Iris_dataset_training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d69dde6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from tensorflow import keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9f18d85e",
   "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>Id</th>\n",
       "      <th>SepalLengthCm</th>\n",
       "      <th>SepalWidthCm</th>\n",
       "      <th>PetalLengthCm</th>\n",
       "      <th>PetalWidthCm</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  SepalLengthCm  SepalWidthCm  PetalLengthCm  PetalWidthCm      Species\n",
       "0   1            5.1           3.5            1.4           0.2  Iris-setosa\n",
       "1   2            4.9           3.0            1.4           0.2  Iris-setosa\n",
       "2   3            4.7           3.2            1.3           0.2  Iris-setosa\n",
       "3   4            4.6           3.1            1.5           0.2  Iris-setosa\n",
       "4   5            5.0           3.6            1.4           0.2  Iris-setosa"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('Irisdataset.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "507fe029",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Iris-setosa        50\n",
       "Iris-versicolor    50\n",
       "Iris-virginica     50\n",
       "Name: Species, dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Species'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "57c3a3f7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 150 entries, 0 to 149\n",
      "Data columns (total 6 columns):\n",
      " #   Column         Non-Null Count  Dtype  \n",
      "---  ------         --------------  -----  \n",
      " 0   Id             150 non-null    int64  \n",
      " 1   SepalLengthCm  150 non-null    float64\n",
      " 2   SepalWidthCm   150 non-null    float64\n",
      " 3   PetalLengthCm  150 non-null    float64\n",
      " 4   PetalWidthCm   150 non-null    float64\n",
      " 5   Species        150 non-null    object \n",
      "dtypes: float64(4), int64(1), object(1)\n",
      "memory usage: 7.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bb6ec125",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Id               0\n",
       "SepalLengthCm    0\n",
       "SepalWidthCm     0\n",
       "PetalLengthCm    0\n",
       "PetalWidthCm     0\n",
       "Species          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8297b332",
   "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>Id</th>\n",
       "      <th>SepalLengthCm</th>\n",
       "      <th>SepalWidthCm</th>\n",
       "      <th>PetalLengthCm</th>\n",
       "      <th>PetalWidthCm</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  SepalLengthCm  SepalWidthCm  PetalLengthCm  PetalWidthCm  Species\n",
       "0   1            5.1           3.5            1.4           0.2        0\n",
       "1   2            4.9           3.0            1.4           0.2        0\n",
       "2   3            4.7           3.2            1.3           0.2        0\n",
       "3   4            4.6           3.1            1.5           0.2        0\n",
       "4   5            5.0           3.6            1.4           0.2        0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "df['Species'] = le.fit_transform(df['Species'])\n",
