mobilePriceClassification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "mobilePriceClassification",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyN2kHGvC35RMPWlz138Vx9a",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Asto7/deep-learning/blob/main/mobilePriceClassification.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lEBaxXTC4aEs",
        "outputId": "de34a762-f9c6-4b34-da37-95e181ed5bac"
      },
      "source": [
        "!pip install kaggle\n"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.7/dist-packages (1.5.10)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.7/dist-packages (from kaggle) (2.8.1)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.7/dist-packages (from kaggle) (1.15.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.7/dist-packages (from kaggle) (4.41.1)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.7/dist-packages (from kaggle) (2020.12.5)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.7/dist-packages (from kaggle) (1.24.3)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.7/dist-packages (from kaggle) (4.0.1)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from kaggle) (2.23.0)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.7/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->kaggle) (3.0.4)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->kaggle) (2.10)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "btHh4rD24jEW"
      },
      "source": [
        "from google.colab import files\n",
        "files.upload()\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dl5r2Eni4mg9"
      },
      "source": [
        "!mkdir -p ~/.kaggle\n",
        "!cp kaggle.json ~/.kaggle\n",
        "!chmod 600 ~/.kaggle/kaggle.json"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tAkqQZA35EZY",
        "outputId": "8aaee406-ac3c-4d13-8d2c-6fc61c80545f"
      },
      "source": [
        "!kaggle datasets download -d iabhishekofficial/mobile-price-classification"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading mobile-price-classification.zip to /content\n",
            "\r  0% 0.00/70.6k [00:00<?, ?B/s]\n",
            "\r100% 70.6k/70.6k [00:00<00:00, 26.3MB/s]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fm8KkbYm5H4c",
        "outputId": "7c3ddfbc-7319-4068-84df-15c206db2679"
      },
      "source": [
        "!unzip mobile-price-classification\n"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Archive:  mobile-price-classification.zip\n",
            "  inflating: test.csv                \n",
            "  inflating: train.csv               \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 422
        },
        "id": "A2S3sLxD9gET",
        "outputId": "8014e77a-5079-4f74-86da-426b31d9ef0a"
      },
      "source": [
        "import pandas as pd\n",
        "df = pd.read_csv('./train.csv')\n",
        "df"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>battery_power</th>\n",
              "      <th>blue</th>\n",
              "      <th>clock_speed</th>\n",
              "      <th>dual_sim</th>\n",
              "      <th>fc</th>\n",
              "      <th>four_g</th>\n",
              "      <th>int_memory</th>\n",
              "      <th>m_dep</th>\n",
              "      <th>mobile_wt</th>\n",
              "      <th>n_cores</th>\n",
              "      <th>pc</th>\n",
              "      <th>px_height</th>\n",
              "      <th>px_width</th>\n",
              "      <th>ram</th>\n",
              "      <th>sc_h</th>\n",
              "      <th>sc_w</th>\n",
              "      <th>talk_time</th>\n",
              "      <th>three_g</th>\n",
              "      <th>touch_screen</th>\n",
              "      <th>wifi</th>\n",
              "      <th>price_range</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>842</td>\n",
              "      <td>0</td>\n",
              "      <td>2.2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0.6</td>\n",
              "      <td>188</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>20</td>\n",
              "      <td>756</td>\n",
              "      <td>2549</td>\n",
              "      <td>9</td>\n",
              "      <td>7</td>\n",
              "      <td>19</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1021</td>\n",
              "      <td>1</td>\n",
              "      <td>0.5</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>53</td>\n",
              "      <td>0.7</td>\n",
              "      <td>136</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>905</td>\n",
              "      <td>1988</td>\n",
              "      <td>2631</td>\n",
              "      <td>17</td>\n",
              "      <td>3</td>\n",
              "      <td>7</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>563</td>\n",
              "      <td>1</td>\n",
              "      <td>0.5</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>41</td>\n",
