evaluation_n_vgg.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from ast import literal_eval\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "df = pd.read_csv(\"vgg_n.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['accuracy', 'activation', 'dataset', 'dropout_rate', 'epochs',\n",
       "       'input_dims', 'loss', 'lr_rate', 'model_name', 'test_accuracy',\n",
       "       'test_loss', 'val_accuracy', 'val_loss', 'opt'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def literal_return(val):\n",
    "    try:\n",
    "        return literal_eval(val)\n",
    "    except (ValueError, SyntaxError) as e:\n",
    "        return val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in list(df.columns):\n",
    "    df[col] = df.apply(lambda x: literal_return(str(x[col])),axis=1)"
   ]
  },
  {
   "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>accuracy</th>\n",
       "      <th>val_accuracy</th>\n",
       "      <th>loss</th>\n",
       "      <th>opt</th>\n",
       "      <th>test_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[0.49962687, 0.9720895, 0.9841791, 0.9889055, ...</td>\n",
       "      <td>[0.9598990082740784, 0.9691919088363647, 0.985...</td>\n",
       "      <td>[1.402208507883395, 0.09357348736690645, 0.053...</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.9951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[0.09793533, 0.0979602, 0.0979602, 0.0979602, ...</td>\n",
       "      <td>[0.10000000149011612, 0.10000000149011612, 0.1...</td>\n",
       "      <td>[14.502014120680776, 14.539163346171971, 14.53...</td>\n",
       "      <td>adam</td>\n",
       "      <td>0.0958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[0.10166667, 0.10189055, 0.10189055, 0.1018905...</td>\n",
       "      <td>[0.10277777910232544, 0.10277777910232544, 0.1...</td>\n",
       "      <td>[14.440315297444661, 14.475813645699724, 14.47...</td>\n",
       "      <td>adagrad</td>\n",
       "      <td>0.1010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[0.17337313, 0.34779105, 0.4544776, 0.5358806,...</td>\n",
       "      <td>[0.27218180894851685, 0.4153939485549927, 0.49...</td>\n",
       "      <td>[2.1972184184914205, 1.7556254367401352, 1.483...</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.7213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[0.09859701, 0.09877612, 0.09877612, 0.0987761...</td>\n",
       "      <td>[0.10248485207557678, 0.10248485207557678, 0.1...</td>\n",
       "      <td>[14.481889267651002, 14.5260122756958, 14.5260...</td>\n",
       "      <td>adam</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[0.098656714, 0.098985076, 0.09877612, 0.09850...</td>\n",
       "      <td>[0.10248485207557678, 0.10248485207557678, 0.1...</td>\n",
       "      <td>[14.480491255802894, 14.522644347916788, 14.52...</td>\n",
       "      <td>adagrad</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            accuracy  \\\n",
       "0  [0.49962687, 0.9720895, 0.9841791, 0.9889055, ...   \n",
       "1  [0.09793533, 0.0979602, 0.0979602, 0.0979602, ...   \n",
       "2  [0.10166667, 0.10189055, 0.10189055, 0.1018905...   \n",
       "3  [0.17337313, 0.34779105, 0.4544776, 0.5358806,...   \n",
       "4  [0.09859701, 0.09877612, 0.09877612, 0.0987761...   \n",
       "5  [0.098656714, 0.098985076, 0.09877612, 0.09850...   \n",
       "\n",
       "                                        val_accuracy  \\\n",
       "0  [0.9598990082740784, 0.9691919088363647, 0.985...   \n",
       "1  [0.10000000149011612, 0.10000000149011612, 0.1...   \n",
       "2  [0.10277777910232544, 0.10277777910232544, 0.1...   \n",
       "3  [0.27218180894851685, 0.4153939485549927, 0.49...   \n",
       "4  [0.10248485207557678, 0.10248485207557678, 0.1...   \n",
       "5  [0.10248485207557678, 0.10248485207557678, 0.1...   \n",
       "\n",
       "                                                loss      opt  test_accuracy  \n",
       "0  [1.402208507883395, 0.09357348736690645, 0.053...      sgd         0.9951  \n",
       "1  [14.502014120680776, 14.539163346171971, 14.53...     adam         0.0958  \n",
       "2  [14.440315297444661, 14.475813645699724, 14.47...  adagrad         0.1010  \n",
       "3  [2.1972184184914205, 1.7556254367401352, 1.483...      sgd         0.7213  \n",
       "4  [14.481889267651002, 14.5260122756958, 14.5260...     adam         0.1000  \n",
       "5  [14.480491255802894, 14.522644347916788, 14.52...  adagrad         0.1000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df[0:3][:,J]\n",
    "d_opt = df[['accuracy','val_accuracy','loss','opt','test_accuracy']][0:6]\n",
    "d_opt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ((ax1, ax2, ax3), (ax4, ax5, ax6)) = plt.subplots(nrows=2, ncols=3, sharex=True, sharey=True, figsize=(9,6))\n",
    "ax1.set_title('sgd_mnist')\n",