    "df.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "96d29f22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Iris-setosa' 'Iris-versicolor' 'Iris-virginica']\n"
     ]
    }
   ],
   "source": [
    "species_name = le.classes_\n",
    "print(species_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "19daea2a",
   "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>SepalLengthCm</th>\n",
       "      <th>SepalWidthCm</th>\n",
       "      <th>PetalLengthCm</th>\n",
       "      <th>PetalWidthCm</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   SepalLengthCm  SepalWidthCm  PetalLengthCm  PetalWidthCm\n",
       "0            5.1           3.5            1.4           0.2\n",
       "1            4.9           3.0            1.4           0.2\n",
       "2            4.7           3.2            1.3           0.2"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.drop(columns=['Id', 'Species'])\n",
    "y = df['Species']\n",
    "X.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fdf5dc5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    0\n",
      "1    0\n",
      "2    0\n",
      "3    0\n",
      "4    0\n",
      "Name: Species, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(y[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6beb56e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.3, shuffle=True, random_state=69)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "376b2667",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(105, 4)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "7091c208",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    SepalLengthCm  SepalWidthCm  PetalLengthCm  PetalWidthCm\n",
      "57            4.9           2.4            3.3           1.0\n",
      "[[-1.14630942 -1.44374444 -0.25650655 -0.258263  ]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "sc = StandardScaler()\n",
    "print(x_train[:1])\n",
    "x_train = sc.fit_transform(x_train)\n",
    "x_test = sc.transform(x_test)\n",
    "print(x_train[:1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "23b8b0cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]]\n"
     ]
    }
   ],
   "source": [
    "y_train = keras.utils.to_categorical(y_train, num_classes=3)\n",
    "print(y_train[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6b203733",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "26e74628",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-05-11 04:56:58.936582: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2023-05-11 04:56:58.937420: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(units=32, activation='relu', input_shape=(x_train.shape[-1], )))\n",
    "model.add(Dense(units=32, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(units=3, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ce7d8a2b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 32)                160       \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 32)                1056      \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 32)                0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 3)                 99        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,315\n",