              "      <td>0.9</td>\n",
              "      <td>145</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>1263</td>\n",
              "      <td>1716</td>\n",
              "      <td>2603</td>\n",
              "      <td>11</td>\n",
              "      <td>2</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>615</td>\n",
              "      <td>1</td>\n",
              "      <td>2.5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "      <td>0.8</td>\n",
              "      <td>131</td>\n",
              "      <td>6</td>\n",
              "      <td>9</td>\n",
              "      <td>1216</td>\n",
              "      <td>1786</td>\n",
              "      <td>2769</td>\n",
              "      <td>16</td>\n",
              "      <td>8</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1821</td>\n",
              "      <td>1</td>\n",
              "      <td>1.2</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>1</td>\n",
              "      <td>44</td>\n",
              "      <td>0.6</td>\n",
              "      <td>141</td>\n",
              "      <td>2</td>\n",
              "      <td>14</td>\n",
              "      <td>1208</td>\n",
              "      <td>1212</td>\n",
              "      <td>1411</td>\n",
              "      <td>8</td>\n",
              "      <td>2</td>\n",
              "      <td>15</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1995</th>\n",
              "      <td>794</td>\n",
              "      <td>1</td>\n",
              "      <td>0.5</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0.8</td>\n",
              "      <td>106</td>\n",
              "      <td>6</td>\n",
              "      <td>14</td>\n",
              "      <td>1222</td>\n",
              "      <td>1890</td>\n",
              "      <td>668</td>\n",
              "      <td>13</td>\n",
              "      <td>4</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1996</th>\n",
              "      <td>1965</td>\n",
              "      <td>1</td>\n",
              "      <td>2.6</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>39</td>\n",
              "      <td>0.2</td>\n",
              "      <td>187</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>915</td>\n",
              "      <td>1965</td>\n",
              "      <td>2032</td>\n",
              "      <td>11</td>\n",
              "      <td>10</td>\n",
              "      <td>16</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1997</th>\n",
              "      <td>1911</td>\n",
              "      <td>0</td>\n",
              "      <td>0.9</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>36</td>\n",
              "      <td>0.7</td>\n",
              "      <td>108</td>\n",
              "      <td>8</td>\n",
              "      <td>3</td>\n",
              "      <td>868</td>\n",
              "      <td>1632</td>\n",
              "      <td>3057</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1998</th>\n",
              "      <td>1512</td>\n",
              "      <td>0</td>\n",
              "      <td>0.9</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>46</td>\n",
              "      <td>0.1</td>\n",
              "      <td>145</td>\n",
              "      <td>5</td>\n",
              "      <td>5</td>\n",
              "      <td>336</td>\n",
              "      <td>670</td>\n",
              "      <td>869</td>\n",
              "      <td>18</td>\n",
              "      <td>10</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1999</th>\n",
              "      <td>510</td>\n",
              "      <td>1</td>\n",
              "      <td>2.0</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>45</td>\n",
              "      <td>0.9</td>\n",
              "      <td>168</td>\n",
              "      <td>6</td>\n",
              "      <td>16</td>\n",
              "      <td>483</td>\n",
              "      <td>754</td>\n",
              "      <td>3919</td>\n",
              "      <td>19</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2000 rows × 21 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      battery_power  blue  clock_speed  ...  touch_screen  wifi  price_range\n",
              "0               842     0          2.2  ...             0     1            1\n",
              "1              1021     1          0.5  ...             1     0            2\n",
              "2               563     1          0.5  ...             1     0            2\n",
              "3               615     1          2.5  ...             0     0            2\n",
              "4              1821     1          1.2  ...             1     0            1\n",
              "...             ...   ...          ...  ...           ...   ...          ...\n",
              "1995            794     1          0.5  ...             1     0            0\n",
