Iris Classification Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Importing necessary libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.python.keras import utils\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sn\n",
    "import missingno as msno\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "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>SepalLength</th>\n",
       "      <th>SepalWidth</th>\n",
       "      <th>PetalLength</th>\n",
       "      <th>PetalWidth</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6.4</td>\n",
       "      <td>2.8</td>\n",
       "      <td>5.6</td>\n",
       "      <td>2.2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.9</td>\n",
       "      <td>2.5</td>\n",
       "      <td>4.5</td>\n",
       "      <td>1.7</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.7</td>\n",
       "      <td>3.8</td>\n",
       "      <td>1.7</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   SepalLength  SepalWidth  PetalLength  PetalWidth  Species\n",
       "0          6.4         2.8          5.6         2.2        2\n",
       "1          5.0         2.3          3.3         1.0        1\n",
       "2          4.9         2.5          4.5         1.7        2\n",
       "3          4.9         3.1          1.5         0.1        0\n",
       "4          5.7         3.8          1.7         0.3        0"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "COLUMN_NAMES = [\n",
    "        'SepalLength', \n",
    "        'SepalWidth',\n",
    "        'PetalLength', \n",
    "        'PetalWidth', \n",
    "        'Species'\n",
    "        ]\n",
    "\n",
    "data = pd.read_csv('iris_data.csv', names=COLUMN_NAMES, header=0)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Types Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SepalLength    float64\n",
       "SepalWidth     float64\n",
       "PetalLength    float64\n",
       "PetalWidth     float64\n",
       "Species          int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Handle Categorical Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SepalLength     float64\n",
       "SepalWidth      float64\n",
       "PetalLength     float64\n",
       "PetalWidth      float64\n",
       "Species        category\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Species'] = data['Species'].astype(\"category\")\n",
    "data.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution = sn.countplot(x=\"Species\", data=data)\n",
    "plt.show(distribution)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Missing Information"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SepalLength    0\n",
      "SepalWidth     0\n",
      "PetalLength    0\n",
      "PetalWidth     0\n",
      "Species        0\n",
      "dtype: int64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10dbd6dd8>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(data.isnull().sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Correlation Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10dc76668>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "corrMatt = data[[\"SepalLength\",\"SepalWidth\",\"PetalLength\",\"PetalWidth\",\"Species\"]].corr()\n",
    "mask = np.array(corrMatt)\n",
    "mask[np.tril_indices_from(mask)] = False\n",
    "fig,ax= plt.subplots()\n",
    "fig.set_size_inches(20,10)\n",
    "sn.heatmap(corrMatt, mask=mask,vmax=.8, square=True,annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Implementation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split Data to Test and Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input training size:  105\n",
      "Output training size:  105\n",
      "Input testing size:  45\n",
      "Output testing size:  45\n"
     ]
    }
   ],
   "source": [
    "output_data = data[\"Species\"]\n",
    "input_data = data.drop(\"Species\",axis=1)\n",
    "X_train, X_test, y_train, y_test = train_test_split(input_data, output_data, test_size=0.3, random_state=42)\n",
    "\n",
    "y_train = utils.to_categorical(y_train) \n",
    "y_test = utils.to_categorical(y_test)\n",
    "\n",
    "print(\"Input training size: \", X_train.shape[0])\n",
    "print(\"Output training size: \", y_train.shape[0])\n",
    "print(\"Input testing size: \", X_test.shape[0])\n",
    "print(\"Output testing size: \", y_test.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "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>SepalLength</th>\n",
       "      <th>SepalWidth</th>\n",
       "      <th>PetalLength</th>\n",
       "      <th>PetalWidth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>6.1</td>\n",
       "      <td>2.9</td>\n",
       "      <td>4.7</td>\n",
       "      <td>1.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>5.5</td>\n",
       "      <td>2.3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>4.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.9</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>4.8</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     SepalLength  SepalWidth  PetalLength  PetalWidth\n",
       "81           6.1         2.9          4.7         1.4\n",