      "Trainable params: 1,315\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.compile(optimizer='Adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "38faf8df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-05-11 04:59:30.836037: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n",
      "2023-05-11 04:59:31.051446: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 - 0s - loss: 1.1700 - accuracy: 0.3810 - 404ms/epoch - 101ms/step\n",
      "Epoch 2/100\n",
      "4/4 - 0s - loss: 1.1580 - accuracy: 0.3714 - 41ms/epoch - 10ms/step\n",
      "Epoch 3/100\n",
      "4/4 - 0s - loss: 1.1554 - accuracy: 0.3429 - 42ms/epoch - 10ms/step\n",
      "Epoch 4/100\n",
      "4/4 - 0s - loss: 1.0570 - accuracy: 0.4762 - 35ms/epoch - 9ms/step\n",
      "Epoch 5/100\n",
      "4/4 - 0s - loss: 1.0030 - accuracy: 0.5333 - 37ms/epoch - 9ms/step\n",
      "Epoch 6/100\n",
      "4/4 - 0s - loss: 0.9586 - accuracy: 0.5238 - 38ms/epoch - 9ms/step\n",
      "Epoch 7/100\n",
      "4/4 - 0s - loss: 0.9080 - accuracy: 0.5714 - 39ms/epoch - 10ms/step\n",
      "Epoch 8/100\n",
      "4/4 - 0s - loss: 0.8406 - accuracy: 0.6190 - 38ms/epoch - 9ms/step\n",
      "Epoch 9/100\n",
      "4/4 - 0s - loss: 0.8503 - accuracy: 0.6571 - 38ms/epoch - 10ms/step\n",
      "Epoch 10/100\n",
      "4/4 - 0s - loss: 0.8130 - accuracy: 0.6571 - 38ms/epoch - 10ms/step\n",
      "Epoch 11/100\n",
      "4/4 - 0s - loss: 0.7529 - accuracy: 0.7333 - 39ms/epoch - 10ms/step\n",
      "Epoch 12/100\n",
      "4/4 - 0s - loss: 0.7364 - accuracy: 0.7238 - 36ms/epoch - 9ms/step\n",
      "Epoch 13/100\n",
      "4/4 - 0s - loss: 0.6500 - accuracy: 0.7905 - 39ms/epoch - 10ms/step\n",
      "Epoch 14/100\n",
      "4/4 - 0s - loss: 0.6467 - accuracy: 0.7524 - 43ms/epoch - 11ms/step\n",
      "Epoch 15/100\n",
      "4/4 - 0s - loss: 0.6397 - accuracy: 0.8000 - 41ms/epoch - 10ms/step\n",
      "Epoch 16/100\n",
      "4/4 - 0s - loss: 0.6348 - accuracy: 0.8000 - 59ms/epoch - 15ms/step\n",
      "Epoch 17/100\n",
      "4/4 - 0s - loss: 0.5651 - accuracy: 0.8000 - 39ms/epoch - 10ms/step\n",
      "Epoch 18/100\n",
      "4/4 - 0s - loss: 0.5705 - accuracy: 0.7619 - 41ms/epoch - 10ms/step\n",
      "Epoch 19/100\n",
      "4/4 - 0s - loss: 0.5456 - accuracy: 0.8000 - 40ms/epoch - 10ms/step\n",
      "Epoch 20/100\n",
      "4/4 - 0s - loss: 0.5371 - accuracy: 0.8095 - 42ms/epoch - 10ms/step\n",
      "Epoch 21/100\n",
      "4/4 - 0s - loss: 0.5407 - accuracy: 0.8571 - 39ms/epoch - 10ms/step\n",
      "Epoch 22/100\n",
      "4/4 - 0s - loss: 0.5193 - accuracy: 0.7619 - 41ms/epoch - 10ms/step\n",
      "Epoch 23/100\n",
      "4/4 - 0s - loss: 0.4768 - accuracy: 0.8381 - 38ms/epoch - 10ms/step\n",
      "Epoch 24/100\n",
      "4/4 - 0s - loss: 0.4880 - accuracy: 0.8476 - 40ms/epoch - 10ms/step\n",
      "Epoch 25/100\n",
      "4/4 - 0s - loss: 0.4531 - accuracy: 0.8286 - 39ms/epoch - 10ms/step\n",
      "Epoch 26/100\n",
      "4/4 - 0s - loss: 0.4400 - accuracy: 0.8286 - 35ms/epoch - 9ms/step\n",
      "Epoch 27/100\n",
      "4/4 - 0s - loss: 0.4360 - accuracy: 0.8286 - 38ms/epoch - 10ms/step\n",
      "Epoch 28/100\n",
      "4/4 - 0s - loss: 0.4411 - accuracy: 0.8571 - 38ms/epoch - 9ms/step\n",
      "Epoch 29/100\n",
      "4/4 - 0s - loss: 0.4404 - accuracy: 0.8095 - 37ms/epoch - 9ms/step\n",
      "Epoch 30/100\n",
      "4/4 - 0s - loss: 0.3695 - accuracy: 0.8667 - 37ms/epoch - 9ms/step\n",
      "Epoch 31/100\n",