              "1996           1965     1          2.6  ...             1     1            2\n",
              "1997           1911     0          0.9  ...             1     0            3\n",
              "1998           1512     0          0.9  ...             1     1            0\n",
              "1999            510     1          2.0  ...             1     1            3\n",
              "\n",
              "[2000 rows x 21 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gp4GW8Be9iDJ"
      },
      "source": [
        "dataset = df.values\n",
        "X  = dataset[:,0:20]\n",
        "Y = dataset[:, 20]"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o-Vl3MZX97YH"
      },
      "source": [
        "from sklearn import preprocessing"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oGmN2OrD97_g"
      },
      "source": [
        "min_max_scaler = preprocessing.MinMaxScaler()\n",
        "X_scale = min_max_scaler.fit_transform(X)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Bnofouv1I_Po",
        "outputId": "013ddae6-87bf-4438-dafa-5eec12cb42bd"
      },
      "source": [
        "X_scale"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.22778891, 0.        , 0.68      , ..., 0.        , 0.        ,\n",
              "        1.        ],\n",
              "       [0.34736139, 1.        , 0.        , ..., 1.        , 1.        ,\n",
              "        0.        ],\n",
              "       [0.04141617, 1.        , 0.        , ..., 1.        , 1.        ,\n",
              "        0.        ],\n",
              "       ...,\n",
              "       [0.94188377, 0.        , 0.16      , ..., 1.        , 1.        ,\n",
              "        0.        ],\n",
              "       [0.6753507 , 0.        , 0.16      , ..., 1.        , 1.        ,\n",
              "        1.        ],\n",
              "       [0.00601202, 1.        , 0.6       , ..., 1.        , 1.        ,\n",
              "        1.        ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JkSw55g4K5ra",
        "outputId": "66e74811-e5b5-4813-f562-cab8fed22d04"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_val_and_test, Y_train, Y_val_and_test = train_test_split(X_scale, Y, test_size=0.3)\n",
        "X_val, X_test, Y_val, Y_test = train_test_split(X_val_and_test, Y_val_and_test, test_size=0.5)\n",
        "print(X_train.shape, X_val.shape, X_test.shape, Y_train.shape, Y_val.shape, Y_test.shape)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1400, 20) (300, 20) (300, 20) (1400,) (300,) (300,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yfaPcBmHUbR_",
        "outputId": "b87ab9b1-7c07-44c9-ec58-1bd30c90660e"
      },
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "C = tf.constant(4, name = \"C\") \n",
        "\n",
        "Y_VAL = tf.one_hot( \n",
        "    Y_val, C, on_value = 1.0, off_value = 0.0, axis =-1) \n",
        "\n",
        "Y_TRAIN = tf.one_hot( \n",
        "    Y_train, C, on_value = 1.0, off_value = 0.0, axis =-1) \n",
        "\n",
        "Y_TEST = tf.one_hot( \n",
        "    Y_test, C, on_value = 1.0, off_value = 0.0, axis =-1) \n",
        "\n",
        "Y_VAL"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(300, 4), dtype=float32, numpy=\n",
              "array([[0., 0., 1., 0.],\n",
              "       [1., 0., 0., 0.],\n",
              "       [0., 0., 0., 1.],\n",
              "       ...,\n",
              "       [0., 1., 0., 0.],\n",
              "       [1., 0., 0., 0.],\n",
              "       [0., 0., 0., 1.]], dtype=float32)>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IrH1wzuYJDX_"
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Dense"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Xe7v7NXJLfN"
      },
      "source": [
        "model = Sequential([\n",
        "    Dense(32, activation='relu', input_shape=(20,)),\n",
        "    Dense(16, activation='relu'),\n",
        "    Dense(4, activation='sigmoid'),\n",
        "])"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6RpS_WMbJSXl"
      },
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jinCppjrKIyM",
        "outputId": "ad331c8a-536c-4fca-e5c6-a2c1249ce675"
      },
      "source": [
        "hist = model.fit(X_train, Y_TRAIN,\n",
        "          batch_size=32, epochs=100,\n",
        "          validation_data=(X_val, Y_VAL))"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "44/44 [==============================] - 3s 14ms/step - loss: 1.3945 - accuracy: 0.2696 - val_loss: 1.3898 - val_accuracy: 0.2433\n",
            "Epoch 2/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 1.3545 - accuracy: 0.3361 - val_loss: 1.3437 - val_accuracy: 0.3467\n",
            "Epoch 3/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 1.2893 - accuracy: 0.4545 - val_loss: 1.2632 - val_accuracy: 0.4267\n",
            "Epoch 4/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 1.1951 - accuracy: 0.5223 - val_loss: 1.1468 - val_accuracy: 0.5133\n",