       "133          5.5         2.3          4.0         1.3\n",
       "137          4.8         3.4          1.9         0.2\n",
       "75           4.9         3.1          1.5         0.1\n",
       "109          4.8         3.1          1.6         0.2"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating & compiling a model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(10, input_dim=4, activation='relu'))\n",
    "model.add(Dense(10, activation='relu'))\n",
    "model.add(Dense(3, activation='softmax'))\n",
    "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training a model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/300\n",
      "105/105 [==============================] - 0s 2ms/step - loss: 1.6388 - acc: 0.2476\n",
      "Epoch 2/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 1.3490 - acc: 0.0000e+00\n",
      "Epoch 3/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 1.2546 - acc: 0.1905\n",
      "Epoch 4/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 1.2030 - acc: 0.3333\n",
      "Epoch 5/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 1.1657 - acc: 0.3333\n",
      "Epoch 6/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 1.1281 - acc: 0.3333\n",
      "Epoch 7/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 1.0956 - acc: 0.3333\n",
      "Epoch 8/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 1.0652 - acc: 0.3333\n",
      "Epoch 9/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 1.0336 - acc: 0.3333\n",
      "Epoch 10/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 1.0076 - acc: 0.3333\n",
      "Epoch 11/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.9860 - acc: 0.3333\n",
      "Epoch 12/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.9649 - acc: 0.3333\n",
      "Epoch 13/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.9424 - acc: 0.3429\n",
      "Epoch 14/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.9169 - acc: 0.5714\n",
      "Epoch 15/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.8914 - acc: 0.7048\n",
      "Epoch 16/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.8545 - acc: 0.7048\n",
      "Epoch 17/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.8157 - acc: 0.7048\n",
      "Epoch 18/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.7853 - acc: 0.7048\n",
      "Epoch 19/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.7477 - acc: 0.7524\n",
      "Epoch 20/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.7129 - acc: 0.8667\n",
      "Epoch 21/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.6841 - acc: 0.7333\n",
      "Epoch 22/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.6457 - acc: 0.8095\n",
      "Epoch 23/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.6155 - acc: 0.8762\n",
      "Epoch 24/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.5854 - acc: 0.8952\n",
      "Epoch 25/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.5542 - acc: 0.8952\n",
      "Epoch 26/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.5238 - acc: 0.9143\n",
      "Epoch 27/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.4937 - acc: 0.9333\n",
      "Epoch 28/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.4679 - acc: 0.8952\n",
      "Epoch 29/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.4475 - acc: 0.9429\n",
      "Epoch 30/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.4257 - acc: 0.9619\n",
      "Epoch 31/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.4056 - acc: 0.9333\n",
      "Epoch 32/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.3881 - acc: 0.9714\n",
      "Epoch 33/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.3738 - acc: 0.9714\n",
      "Epoch 34/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.3602 - acc: 0.9714\n",
      "Epoch 35/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.3453 - acc: 0.9714\n",
      "Epoch 36/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.3332 - acc: 0.9714\n",
      "Epoch 37/300\n",
      "105/105 [==============================] - 0s 124us/step - loss: 0.3209 - acc: 0.9714\n",
      "Epoch 38/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.3105 - acc: 0.9714\n",
      "Epoch 39/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2989 - acc: 0.9714\n",
      "Epoch 40/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.2892 - acc: 0.9714\n",
      "Epoch 41/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.2818 - acc: 0.9714\n",
      "Epoch 42/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2719 - acc: 0.9714\n",
      "Epoch 43/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.2632 - acc: 0.9714\n",
      "Epoch 44/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2556 - acc: 0.9714\n",
      "Epoch 45/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.2492 - acc: 0.9714\n",
      "Epoch 46/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.2402 - acc: 0.9714\n",
      "Epoch 47/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.2352 - acc: 0.9810\n",
      "Epoch 48/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2287 - acc: 0.9810\n",
      "Epoch 49/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.2227 - acc: 0.9714\n",