      "4/4 - 0s - loss: 0.4026 - accuracy: 0.8190 - 37ms/epoch - 9ms/step\n",
      "Epoch 32/100\n",
      "4/4 - 0s - loss: 0.3770 - accuracy: 0.8381 - 37ms/epoch - 9ms/step\n",
      "Epoch 33/100\n",
      "4/4 - 0s - loss: 0.3883 - accuracy: 0.8381 - 37ms/epoch - 9ms/step\n",
      "Epoch 34/100\n",
      "4/4 - 0s - loss: 0.3661 - accuracy: 0.8190 - 34ms/epoch - 9ms/step\n",
      "Epoch 35/100\n",
      "4/4 - 0s - loss: 0.3877 - accuracy: 0.8190 - 38ms/epoch - 10ms/step\n",
      "Epoch 36/100\n",
      "4/4 - 0s - loss: 0.3610 - accuracy: 0.8381 - 35ms/epoch - 9ms/step\n",
      "Epoch 37/100\n",
      "4/4 - 0s - loss: 0.3523 - accuracy: 0.8571 - 39ms/epoch - 10ms/step\n",
      "Epoch 38/100\n",
      "4/4 - 0s - loss: 0.3875 - accuracy: 0.8762 - 35ms/epoch - 9ms/step\n",
      "Epoch 39/100\n",
      "4/4 - 0s - loss: 0.3388 - accuracy: 0.8571 - 35ms/epoch - 9ms/step\n",
      "Epoch 40/100\n",
      "4/4 - 0s - loss: 0.3633 - accuracy: 0.8571 - 39ms/epoch - 10ms/step\n",
      "Epoch 41/100\n",
      "4/4 - 0s - loss: 0.3528 - accuracy: 0.8190 - 41ms/epoch - 10ms/step\n",
      "Epoch 42/100\n",
      "4/4 - 0s - loss: 0.3202 - accuracy: 0.8571 - 39ms/epoch - 10ms/step\n",
      "Epoch 43/100\n",
      "4/4 - 0s - loss: 0.3443 - accuracy: 0.8476 - 38ms/epoch - 9ms/step\n",
      "Epoch 44/100\n",
      "4/4 - 0s - loss: 0.3172 - accuracy: 0.8762 - 39ms/epoch - 10ms/step\n",
      "Epoch 45/100\n",
      "4/4 - 0s - loss: 0.3104 - accuracy: 0.8952 - 40ms/epoch - 10ms/step\n",
      "Epoch 46/100\n",
      "4/4 - 0s - loss: 0.3664 - accuracy: 0.8286 - 39ms/epoch - 10ms/step\n",
      "Epoch 47/100\n",
      "4/4 - 0s - loss: 0.3279 - accuracy: 0.8952 - 36ms/epoch - 9ms/step\n",
      "Epoch 48/100\n",
      "4/4 - 0s - loss: 0.3173 - accuracy: 0.8857 - 40ms/epoch - 10ms/step\n",
      "Epoch 49/100\n",
      "4/4 - 0s - loss: 0.3297 - accuracy: 0.8762 - 36ms/epoch - 9ms/step\n",
      "Epoch 50/100\n",
      "4/4 - 0s - loss: 0.3438 - accuracy: 0.8190 - 35ms/epoch - 9ms/step\n",
      "Epoch 51/100\n",
      "4/4 - 0s - loss: 0.2985 - accuracy: 0.8571 - 37ms/epoch - 9ms/step\n",
      "Epoch 52/100\n",
      "4/4 - 0s - loss: 0.3051 - accuracy: 0.8762 - 37ms/epoch - 9ms/step\n",
      "Epoch 53/100\n",
      "4/4 - 0s - loss: 0.2933 - accuracy: 0.9048 - 38ms/epoch - 10ms/step\n",
      "Epoch 54/100\n",
      "4/4 - 0s - loss: 0.3092 - accuracy: 0.8857 - 38ms/epoch - 9ms/step\n",
      "Epoch 55/100\n",
      "4/4 - 0s - loss: 0.3524 - accuracy: 0.8286 - 37ms/epoch - 9ms/step\n",
      "Epoch 56/100\n",
      "4/4 - 0s - loss: 0.2929 - accuracy: 0.8952 - 35ms/epoch - 9ms/step\n",
      "Epoch 57/100\n",
      "4/4 - 0s - loss: 0.2805 - accuracy: 0.8762 - 34ms/epoch - 8ms/step\n",
      "Epoch 58/100\n",
      "4/4 - 0s - loss: 0.3129 - accuracy: 0.8762 - 37ms/epoch - 9ms/step\n",
      "Epoch 59/100\n",
      "4/4 - 0s - loss: 0.2855 - accuracy: 0.8762 - 36ms/epoch - 9ms/step\n",
      "Epoch 60/100\n",
      "4/4 - 0s - loss: 0.2826 - accuracy: 0.8857 - 39ms/epoch - 10ms/step\n",
      "Epoch 61/100\n",
      "4/4 - 0s - loss: 0.2457 - accuracy: 0.9524 - 36ms/epoch - 9ms/step\n",
      "Epoch 62/100\n",
      "4/4 - 0s - loss: 0.2495 - accuracy: 0.9048 - 38ms/epoch - 9ms/step\n",
      "Epoch 63/100\n",
      "4/4 - 0s - loss: 0.2380 - accuracy: 0.9333 - 35ms/epoch - 9ms/step\n",