    "ax1.plot(d_opt['accuracy'][0] ,'r')\n",
    "ax1.plot(d_opt['val_accuracy'][0] ,'b')\n",
    "# ax1.plot(d_opt['loss'][0], 'g' )\n",
    "\n",
    "ax2.set_title('adam_mnist')\n",
    "ax2.plot(d_opt['accuracy'][1],'r')\n",
    "ax2.plot(d_opt['val_accuracy'][1] ,'b')\n",
    "# ax2.plot(d_opt['loss'][1] ,'g')\n",
    "\n",
    "ax3.set_title('adagrad_mnist')\n",
    "ax3.plot(d_opt['accuracy'][2] , 'r')\n",
    "ax3.plot(d_opt['val_accuracy'][2] ,'b')\n",
    "# ax3.plot(d_opt['loss'][2] ,'g')\n",
    "\n",
    "ax4.set_title('sgd_cifar10')\n",
    "ax4.plot(d_opt['accuracy'][3] ,'r')\n",
    "ax4.plot(d_opt['val_accuracy'][3] ,'b')\n",
    "# ax4.plot(d_opt['loss'][3] ,'g')\n",
    "\n",
    "ax5.set_title('adam_cifar10')\n",
    "ax5.plot(d_opt['accuracy'][4] ,'r')\n",
    "ax5.plot(d_opt['val_accuracy'][4] ,'b')\n",
    "# ax5.plot(d_opt['loss'][4] ,'g')\n",
    "\n",
    "ax6.set_title('adagrad_cifar10')\n",
    "ax6.plot(d_opt['accuracy'][5] ,'r')\n",
    "ax6.plot(d_opt['val_accuracy'][5] ,'b')\n",
    "# ax6.plot(d_opt['loss'][5] ,'g')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ((ax1, ax2)) = plt.subplots(nrows=1, ncols=2, sharex=True, sharey=True,figsize=(6,3))\n",
    "ax1.set_title('MNIST Loss')\n",
    "ax1.plot(d_opt['loss'][0],'r') # sgd\n",
    "ax1.plot(d_opt['loss'][1],'g') # adam\n",
    "ax1.plot(d_opt['loss'][2],'b') # adagrad\n",
    "ax2.set_title('CIFAR10 Loss') \n",
    "ax2.plot(d_opt['loss'][3],'r')  # sgd\n",
    "ax2.plot(d_opt['loss'][4],'g') #adam\n",
    "ax2.plot(d_opt['loss'][5],'b') # adagrad\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 144x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ((ax1)) = plt.subplots(nrows=1, ncols=1, sharex=True, sharey=True,figsize=(2,2))\n",
    "width = 0.3\n",
    "ind= np.arange(len(d_opt[['opt']])/2)\n",
    "ax1.bar(ind, list(d_opt['test_accuracy'].values)[0:3],width,color='g')\n",
    "ax1.bar(ind + width, list(d_opt['test_accuracy'].values)[3:6],width,color='b')\n",
    "ax1.set_xticks(ind)\n",
    "ax1.set_xticklabels(list(d_opt['opt'].values))\n",
    "plt.show()"
   ]
  },
  {
   "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>opt</th>\n",
       "      <th>dataset</th>\n",
       "      <th>lr_rate</th>\n",
       "      <th>test_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.4935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.6647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>0.7213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   opt  dataset  lr_rate  test_accuracy\n",
       "7  sgd  cifar10   0.0001         0.4935\n",
       "6  sgd  cifar10   0.0010         0.6647\n",
       "3  sgd  cifar10   0.0100         0.7213\n",
       "8  sgd  cifar10   0.1000         0.1000\n",
       "9  sgd  cifar10   0.5000         0.1000"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "frames = [df[['opt','dataset','lr_rate','test_accuracy']][3:4],df[['opt','dataset','lr_rate','test_accuracy']][6:10]]\n",
    "df_lr_rate = pd.concat(frames,ignore_index=False)#.reset_index(drop=True)\n",
    "df_lr_rate = df_lr_rate.sort_values(by=['lr_rate'])\n",
    "ax = df_lr_rate.plot.barh(x='lr_rate', y='test_accuracy',figsize=(4,2),color='b',title='CIFAR10 test accuracy on different Learning Rates')\n",
    "df_lr_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>opt</th>\n",
       "      <th>dataset</th>\n",
       "      <th>activation</th>\n",
       "      <th>test_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>relu</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>tanh</td>\n",
       "      <td>0.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>0.7130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sgd</td>\n",
       "      <td>mnist</td>\n",
       "      <td>softmax</td>\n",
       "      <td>0.9951</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    opt  dataset activation  test_accuracy\n",
       "10  sgd  cifar10       relu         0.1000\n",
       "11  sgd  cifar10       tanh         0.1000\n",
       "12  sgd  cifar10    sigmoid         0.7130\n",
       "0   sgd    mnist    softmax         0.9951"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "frames = [df[['opt','dataset','activation','test_accuracy']][0:1],df[['opt','dataset','activation','test_accuracy']][10:13]]\n",
    "d_act_rate = pd.concat(frames,ignore_index=False)#.reset_index(drop=True)\n",
    "d_act_rate = d_act_rate.sort_values(by=['test_accuracy'])\n",
    "ax = d_act_rate.plot.barh(x='activation', y='test_accuracy',figsize=(4,2),color='b',title='CIFAR10 Test Accuracy on different Activation Functions')\n",
    "d_act_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>opt</th>\n",
       "      <th>dataset</th>\n",
       "      <th>dropout_rate</th>\n",