      "Epoch 50/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2157 - acc: 0.9714\n",
      "Epoch 51/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.2099 - acc: 0.9810\n",
      "Epoch 52/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.2049 - acc: 0.9810\n",
      "Epoch 53/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1997 - acc: 0.9714\n",
      "Epoch 54/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1975 - acc: 0.9810\n",
      "Epoch 55/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1903 - acc: 0.9714\n",
      "Epoch 56/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.1858 - acc: 0.9810\n",
      "Epoch 57/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1830 - acc: 0.9810\n",
      "Epoch 58/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1817 - acc: 0.9810\n",
      "Epoch 59/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.1739 - acc: 0.9810\n",
      "Epoch 60/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1722 - acc: 0.9810\n",
      "Epoch 61/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1705 - acc: 0.9714\n",
      "Epoch 62/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.1642 - acc: 0.9810\n",
      "Epoch 63/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.1607 - acc: 0.9810\n",
      "Epoch 64/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1567 - acc: 0.9810\n",
      "Epoch 65/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1541 - acc: 0.9810\n",
      "Epoch 66/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1522 - acc: 0.9810\n",
      "Epoch 67/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1493 - acc: 0.9810\n",
      "Epoch 68/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1503 - acc: 0.9810\n",
      "Epoch 69/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1454 - acc: 0.9714\n",
      "Epoch 70/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1459 - acc: 0.9810\n",
      "Epoch 71/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1423 - acc: 0.9810\n",
      "Epoch 72/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1362 - acc: 0.9810\n",
      "Epoch 73/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1340 - acc: 0.9810\n",
      "Epoch 74/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1320 - acc: 0.9810\n",
      "Epoch 75/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.1304 - acc: 0.9810\n",
      "Epoch 76/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1305 - acc: 0.9810\n",
      "Epoch 77/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1272 - acc: 0.9810\n",
      "Epoch 78/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1283 - acc: 0.9810\n",
      "Epoch 79/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1219 - acc: 0.9810\n",
      "Epoch 80/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.1224 - acc: 0.9810\n",
      "Epoch 81/300\n",
      "105/105 [==============================] - 0s 110us/step - loss: 0.1238 - acc: 0.9810\n",
      "Epoch 82/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1181 - acc: 0.9810\n",
      "Epoch 83/300\n",
      "105/105 [==============================] - 0s 81us/step - loss: 0.1172 - acc: 0.9810\n",
      "Epoch 84/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1149 - acc: 0.9810\n",
      "Epoch 85/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1140 - acc: 0.9810\n",
      "Epoch 86/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1129 - acc: 0.9810\n",
      "Epoch 87/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1108 - acc: 0.9810\n",
      "Epoch 88/300\n",
      "105/105 [==============================] - 0s 81us/step - loss: 0.1143 - acc: 0.9905\n",
      "Epoch 89/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1095 - acc: 0.9810\n",
      "Epoch 90/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.1075 - acc: 0.9810\n",
      "Epoch 91/300\n",
      "105/105 [==============================] - 0s 114us/step - loss: 0.1061 - acc: 0.9810\n",
      "Epoch 92/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1043 - acc: 0.9810\n",
      "Epoch 93/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1083 - acc: 0.9714\n",
      "Epoch 94/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1022 - acc: 0.9810\n",
      "Epoch 95/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1028 - acc: 0.9810\n",
      "Epoch 96/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0998 - acc: 0.9810\n",
      "Epoch 97/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.1006 - acc: 0.9810\n",
      "Epoch 98/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0986 - acc: 0.9810\n",
      "Epoch 99/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0978 - acc: 0.9810\n",
      "Epoch 100/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0973 - acc: 0.9810\n",
      "Epoch 101/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0954 - acc: 0.9810\n",
      "Epoch 102/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0949 - acc: 0.9810\n",
      "Epoch 103/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0983 - acc: 0.9810\n",
      "Epoch 104/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.1001 - acc: 0.9810\n",
