>0.989278</td>\n",
       "      <td>0.676311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>37</td>\n",
       "      <td>0.957720</td>\n",
       "      <td>0.687026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>38</td>\n",
       "      <td>0.928484</td>\n",
       "      <td>0.695703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>39</td>\n",
       "      <td>0.907364</td>\n",
       "      <td>0.702065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>40</td>\n",
       "      <td>0.881311</td>\n",
       "      <td>0.712277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>41</td>\n",
       "      <td>0.861582</td>\n",
       "      <td>0.721122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>42</td>\n",
       "      <td>0.836730</td>\n",
       "      <td>0.727567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>43</td>\n",
       "      <td>0.812460</td>\n",
       "      <td>0.734040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>44</td>\n",
       "      <td>0.786593</td>\n",
       "      <td>0.742327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>45</td>\n",
       "      <td>0.770886</td>\n",
       "      <td>0.750391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>46</td>\n",
       "      <td>0.759511</td>\n",
       "      <td>0.753739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>47</td>\n",
       "      <td>0.735362</td>\n",
       "      <td>0.761217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>48</td>\n",
       "      <td>0.711351</td>\n",
       "      <td>0.769671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>49</td>\n",
       "      <td>0.692834</td>\n",
       "      <td>0.773298</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>50</td>\n",
       "      <td>0.674972</td>\n",
       "      <td>0.781055</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    epoch      loss       acc\n",
       "0       1  3.518767  0.141574\n",
       "1       2  3.167011  0.163365\n",
       "2       3  2.948693  0.208929\n",
       "3       4  2.617472  0.279911\n",
       "4       5  2.348704  0.340681\n",
       "5       6  2.138005  0.389816\n",
       "6       7  2.003399  0.423410\n",
       "7       8  1.909276  0.447824\n",
       "8       9  1.841778  0.464118\n",
       "9      10  1.764172  0.485296\n",
       "10     11  1.700561  0.502651\n",
       "11     12  1.640164  0.513867\n",
       "12     13  1.582906  0.523661\n",
       "13     14  1.529423  0.535128\n",
       "14     15  1.494608  0.542076\n",
       "15     16  1.457570  0.546680\n",
       "16     17  1.425620  0.553990\n",
       "17     18  1.397976  0.559626\n",
       "18     19  1.367130  0.566797\n",
       "19     20  1.344677  0.573800\n",
       "20     21  1.313548  0.582366\n",
       "21     22  1.293113  0.584794\n",
       "22     23  1.264488  0.589872\n",
       "23     24  1.238285  0.598242\n",
       "24     25  1.220307  0.602790\n",
       "25     26  1.197404  0.608064\n",
       "26     27  1.172865  0.617997\n",
       "27     28  1.157766  0.621122\n",
       "28     29  1.142872  0.626702\n",
       "29     30  1.115034  0.635575\n",
       "30     31  1.099478  0.638756\n",
       "31     32  1.075674  0.648493\n",
       "32     33  1.055952  0.651758\n",
       "33     34  1.028288  0.664927\n",
       "34     35  1.012214  0.667690\n",
       "35     36  0.989278  0.676311\n",
       "36     37  0.957720  0.687026\n",
       "37     38  0.928484  0.695703\n",
       "38     39  0.907364  0.702065\n",
       "39     40  0.881311  0.712277\n",
       "40     41  0.861582  0.721122\n",
       "41     42  0.836730  0.727567\n",
       "42     43  0.812460  0.734040\n",
       "43     44  0.786593  0.742327\n",
       "44     45  0.770886  0.750391\n",
       "45     46  0.759511  0.753739\n",
       "46     47  0.735362  0.761217\n",
       "47     48  0.711351  0.769671\n",
       "48     49  0.692834  0.773298\n",
       "49     50  0.674972  0.781055"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('logs/training_log.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAHpCAYAAABA0XIiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10e71bb10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(df['epoch'], df['loss'])\n",
    "plt.title('Training Loss over Epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAHpCAYAAABA0XIiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10e82c350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(df['epoch'], df['acc'])\n",
    "plt.title('Training Accuracy over Epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}