      "Epoch 64/100\n",
      "4/4 - 0s - loss: 0.2578 - accuracy: 0.9333 - 38ms/epoch - 9ms/step\n",
      "Epoch 65/100\n",
      "4/4 - 0s - loss: 0.2513 - accuracy: 0.9619 - 42ms/epoch - 10ms/step\n",
      "Epoch 66/100\n",
      "4/4 - 0s - loss: 0.2466 - accuracy: 0.9238 - 44ms/epoch - 11ms/step\n",
      "Epoch 67/100\n",
      "4/4 - 0s - loss: 0.2682 - accuracy: 0.9048 - 40ms/epoch - 10ms/step\n",
      "Epoch 68/100\n",
      "4/4 - 0s - loss: 0.2635 - accuracy: 0.9333 - 42ms/epoch - 10ms/step\n",
      "Epoch 69/100\n",
      "4/4 - 0s - loss: 0.2687 - accuracy: 0.9238 - 39ms/epoch - 10ms/step\n",
      "Epoch 70/100\n",
      "4/4 - 0s - loss: 0.2571 - accuracy: 0.8952 - 41ms/epoch - 10ms/step\n",
      "Epoch 71/100\n",
      "4/4 - 0s - loss: 0.2160 - accuracy: 0.9238 - 40ms/epoch - 10ms/step\n",
      "Epoch 72/100\n",
      "4/4 - 0s - loss: 0.2291 - accuracy: 0.9143 - 38ms/epoch - 9ms/step\n",
      "Epoch 73/100\n",
      "4/4 - 0s - loss: 0.2485 - accuracy: 0.9143 - 37ms/epoch - 9ms/step\n",
      "Epoch 74/100\n",
      "4/4 - 0s - loss: 0.2437 - accuracy: 0.9143 - 38ms/epoch - 9ms/step\n",
      "Epoch 75/100\n",
      "4/4 - 0s - loss: 0.2336 - accuracy: 0.9333 - 40ms/epoch - 10ms/step\n",
      "Epoch 76/100\n",
      "4/4 - 0s - loss: 0.2230 - accuracy: 0.9143 - 39ms/epoch - 10ms/step\n",
      "Epoch 77/100\n",
      "4/4 - 0s - loss: 0.2048 - accuracy: 0.9143 - 37ms/epoch - 9ms/step\n",
      "Epoch 78/100\n",
      "4/4 - 0s - loss: 0.2232 - accuracy: 0.9333 - 38ms/epoch - 10ms/step\n",
      "Epoch 79/100\n",
      "4/4 - 0s - loss: 0.2209 - accuracy: 0.9333 - 38ms/epoch - 9ms/step\n",
      "Epoch 80/100\n",
      "4/4 - 0s - loss: 0.2085 - accuracy: 0.9333 - 39ms/epoch - 10ms/step\n",
      "Epoch 81/100\n",
      "4/4 - 0s - loss: 0.1993 - accuracy: 0.9524 - 38ms/epoch - 10ms/step\n",
      "Epoch 82/100\n",
      "4/4 - 0s - loss: 0.1857 - accuracy: 0.9619 - 37ms/epoch - 9ms/step\n",
      "Epoch 83/100\n",
      "4/4 - 0s - loss: 0.1863 - accuracy: 0.9524 - 39ms/epoch - 10ms/step\n",
      "Epoch 84/100\n",
      "4/4 - 0s - loss: 0.1832 - accuracy: 0.9619 - 38ms/epoch - 10ms/step\n",
      "Epoch 85/100\n",
      "4/4 - 0s - loss: 0.1603 - accuracy: 0.9429 - 40ms/epoch - 10ms/step\n",
      "Epoch 86/100\n",
      "4/4 - 0s - loss: 0.1791 - accuracy: 0.9810 - 38ms/epoch - 9ms/step\n",
      "Epoch 87/100\n",
      "4/4 - 0s - loss: 0.1645 - accuracy: 0.9524 - 35ms/epoch - 9ms/step\n",
      "Epoch 88/100\n",
      "4/4 - 0s - loss: 0.2116 - accuracy: 0.9238 - 38ms/epoch - 10ms/step\n",
      "Epoch 89/100\n",
      "4/4 - 0s - loss: 0.1708 - accuracy: 0.9429 - 43ms/epoch - 11ms/step\n",
      "Epoch 90/100\n",
      "4/4 - 0s - loss: 0.1754 - accuracy: 0.9714 - 42ms/epoch - 10ms/step\n",
      "Epoch 91/100\n",
      "4/4 - 0s - loss: 0.1552 - accuracy: 0.9524 - 40ms/epoch - 10ms/step\n",
      "Epoch 92/100\n",
      "4/4 - 0s - loss: 0.1490 - accuracy: 0.9619 - 38ms/epoch - 9ms/step\n",
      "Epoch 93/100\n",
      "4/4 - 0s - loss: 0.1898 - accuracy: 0.9238 - 38ms/epoch - 10ms/step\n",
      "Epoch 94/100\n",
      "4/4 - 0s - loss: 0.1488 - accuracy: 0.9714 - 38ms/epoch - 9ms/step\n",
      "Epoch 95/100\n",
      "4/4 - 0s - loss: 0.1877 - accuracy: 0.9429 - 39ms/epoch - 10ms/step\n",
      "Epoch 96/100\n",