      "Epoch 105/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0931 - acc: 0.9905\n",
      "Epoch 106/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0921 - acc: 0.9810\n",
      "Epoch 107/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0928 - acc: 0.9810\n",
      "Epoch 108/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0905 - acc: 0.9810\n",
      "Epoch 109/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0937 - acc: 0.9810\n",
      "Epoch 110/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0903 - acc: 0.9810\n",
      "Epoch 111/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0884 - acc: 0.9810\n",
      "Epoch 112/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0872 - acc: 0.9810\n",
      "Epoch 113/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0880 - acc: 0.9810\n",
      "Epoch 114/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0861 - acc: 0.9810\n",
      "Epoch 115/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0883 - acc: 0.9810\n",
      "Epoch 116/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0851 - acc: 0.9810\n",
      "Epoch 117/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0856 - acc: 0.9810\n",
      "Epoch 118/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0852 - acc: 0.9905\n",
      "Epoch 119/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0878 - acc: 0.9714\n",
      "Epoch 120/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0863 - acc: 0.9810\n",
      "Epoch 121/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0861 - acc: 0.9810\n",
      "Epoch 122/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0821 - acc: 0.9810\n",
      "Epoch 123/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0819 - acc: 0.9810\n",
      "Epoch 124/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0832 - acc: 0.9810\n",
      "Epoch 125/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0807 - acc: 0.9810\n",
      "Epoch 126/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0805 - acc: 0.9810\n",
      "Epoch 127/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0785 - acc: 0.9810\n",
      "Epoch 128/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0788 - acc: 0.9810\n",
      "Epoch 129/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0788 - acc: 0.9810\n",
      "Epoch 130/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0780 - acc: 0.9810\n",
      "Epoch 131/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0782 - acc: 0.9810\n",
      "Epoch 132/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0782 - acc: 0.9905\n",
      "Epoch 133/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0768 - acc: 0.9810\n",
      "Epoch 134/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.0758 - acc: 0.9810\n",
      "Epoch 135/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0760 - acc: 0.9810\n",
      "Epoch 136/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0799 - acc: 0.9810\n",
      "Epoch 137/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0783 - acc: 0.9714\n",
      "Epoch 138/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0774 - acc: 0.9905\n",
      "Epoch 139/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0762 - acc: 0.9810\n",
      "Epoch 140/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0752 - acc: 0.9810\n",
      "Epoch 141/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0752 - acc: 0.9810\n",
      "Epoch 142/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0749 - acc: 0.9810\n",
      "Epoch 143/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0737 - acc: 0.9810\n",
      "Epoch 144/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0745 - acc: 0.9810\n",
      "Epoch 145/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0727 - acc: 0.9810\n",
      "Epoch 146/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0730 - acc: 0.9810\n",
      "Epoch 147/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0759 - acc: 0.9810\n",
      "Epoch 148/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0726 - acc: 0.9810\n",
      "Epoch 149/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0716 - acc: 0.9810\n",
      "Epoch 150/300\n",
      "105/105 [==============================] - 0s 124us/step - loss: 0.0721 - acc: 0.9810\n",
      "Epoch 151/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0713 - acc: 0.9905\n",
      "Epoch 152/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0735 - acc: 0.9810\n",
      "Epoch 153/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0721 - acc: 0.9810\n",
      "Epoch 154/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0708 - acc: 0.9810\n",
      "Epoch 155/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0723 - acc: 0.9810\n",
      "Epoch 156/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0705 - acc: 0.9810\n",
      "Epoch 157/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.0705 - acc: 0.9810\n",
      "Epoch 158/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0699 - acc: 0.9810\n",
      "Epoch 159/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0706 - acc: 0.9810\n",