            "Epoch 5/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 1.0539 - accuracy: 0.6171 - val_loss: 0.9938 - val_accuracy: 0.6200\n",
            "Epoch 6/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.9205 - accuracy: 0.6893 - val_loss: 0.8714 - val_accuracy: 0.6767\n",
            "Epoch 7/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.8122 - accuracy: 0.7094 - val_loss: 0.7736 - val_accuracy: 0.7367\n",
            "Epoch 8/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.7062 - accuracy: 0.7484 - val_loss: 0.6976 - val_accuracy: 0.7600\n",
            "Epoch 9/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.6238 - accuracy: 0.7895 - val_loss: 0.6385 - val_accuracy: 0.8200\n",
            "Epoch 10/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.5957 - accuracy: 0.8282 - val_loss: 0.5939 - val_accuracy: 0.8133\n",
            "Epoch 11/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.5331 - accuracy: 0.8323 - val_loss: 0.5539 - val_accuracy: 0.8300\n",
            "Epoch 12/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.4830 - accuracy: 0.8666 - val_loss: 0.5135 - val_accuracy: 0.8167\n",
            "Epoch 13/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.4601 - accuracy: 0.8723 - val_loss: 0.4859 - val_accuracy: 0.8467\n",
            "Epoch 14/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.4267 - accuracy: 0.8815 - val_loss: 0.4741 - val_accuracy: 0.8300\n",
            "Epoch 15/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.4090 - accuracy: 0.8830 - val_loss: 0.4595 - val_accuracy: 0.8267\n",
            "Epoch 16/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.3859 - accuracy: 0.8949 - val_loss: 0.4138 - val_accuracy: 0.8767\n",
            "Epoch 17/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.3561 - accuracy: 0.9083 - val_loss: 0.4161 - val_accuracy: 0.8533\n",
            "Epoch 18/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.3249 - accuracy: 0.9195 - val_loss: 0.3868 - val_accuracy: 0.8733\n",
            "Epoch 19/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.3252 - accuracy: 0.9034 - val_loss: 0.3680 - val_accuracy: 0.8600\n",
            "Epoch 20/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2995 - accuracy: 0.9224 - val_loss: 0.3579 - val_accuracy: 0.8833\n",
            "Epoch 21/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2837 - accuracy: 0.9268 - val_loss: 0.3453 - val_accuracy: 0.8733\n",
            "Epoch 22/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2762 - accuracy: 0.9151 - val_loss: 0.3303 - val_accuracy: 0.8833\n",
            "Epoch 23/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2782 - accuracy: 0.9189 - val_loss: 0.3612 - val_accuracy: 0.8633\n",
            "Epoch 24/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2486 - accuracy: 0.9281 - val_loss: 0.3040 - val_accuracy: 0.8833\n",
            "Epoch 25/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2392 - accuracy: 0.9394 - val_loss: 0.2984 - val_accuracy: 0.8767\n",
            "Epoch 26/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2308 - accuracy: 0.9407 - val_loss: 0.2892 - val_accuracy: 0.8800\n",
            "Epoch 27/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2040 - accuracy: 0.9478 - val_loss: 0.2839 - val_accuracy: 0.8833\n",
            "Epoch 28/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2007 - accuracy: 0.9659 - val_loss: 0.2776 - val_accuracy: 0.8833\n",
            "Epoch 29/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.2046 - accuracy: 0.9503 - val_loss: 0.2699 - val_accuracy: 0.8967\n",
            "Epoch 30/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1837 - accuracy: 0.9616 - val_loss: 0.2603 - val_accuracy: 0.8867\n",
            "Epoch 31/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1826 - accuracy: 0.9642 - val_loss: 0.2832 - val_accuracy: 0.8867\n",
            "Epoch 32/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.1853 - accuracy: 0.9502 - val_loss: 0.2526 - val_accuracy: 0.9000\n",
            "Epoch 33/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.1784 - accuracy: 0.9688 - val_loss: 0.2545 - val_accuracy: 0.9000\n",
            "Epoch 34/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1595 - accuracy: 0.9662 - val_loss: 0.2546 - val_accuracy: 0.8833\n",
            "Epoch 35/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1614 - accuracy: 0.9611 - val_loss: 0.2414 - val_accuracy: 0.9000\n",
            "Epoch 36/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1482 - accuracy: 0.9690 - val_loss: 0.2346 - val_accuracy: 0.8933\n",
            "Epoch 37/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1518 - accuracy: 0.9663 - val_loss: 0.2336 - val_accuracy: 0.9000\n",
            "Epoch 38/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1473 - accuracy: 0.9708 - val_loss: 0.2344 - val_accuracy: 0.8933\n",