       "      <th>test_accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.7061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.7213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.7372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>sgd</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.6904</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    opt  dataset  dropout_rate  test_accuracy\n",
       "13  sgd  cifar10           0.1         0.7061\n",
       "3   sgd  cifar10           0.2         0.7213\n",
       "14  sgd  cifar10           0.5         0.7372\n",
       "15  sgd  cifar10           0.9         0.6904"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "frames = [df[['opt','dataset','dropout_rate','test_accuracy']][3:4],df[['opt','dataset','dropout_rate','test_accuracy']][13:16]]\n",
    "d_drp_rate = pd.concat(frames,ignore_index=False)#.reset_index(drop=True)\n",
    "d_drp_rate = d_drp_rate.sort_values(by=['dropout_rate'])\n",
    "list(d_drp_rate['dropout_rate'].values)\n",
    "ax = d_drp_rate.plot.scatter(x='dropout_rate', y='test_accuracy',figsize=(4,2),color='b',xticks=list(d_drp_rate['dropout_rate'].values),xlim=[0,1],title='CIFAR10 Test Accuracy on different Dropout Rate')\n",
    "d_drp_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "vector_cols = ['accuracy', 'val_accuracy', 'val_loss', 'loss']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in vector_cols:\n",
    "    df[col] = df.apply(lambda x: x[col][-1],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>model_name</th>\n",
       "      <th>activation</th>\n",
       "      <th>dataset</th>\n",
       "      <th>dropout_rate</th>\n",
       "      <th>epochs</th>\n",
       "      <th>input_dims</th>\n",
       "      <th>lr_rate</th>\n",
       "      <th>opt</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>val_accuracy</th>\n",
       "      <th>loss</th>\n",
       "      <th>test_accuracy</th>\n",
       "      <th>test_loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>mnist</td>\n",
       "      <td>0.2</td>\n",
       "      <td>15.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.999702</td>\n",
       "      <td>0.993687</td>\n",
       "      <td>1.088761e-03</td>\n",
       "      <td>0.9951</td>\n",
       "      <td>2.224292e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>mnist</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>adam</td>\n",
       "      <td>0.097960</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>1.453916e+01</td>\n",
       "      <td>0.0958</td>\n",
       "      <td>1.457398e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>mnist</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>adagrad</td>\n",
       "      <td>0.101891</td>\n",
       "      <td>0.102778</td>\n",
       "      <td>1.447581e+01</td>\n",
       "      <td>0.1010</td>\n",
       "      <td>1.449017e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.825284</td>\n",
       "      <td>0.733697</td>\n",
       "      <td>4.982573e-01</td>\n",
       "      <td>0.7213</td>\n",
       "      <td>8.661988e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>adam</td>\n",
       "      <td>0.098776</td>\n",
       "      <td>0.102485</td>\n",
       "      <td>1.452601e+01</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>1.450629e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>adagrad</td>\n",
       "      <td>0.098567</td>\n",
       "      <td>0.102485</td>\n",
       "      <td>1.452938e+01</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>1.450629e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.727194</td>\n",
       "      <td>0.666061</td>\n",
       "      <td>7.761270e-01</td>\n",
       "      <td>0.6647</td>\n",
       "      <td>9.896639e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.472657</td>\n",
       "      <td>0.493212</td>\n",
       "      <td>1.477908e+00</td>\n",
       "      <td>0.4935</td>\n",
       "      <td>1.420535e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.103552</td>\n",
       "      <td>0.095879</td>\n",
       "      <td>2.304036e+00</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>2.303477e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.098896</td>\n",
       "      <td>0.100727</td>\n",
       "      <td>1.452409e+01</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>1.450629e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>relu</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.098716</td>\n",
       "      <td>0.102606</td>\n",
       "      <td>2.303384e+00</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>2.302689e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>tanh</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.100657</td>\n",
       "      <td>0.098667</td>\n",
       "      <td>1.192093e-07</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>1.192093e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>sigmoid</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.2</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.798328</td>\n",