      "Epoch 160/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0679 - acc: 0.9810\n",
      "Epoch 161/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0700 - acc: 0.9905\n",
      "Epoch 162/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0715 - acc: 0.9905\n",
      "Epoch 163/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0688 - acc: 0.9810\n",
      "Epoch 164/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0673 - acc: 0.9810\n",
      "Epoch 165/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0685 - acc: 0.9905\n",
      "Epoch 166/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0668 - acc: 0.9810\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 167/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0678 - acc: 0.9810\n",
      "Epoch 168/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0675 - acc: 0.9810\n",
      "Epoch 169/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0671 - acc: 0.9810\n",
      "Epoch 170/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0683 - acc: 0.9810\n",
      "Epoch 171/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0661 - acc: 0.9810\n",
      "Epoch 172/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0675 - acc: 0.9905\n",
      "Epoch 173/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0670 - acc: 0.9810\n",
      "Epoch 174/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0656 - acc: 0.9810\n",
      "Epoch 175/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0696 - acc: 0.9810\n",
      "Epoch 176/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0692 - acc: 0.9810\n",
      "Epoch 177/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0676 - acc: 0.9810\n",
      "Epoch 178/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0677 - acc: 0.9810\n",
      "Epoch 179/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0657 - acc: 0.9810\n",
      "Epoch 180/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0661 - acc: 0.9810\n",
      "Epoch 181/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0665 - acc: 0.9810\n",
      "Epoch 182/300\n",
      "105/105 [==============================] - 0s 110us/step - loss: 0.0653 - acc: 0.9810\n",
      "Epoch 183/300\n",
      "105/105 [==============================] - 0s 81us/step - loss: 0.0657 - acc: 0.9905\n",
      "Epoch 184/300\n",
      "105/105 [==============================] - 0s 100us/step - loss: 0.0708 - acc: 0.9810\n",
      "Epoch 185/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0663 - acc: 0.9810\n",
      "Epoch 186/300\n",
      "105/105 [==============================] - 0s 81us/step - loss: 0.0660 - acc: 0.9810\n",
      "Epoch 187/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0642 - acc: 0.9810\n",
      "Epoch 188/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0636 - acc: 0.9810\n",
      "Epoch 189/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0641 - acc: 0.9810\n",
      "Epoch 190/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0634 - acc: 0.9810\n",
      "Epoch 191/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0633 - acc: 0.9810\n",
      "Epoch 192/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0646 - acc: 0.9810\n",
      "Epoch 193/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0639 - acc: 0.9905\n",
      "Epoch 194/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0659 - acc: 0.9905\n",
      "Epoch 195/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0668 - acc: 0.9810\n",
      "Epoch 196/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0624 - acc: 0.9810\n",
      "Epoch 197/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0634 - acc: 0.9810\n",
      "Epoch 198/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0686 - acc: 0.9905\n",
      "Epoch 199/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0635 - acc: 0.9810\n",
      "Epoch 200/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0635 - acc: 0.9810\n",
      "Epoch 201/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0629 - acc: 0.9905\n",
      "Epoch 202/300\n",
      "105/105 [==============================] - 0s 114us/step - loss: 0.0632 - acc: 0.9905\n",
      "Epoch 203/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.0617 - acc: 0.9905\n",
      "Epoch 204/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0615 - acc: 0.9810\n",
      "Epoch 205/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.0619 - acc: 0.9810\n",
      "Epoch 206/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0612 - acc: 0.9810\n",
      "Epoch 207/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0679 - acc: 0.9905\n",
      "Epoch 208/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0596 - acc: 0.9905\n",
      "Epoch 209/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0626 - acc: 0.9810\n",
      "Epoch 210/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0620 - acc: 0.9810\n",
      "Epoch 211/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0630 - acc: 0.9810\n",
      "Epoch 212/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0603 - acc: 0.9810\n",
      "Epoch 213/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0609 - acc: 0.9810\n",