            "Epoch 39/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1441 - accuracy: 0.9598 - val_loss: 0.2279 - val_accuracy: 0.8967\n",
            "Epoch 40/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1472 - accuracy: 0.9692 - val_loss: 0.2366 - val_accuracy: 0.8900\n",
            "Epoch 41/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1352 - accuracy: 0.9700 - val_loss: 0.2283 - val_accuracy: 0.8967\n",
            "Epoch 42/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1352 - accuracy: 0.9661 - val_loss: 0.2228 - val_accuracy: 0.8933\n",
            "Epoch 43/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1244 - accuracy: 0.9632 - val_loss: 0.2173 - val_accuracy: 0.9100\n",
            "Epoch 44/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.1207 - accuracy: 0.9797 - val_loss: 0.2223 - val_accuracy: 0.8933\n",
            "Epoch 45/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1254 - accuracy: 0.9763 - val_loss: 0.2213 - val_accuracy: 0.9000\n",
            "Epoch 46/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1215 - accuracy: 0.9716 - val_loss: 0.2302 - val_accuracy: 0.8867\n",
            "Epoch 47/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1241 - accuracy: 0.9699 - val_loss: 0.2106 - val_accuracy: 0.9000\n",
            "Epoch 48/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1131 - accuracy: 0.9775 - val_loss: 0.2130 - val_accuracy: 0.9033\n",
            "Epoch 49/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.1132 - accuracy: 0.9823 - val_loss: 0.2111 - val_accuracy: 0.8967\n",
            "Epoch 50/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1015 - accuracy: 0.9817 - val_loss: 0.2204 - val_accuracy: 0.8867\n",
            "Epoch 51/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1092 - accuracy: 0.9804 - val_loss: 0.2141 - val_accuracy: 0.8900\n",
            "Epoch 52/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0965 - accuracy: 0.9838 - val_loss: 0.2073 - val_accuracy: 0.8967\n",
            "Epoch 53/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1001 - accuracy: 0.9786 - val_loss: 0.2078 - val_accuracy: 0.9000\n",
            "Epoch 54/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1039 - accuracy: 0.9779 - val_loss: 0.2119 - val_accuracy: 0.8900\n",
            "Epoch 55/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.1056 - accuracy: 0.9787 - val_loss: 0.2113 - val_accuracy: 0.9067\n",
            "Epoch 56/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0834 - accuracy: 0.9912 - val_loss: 0.2104 - val_accuracy: 0.9033\n",
            "Epoch 57/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0933 - accuracy: 0.9784 - val_loss: 0.1991 - val_accuracy: 0.9000\n",
            "Epoch 58/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0907 - accuracy: 0.9833 - val_loss: 0.1950 - val_accuracy: 0.9233\n",
            "Epoch 59/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0879 - accuracy: 0.9838 - val_loss: 0.1969 - val_accuracy: 0.9000\n",
            "Epoch 60/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0891 - accuracy: 0.9831 - val_loss: 0.2166 - val_accuracy: 0.9033\n",
            "Epoch 61/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0933 - accuracy: 0.9776 - val_loss: 0.1995 - val_accuracy: 0.8900\n",
            "Epoch 62/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0789 - accuracy: 0.9887 - val_loss: 0.2028 - val_accuracy: 0.8900\n",
            "Epoch 63/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0866 - accuracy: 0.9854 - val_loss: 0.1952 - val_accuracy: 0.9100\n",
            "Epoch 64/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.0745 - accuracy: 0.9890 - val_loss: 0.2271 - val_accuracy: 0.8800\n",
            "Epoch 65/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.0799 - accuracy: 0.9849 - val_loss: 0.1957 - val_accuracy: 0.9067\n",
            "Epoch 66/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0848 - accuracy: 0.9827 - val_loss: 0.2012 - val_accuracy: 0.8933\n",
            "Epoch 67/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0820 - accuracy: 0.9824 - val_loss: 0.1910 - val_accuracy: 0.8967\n",
            "Epoch 68/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0705 - accuracy: 0.9913 - val_loss: 0.1971 - val_accuracy: 0.8933\n",
            "Epoch 69/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0721 - accuracy: 0.9923 - val_loss: 0.1861 - val_accuracy: 0.9100\n",
            "Epoch 70/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0708 - accuracy: 0.9929 - val_loss: 0.1985 - val_accuracy: 0.9033\n",
            "Epoch 71/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0665 - accuracy: 0.9875 - val_loss: 0.2061 - val_accuracy: 0.9067\n",
            "Epoch 72/100\n",
            "44/44 [==============================] - 0s 5ms/step - loss: 0.0702 - accuracy: 0.9860 - val_loss: 0.1998 - val_accuracy: 0.8900\n",