       "      <td>0.717697</td>\n",
       "      <td>5.752599e-01</td>\n",
       "      <td>0.7130</td>\n",
       "      <td>8.386861e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.803493</td>\n",
       "      <td>0.707030</td>\n",
       "      <td>5.636663e-01</td>\n",
       "      <td>0.7061</td>\n",
       "      <td>8.835897e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.5</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.828657</td>\n",
       "      <td>0.740667</td>\n",
       "      <td>4.955358e-01</td>\n",
       "      <td>0.7372</td>\n",
       "      <td>8.148351e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>VGG16</td>\n",
       "      <td>softmax</td>\n",
       "      <td>cifar10</td>\n",
       "      <td>0.9</td>\n",
       "      <td>10.0</td>\n",
       "      <td>(224, 224, 3)</td>\n",
       "      <td>0.0100</td>\n",
       "      <td>sgd</td>\n",
       "      <td>0.731373</td>\n",
       "      <td>0.692485</td>\n",
       "      <td>7.939504e-01</td>\n",
       "      <td>0.6904</td>\n",
       "      <td>9.208996e-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   model_name activation  dataset  dropout_rate  epochs     input_dims  \\\n",
       "0       VGG16    softmax    mnist           0.2    15.0  (224, 224, 3)   \n",
       "1       VGG16    softmax    mnist           0.2    10.0  (224, 224, 3)   \n",
       "2       VGG16    softmax    mnist           0.2    10.0  (224, 224, 3)   \n",
       "3       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "4       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "5       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "6       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "7       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "8       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "9       VGG16    softmax  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "10      VGG16       relu  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "11      VGG16       tanh  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "12      VGG16    sigmoid  cifar10           0.2    10.0  (224, 224, 3)   \n",
       "13      VGG16    softmax  cifar10           0.1    10.0  (224, 224, 3)   \n",
       "14      VGG16    softmax  cifar10           0.5    10.0  (224, 224, 3)   \n",
       "15      VGG16    softmax  cifar10           0.9    10.0  (224, 224, 3)   \n",
       "\n",
       "    lr_rate      opt  accuracy  val_accuracy          loss  test_accuracy  \\\n",
       "0    0.0100      sgd  0.999702      0.993687  1.088761e-03         0.9951   \n",
       "1    0.0100     adam  0.097960      0.100000  1.453916e+01         0.0958   \n",
       "2    0.0100  adagrad  0.101891      0.102778  1.447581e+01         0.1010   \n",
       "3    0.0100      sgd  0.825284      0.733697  4.982573e-01         0.7213   \n",
       "4    0.0100     adam  0.098776      0.102485  1.452601e+01         0.1000   \n",
       "5    0.0100  adagrad  0.098567      0.102485  1.452938e+01         0.1000   \n",
       "6    0.0010      sgd  0.727194      0.666061  7.761270e-01         0.6647   \n",
       "7    0.0001      sgd  0.472657      0.493212  1.477908e+00         0.4935   \n",
       "8    0.1000      sgd  0.103552      0.095879  2.304036e+00         0.1000   \n",
       "9    0.5000      sgd  0.098896      0.100727  1.452409e+01         0.1000   \n",
       "10   0.0100      sgd  0.098716      0.102606  2.303384e+00         0.1000   \n",
       "11   0.0100      sgd  0.100657      0.098667  1.192093e-07         0.1000   \n",
       "12   0.0100      sgd  0.798328      0.717697  5.752599e-01         0.7130   \n",
       "13   0.0100      sgd  0.803493      0.707030  5.636663e-01         0.7061   \n",
       "14   0.0100      sgd  0.828657      0.740667  4.955358e-01         0.7372   \n",
       "15   0.0100      sgd  0.731373      0.692485  7.939504e-01         0.6904   \n",
       "\n",
       "       test_loss  \n",
       "0   2.224292e-02  \n",
       "1   1.457398e+01  \n",
       "2   1.449017e+01  \n",
       "3   8.661988e-01  \n",
       "4   1.450629e+01  \n",
       "5   1.450629e+01  \n",
       "6   9.896639e-01  \n",
       "7   1.420535e+00  \n",
       "8   2.303477e+00  \n",
       "9   1.450629e+01  \n",
       "10  2.302689e+00  \n",
       "11  1.192093e-07  \n",
       "12  8.386861e-01  \n",
       "13  8.835897e-01  \n",
       "14  8.148351e-01  \n",
       "15  9.208996e-01  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['model_name','activation','dataset','dropout_rate' ,'epochs','input_dims','lr_rate', 'opt','accuracy','val_accuracy','loss','test_accuracy','test_loss']]"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}