      "4/4 - 0s - loss: 0.1538 - accuracy: 0.9619 - 39ms/epoch - 10ms/step\n",
      "Epoch 97/100\n",
      "4/4 - 0s - loss: 0.1857 - accuracy: 0.9429 - 37ms/epoch - 9ms/step\n",
      "Epoch 98/100\n",
      "4/4 - 0s - loss: 0.1763 - accuracy: 0.9333 - 38ms/epoch - 9ms/step\n",
      "Epoch 99/100\n",
      "4/4 - 0s - loss: 0.1735 - accuracy: 0.9524 - 50ms/epoch - 12ms/step\n",
      "Epoch 100/100\n",
      "4/4 - 0s - loss: 0.1362 - accuracy: 0.9619 - 37ms/epoch - 9ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x16ccb4b50>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x_train, y_train, epochs=100, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "abbef3be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 [==============================] - 0s 18ms/step\n",
      "[[9.9946219e-01 4.8392054e-04 5.3865653e-05]\n",
      " [3.4496377e-03 9.3944883e-01 5.7101529e-02]\n",
      " [9.9754477e-01 2.3062190e-03 1.4900243e-04]\n",
      " [4.0955580e-05 1.3634584e-02 9.8632449e-01]\n",
      " [9.9670237e-01 3.0157326e-03 2.8197953e-04]]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-05-11 05:00:12.708082: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    }
   ],
   "source": [
    "prediction = model.predict(x_test)\n",
    "print(prediction[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "e76053da",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 0 2 0]\n"
     ]
    }
   ],
   "source": [
    "prediction = np.argmax(prediction, axis=-1)\n",
    "print(prediction[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d73db658",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4      0\n",
      "58     1\n",
      "20     0\n",
      "114    2\n",
      "31     0\n",
      "Name: Species, dtype: int64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/75/fyf8yn3s0h3chmlrq431v9kr0000gn/T/ipykernel_24405/4125655240.py:1: FutureWarning: The behavior of `series[i:j]` with an integer-dtype index is deprecated. In a future version, this will be treated as *label-based* indexing, consistent with e.g. `series[i]` lookups. To retain the old behavior, use `series.iloc[i:j]`. To get the future behavior, use `series.loc[i:j]`.\n",
      "  print(y_test[:5])\n"
     ]
    }
   ],
   "source": [
    "print(y_test[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "74b91d1c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9777777777777777\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, confusion_matrix\n",
    "print(accuracy_score(y_test, prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "489c52d3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[16  0  0]\n",
      " [ 0 12  0]\n",
      " [ 0  1 16]]\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, prediction)\n",
    "print(cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "9cd87323",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(50.72222222222221, 0.5, 'actual')"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.heatmap(cm, annot=True, fmt='d', cmap='Reds', xticklabels=species_name, yticklabels=species_name)\n",
    "ax.set_title('confusion matrix for irsis dataset prediction')\n",
    "ax.set_xlabel('prediction', fontsize=14)\n",
    "ax.set_ylabel('actual', fontsize=14)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2582fba4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}