            "Epoch 73/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0680 - accuracy: 0.9858 - val_loss: 0.1926 - val_accuracy: 0.9033\n",
            "Epoch 74/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0665 - accuracy: 0.9929 - val_loss: 0.1822 - val_accuracy: 0.9133\n",
            "Epoch 75/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0612 - accuracy: 0.9913 - val_loss: 0.2045 - val_accuracy: 0.9100\n",
            "Epoch 76/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0686 - accuracy: 0.9865 - val_loss: 0.1866 - val_accuracy: 0.9100\n",
            "Epoch 77/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0660 - accuracy: 0.9937 - val_loss: 0.1875 - val_accuracy: 0.9167\n",
            "Epoch 78/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0677 - accuracy: 0.9896 - val_loss: 0.1946 - val_accuracy: 0.9200\n",
            "Epoch 79/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0586 - accuracy: 0.9914 - val_loss: 0.1966 - val_accuracy: 0.8900\n",
            "Epoch 80/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0561 - accuracy: 0.9929 - val_loss: 0.1833 - val_accuracy: 0.9100\n",
            "Epoch 81/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0535 - accuracy: 0.9964 - val_loss: 0.1977 - val_accuracy: 0.9000\n",
            "Epoch 82/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0564 - accuracy: 0.9930 - val_loss: 0.1832 - val_accuracy: 0.9067\n",
            "Epoch 83/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0538 - accuracy: 0.9977 - val_loss: 0.1951 - val_accuracy: 0.8967\n",
            "Epoch 84/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0547 - accuracy: 0.9986 - val_loss: 0.1939 - val_accuracy: 0.9000\n",
            "Epoch 85/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0558 - accuracy: 0.9938 - val_loss: 0.1879 - val_accuracy: 0.9100\n",
            "Epoch 86/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0468 - accuracy: 0.9939 - val_loss: 0.1843 - val_accuracy: 0.9133\n",
            "Epoch 87/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0495 - accuracy: 0.9941 - val_loss: 0.1926 - val_accuracy: 0.9033\n",
            "Epoch 88/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0581 - accuracy: 0.9858 - val_loss: 0.2061 - val_accuracy: 0.8933\n",
            "Epoch 89/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0512 - accuracy: 0.9944 - val_loss: 0.2058 - val_accuracy: 0.9033\n",
            "Epoch 90/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0581 - accuracy: 0.9906 - val_loss: 0.2008 - val_accuracy: 0.9000\n",
            "Epoch 91/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0480 - accuracy: 0.9968 - val_loss: 0.1840 - val_accuracy: 0.9067\n",
            "Epoch 92/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0456 - accuracy: 0.9966 - val_loss: 0.1931 - val_accuracy: 0.9067\n",
            "Epoch 93/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0509 - accuracy: 0.9963 - val_loss: 0.1912 - val_accuracy: 0.8967\n",
            "Epoch 94/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0456 - accuracy: 0.9957 - val_loss: 0.1856 - val_accuracy: 0.9033\n",
            "Epoch 95/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0411 - accuracy: 0.9984 - val_loss: 0.2016 - val_accuracy: 0.9033\n",
            "Epoch 96/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0432 - accuracy: 0.9988 - val_loss: 0.1924 - val_accuracy: 0.8933\n",
            "Epoch 97/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0448 - accuracy: 0.9946 - val_loss: 0.1898 - val_accuracy: 0.9067\n",
            "Epoch 98/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0450 - accuracy: 0.9978 - val_loss: 0.1897 - val_accuracy: 0.9000\n",
            "Epoch 99/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0412 - accuracy: 0.9926 - val_loss: 0.2257 - val_accuracy: 0.9133\n",
            "Epoch 100/100\n",
            "44/44 [==============================] - 0s 4ms/step - loss: 0.0538 - accuracy: 0.9840 - val_loss: 0.2047 - val_accuracy: 0.8967\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KDFeSasBQz-6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "944e6666-2073-4638-b976-4f85588911e4"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.plot(hist.history['loss'])\n",
        "plt.plot(hist.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "3rhhryEnvTmk",
        "outputId": "4f702f45-9ca9-49e1-a60d-70eb23d93f13"
      },
      "source": [
        "plt.plot(hist.history['accuracy'])\n",
        "plt.plot(hist.history['val_accuracy'])\n",
        "plt.title('Model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='lower right')\n",
        "plt.show()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YMhRozokw5YQ",
        "outputId": "51622f3a-0970-4305-ef89-508b8382a87b"
      },
      "source": [
        "from keras.layers import Dropout\n",