      "Epoch 214/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0622 - acc: 0.9810\n",
      "Epoch 215/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0600 - acc: 0.9905\n",
      "Epoch 216/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0618 - acc: 0.9810\n",
      "Epoch 217/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0605 - acc: 0.9810\n",
      "Epoch 218/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0609 - acc: 0.9905\n",
      "Epoch 219/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0609 - acc: 0.9905\n",
      "Epoch 220/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0658 - acc: 0.9810\n",
      "Epoch 221/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0614 - acc: 0.9905\n",
      "Epoch 222/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0604 - acc: 0.9810\n",
      "Epoch 223/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0660 - acc: 0.9810\n",
      "Epoch 224/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0661 - acc: 0.9810\n",
      "Epoch 225/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0610 - acc: 0.9810\n",
      "Epoch 226/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0593 - acc: 0.9810\n",
      "Epoch 227/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0606 - acc: 0.9810\n",
      "Epoch 228/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0646 - acc: 0.9810\n",
      "Epoch 229/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0584 - acc: 0.9905\n",
      "Epoch 230/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0592 - acc: 0.9810\n",
      "Epoch 231/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0612 - acc: 0.9810\n",
      "Epoch 232/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0595 - acc: 0.9810\n",
      "Epoch 233/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0611 - acc: 0.9905\n",
      "Epoch 234/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0576 - acc: 0.9905\n",
      "Epoch 235/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0641 - acc: 0.9810\n",
      "Epoch 236/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0582 - acc: 0.9810\n",
      "Epoch 237/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0592 - acc: 0.9905\n",
      "Epoch 238/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0590 - acc: 0.9810\n",
      "Epoch 239/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0586 - acc: 0.9810\n",
      "Epoch 240/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0586 - acc: 0.9810\n",
      "Epoch 241/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0574 - acc: 0.9905\n",
      "Epoch 242/300\n",
      "105/105 [==============================] - 0s 67us/step - loss: 0.0576 - acc: 0.9810\n",
      "Epoch 243/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0595 - acc: 0.9905\n",
      "Epoch 244/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0590 - acc: 0.9810\n",
      "Epoch 245/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0585 - acc: 0.9810\n",
      "Epoch 246/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0568 - acc: 0.9810\n",
      "Epoch 247/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0567 - acc: 0.9905\n",
      "Epoch 248/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0577 - acc: 0.9905\n",
      "Epoch 249/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0573 - acc: 0.9810\n",
      "Epoch 250/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0582 - acc: 0.9905\n",
      "Epoch 251/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0568 - acc: 0.9905\n",
      "Epoch 252/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0560 - acc: 0.9810\n",
      "Epoch 253/300\n",
      "105/105 [==============================] - 0s 76us/step - loss: 0.0583 - acc: 0.9810\n",
      "Epoch 254/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0611 - acc: 0.9810\n",
      "Epoch 255/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0581 - acc: 0.9905\n",
      "Epoch 256/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0571 - acc: 0.9810\n",
      "Epoch 257/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0582 - acc: 0.9905\n",
      "Epoch 258/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0570 - acc: 0.9810\n",
      "Epoch 259/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0558 - acc: 0.9810\n",
      "Epoch 260/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0557 - acc: 0.9905\n",
      "Epoch 261/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0577 - acc: 0.9810\n",
      "Epoch 262/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0562 - acc: 0.9810\n",
      "Epoch 263/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0554 - acc: 0.9905\n",
      "Epoch 264/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0559 - acc: 0.9905\n",
      "Epoch 265/300\n",
      "105/105 [==============================] - 0s 114us/step - loss: 0.0596 - acc: 0.9810\n",
      "Epoch 266/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0565 - acc: 0.9905\n",
      "Epoch 267/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0553 - acc: 0.9905\n",
      "Epoch 268/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0552 - acc: 0.9810\n",