        "from keras import regularizers\n",
        "model_2 = Sequential([\n",
        "    Dense(1000, activation='relu', input_shape=(20,), kernel_regularizer = regularizers.l2(0.01)),\n",
        "    Dropout(0.3),\n",
        "    Dense(1000, activation='relu', kernel_regularizer = regularizers.l2(0.01)),\n",
        "    Dropout(0.3),\n",
        "    Dense(1000, activation='relu', kernel_regularizer = regularizers.l2(0.01)),\n",
        "    Dropout(0.3),\n",
        "    Dense(1000, activation='relu', kernel_regularizer = regularizers.l2(0.01)),\n",
        "    Dropout(0.3),\n",
        "    Dense(4, activation='sigmoid', kernel_regularizer = regularizers.l2(0.01)),\n",
        "])\n",
        "\n",
        "model_2.compile(optimizer='adam',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "hist_2 = model_2.fit(X_train, Y_TRAIN,\n",
        "          batch_size=32, epochs=100,\n",
        "          validation_data=(X_val, Y_VAL))\n",
        "# hist_2 = model_2.fit(X_train, Y_TRAIN, batch_size=32, epochs=100, validation_data=(X_val, Y_VAL))"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "44/44 [==============================] - 1s 14ms/step - loss: 20.3027 - accuracy: 0.2889 - val_loss: 2.7047 - val_accuracy: 0.5133\n",
            "Epoch 2/100\n",
            "44/44 [==============================] - 0s 9ms/step - loss: 2.0159 - accuracy: 0.5722 - val_loss: 1.0636 - val_accuracy: 0.7333\n",
            "Epoch 3/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 1.0959 - accuracy: 0.7209 - val_loss: 1.0253 - val_accuracy: 0.7633\n",
            "Epoch 4/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 1.0150 - accuracy: 0.7390 - val_loss: 0.8224 - val_accuracy: 0.8467\n",
            "Epoch 5/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.8466 - accuracy: 0.8323 - val_loss: 0.7803 - val_accuracy: 0.8733\n",
            "Epoch 6/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.8102 - accuracy: 0.8350 - val_loss: 0.8123 - val_accuracy: 0.8467\n",
            "Epoch 7/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7887 - accuracy: 0.8625 - val_loss: 0.7164 - val_accuracy: 0.9033\n",
            "Epoch 8/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7791 - accuracy: 0.8352 - val_loss: 0.7367 - val_accuracy: 0.8900\n",
            "Epoch 9/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7288 - accuracy: 0.8824 - val_loss: 0.7940 - val_accuracy: 0.8100\n",
            "Epoch 10/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7426 - accuracy: 0.8681 - val_loss: 0.7010 - val_accuracy: 0.8967\n",
            "Epoch 11/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7115 - accuracy: 0.8849 - val_loss: 0.6588 - val_accuracy: 0.9233\n",
            "Epoch 12/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6964 - accuracy: 0.8663 - val_loss: 0.6794 - val_accuracy: 0.8967\n",
            "Epoch 13/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7088 - accuracy: 0.8742 - val_loss: 0.6470 - val_accuracy: 0.9467\n",
            "Epoch 14/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.7404 - accuracy: 0.8522 - val_loss: 0.7184 - val_accuracy: 0.8333\n",
            "Epoch 15/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6961 - accuracy: 0.8771 - val_loss: 0.6460 - val_accuracy: 0.9100\n",
            "Epoch 16/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6753 - accuracy: 0.8769 - val_loss: 0.6235 - val_accuracy: 0.9533\n",
            "Epoch 17/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6801 - accuracy: 0.8689 - val_loss: 0.7529 - val_accuracy: 0.8033\n",
            "Epoch 18/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6394 - accuracy: 0.9022 - val_loss: 0.6604 - val_accuracy: 0.8933\n",
            "Epoch 19/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6464 - accuracy: 0.9000 - val_loss: 0.6904 - val_accuracy: 0.8700\n",
            "Epoch 20/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6712 - accuracy: 0.8881 - val_loss: 0.6273 - val_accuracy: 0.9200\n",
            "Epoch 21/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6470 - accuracy: 0.8922 - val_loss: 0.6840 - val_accuracy: 0.8933\n",
            "Epoch 22/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6913 - accuracy: 0.8605 - val_loss: 0.7812 - val_accuracy: 0.7900\n",
            "Epoch 23/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.7017 - accuracy: 0.8540 - val_loss: 0.7081 - val_accuracy: 0.8267\n",
            "Epoch 24/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6971 - accuracy: 0.8540 - val_loss: 0.6420 - val_accuracy: 0.8800\n",
            "Epoch 25/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6412 - accuracy: 0.8938 - val_loss: 0.6082 - val_accuracy: 0.9167\n",
            "Epoch 26/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6421 - accuracy: 0.8923 - val_loss: 0.6313 - val_accuracy: 0.9000\n",