      "Epoch 269/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0558 - acc: 0.9905\n",
      "Epoch 270/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0560 - acc: 0.9905\n",
      "Epoch 271/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0554 - acc: 0.9810\n",
      "Epoch 272/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0559 - acc: 0.9810\n",
      "Epoch 273/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0569 - acc: 0.9810\n",
      "Epoch 274/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0546 - acc: 0.9810\n",
      "Epoch 275/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0586 - acc: 0.9905\n",
      "Epoch 276/300\n",
      "105/105 [==============================] - 0s 114us/step - loss: 0.0545 - acc: 0.9905\n",
      "Epoch 277/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0554 - acc: 0.9810\n",
      "Epoch 278/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0544 - acc: 0.9810\n",
      "Epoch 279/300\n",
      "105/105 [==============================] - 0s 105us/step - loss: 0.0553 - acc: 0.9905\n",
      "Epoch 280/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0567 - acc: 0.9810\n",
      "Epoch 281/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0559 - acc: 0.9905\n",
      "Epoch 282/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0543 - acc: 0.9905\n",
      "Epoch 283/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0584 - acc: 0.9810\n",
      "Epoch 284/300\n",
      "105/105 [==============================] - 0s 96us/step - loss: 0.0578 - acc: 0.9905\n",
      "Epoch 285/300\n",
      "105/105 [==============================] - 0s 91us/step - loss: 0.0584 - acc: 0.9810\n",
      "Epoch 286/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0555 - acc: 0.9905\n",
      "Epoch 287/300\n",
      "105/105 [==============================] - 0s 100us/step - loss: 0.0541 - acc: 0.9905\n",
      "Epoch 288/300\n",
      "105/105 [==============================] - 0s 91us/step - loss: 0.0539 - acc: 0.9905\n",
      "Epoch 289/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0541 - acc: 0.9810\n",
      "Epoch 290/300\n",
      "105/105 [==============================] - 0s 91us/step - loss: 0.0552 - acc: 0.9905\n",
      "Epoch 291/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0541 - acc: 0.9905\n",
      "Epoch 292/300\n",
      "105/105 [==============================] - 0s 91us/step - loss: 0.0528 - acc: 0.9905\n",
      "Epoch 293/300\n",
      "105/105 [==============================] - 0s 91us/step - loss: 0.0568 - acc: 0.9810\n",
      "Epoch 294/300\n",
      "105/105 [==============================] - 0s 96us/step - loss: 0.0528 - acc: 0.9905\n",
      "Epoch 295/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0536 - acc: 0.9810\n",
      "Epoch 296/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0549 - acc: 0.9905\n",
      "Epoch 297/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0532 - acc: 0.9905\n",
      "Epoch 298/300\n",
      "105/105 [==============================] - 0s 95us/step - loss: 0.0529 - acc: 0.9905\n",
      "Epoch 299/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0528 - acc: 0.9905\n",
      "Epoch 300/300\n",
      "105/105 [==============================] - 0s 86us/step - loss: 0.0569 - acc: 0.9905\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x10dc61390>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs=300, batch_size=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "45/45 [==============================] - 0s 756us/step\n",
      "\n",
      "Accuracy: 95.56%\n"
     ]
    }
   ],
   "source": [
    "scores = model.evaluate(X_test, y_test)\n",
    "print(\"\\nAccuracy: %.2f%%\" % (scores[1]*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "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>IRIS1</th>\n",
       "      <th>IRIS2</th>\n",
       "      <th>IRIS3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>0.9977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0084</td>\n",
       "      <td>0.9916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.9989</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.9558</td>\n",
       "      <td>0.0436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0017</td>\n",
       "      <td>0.9958</td>\n",
       "      <td>0.0025</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    IRIS1   IRIS2   IRIS3\n",
       "0  0.0000  0.0023  0.9977\n",
       "1  0.0000  0.0084  0.9916\n",
       "2  0.9989  0.0011  0.0000\n",
       "3  0.0007  0.9558  0.0436\n",
       "4  0.0017  0.9958  0.0025"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction = model.predict(X_test)\n",
    "prediction1 = pd.DataFrame({'IRIS1':prediction[:,0],'IRIS2':prediction[:,1], 'IRIS3':prediction[:,2]})\n",
    "prediction1.round(decimals=4).head()"
   ]
  },
  {
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1    2\n",
       "0  0.0  0.0  1.0\n",
       "1  0.0  0.0  1.0\n",
       "2  1.0  0.0  0.0\n",
       "3  0.0  1.0  0.0\n",
       "4  0.0  1.0  0.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "real_values = pd.DataFrame(y_test)\n",
    "real_values.head()"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}