            "Epoch 27/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6522 - accuracy: 0.8739 - val_loss: 0.6011 - val_accuracy: 0.9200\n",
            "Epoch 28/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6225 - accuracy: 0.9028 - val_loss: 0.6028 - val_accuracy: 0.9233\n",
            "Epoch 29/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6556 - accuracy: 0.8639 - val_loss: 0.6666 - val_accuracy: 0.8600\n",
            "Epoch 30/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6440 - accuracy: 0.8796 - val_loss: 0.6072 - val_accuracy: 0.9200\n",
            "Epoch 31/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6245 - accuracy: 0.9016 - val_loss: 0.5695 - val_accuracy: 0.9567\n",
            "Epoch 32/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5989 - accuracy: 0.9172 - val_loss: 0.6048 - val_accuracy: 0.9300\n",
            "Epoch 33/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6596 - accuracy: 0.8779 - val_loss: 0.6393 - val_accuracy: 0.9033\n",
            "Epoch 34/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6825 - accuracy: 0.8565 - val_loss: 0.5867 - val_accuracy: 0.9333\n",
            "Epoch 35/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6425 - accuracy: 0.8677 - val_loss: 0.7244 - val_accuracy: 0.7867\n",
            "Epoch 36/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6660 - accuracy: 0.8685 - val_loss: 0.6074 - val_accuracy: 0.8933\n",
            "Epoch 37/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6618 - accuracy: 0.8696 - val_loss: 0.5891 - val_accuracy: 0.9133\n",
            "Epoch 38/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5819 - accuracy: 0.9161 - val_loss: 0.6015 - val_accuracy: 0.9167\n",
            "Epoch 39/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6250 - accuracy: 0.8882 - val_loss: 0.5942 - val_accuracy: 0.9300\n",
            "Epoch 40/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6267 - accuracy: 0.8854 - val_loss: 0.6178 - val_accuracy: 0.8933\n",
            "Epoch 41/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6178 - accuracy: 0.8978 - val_loss: 0.5884 - val_accuracy: 0.9133\n",
            "Epoch 42/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6472 - accuracy: 0.8722 - val_loss: 0.5789 - val_accuracy: 0.9367\n",
            "Epoch 43/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5857 - accuracy: 0.9216 - val_loss: 0.5707 - val_accuracy: 0.9400\n",
            "Epoch 44/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6112 - accuracy: 0.8943 - val_loss: 0.7544 - val_accuracy: 0.8433\n",
            "Epoch 45/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6230 - accuracy: 0.9012 - val_loss: 0.5714 - val_accuracy: 0.9133\n",
            "Epoch 46/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6695 - accuracy: 0.8610 - val_loss: 0.6677 - val_accuracy: 0.8467\n",
            "Epoch 47/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6163 - accuracy: 0.8876 - val_loss: 0.6441 - val_accuracy: 0.8367\n",
            "Epoch 48/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6081 - accuracy: 0.9059 - val_loss: 0.6554 - val_accuracy: 0.8333\n",
            "Epoch 49/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6343 - accuracy: 0.8823 - val_loss: 0.6119 - val_accuracy: 0.8767\n",
            "Epoch 50/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6005 - accuracy: 0.8953 - val_loss: 0.6533 - val_accuracy: 0.8433\n",
            "Epoch 51/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6062 - accuracy: 0.9001 - val_loss: 0.5896 - val_accuracy: 0.9233\n",
            "Epoch 52/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5983 - accuracy: 0.8933 - val_loss: 0.6273 - val_accuracy: 0.8767\n",
            "Epoch 53/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6138 - accuracy: 0.8843 - val_loss: 0.5831 - val_accuracy: 0.9100\n",
            "Epoch 54/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6193 - accuracy: 0.8865 - val_loss: 0.5922 - val_accuracy: 0.9200\n",
            "Epoch 55/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6072 - accuracy: 0.8921 - val_loss: 0.5849 - val_accuracy: 0.9000\n",
            "Epoch 56/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5659 - accuracy: 0.9190 - val_loss: 0.6285 - val_accuracy: 0.8767\n",
            "Epoch 57/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6134 - accuracy: 0.8959 - val_loss: 0.5950 - val_accuracy: 0.9067\n",
            "Epoch 58/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6035 - accuracy: 0.8943 - val_loss: 0.6890 - val_accuracy: 0.8433\n",
            "Epoch 59/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5876 - accuracy: 0.9010 - val_loss: 0.5869 - val_accuracy: 0.9200\n",
            "Epoch 60/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6501 - accuracy: 0.8652 - val_loss: 0.6125 - val_accuracy: 0.9300\n",
            "Epoch 61/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6129 - accuracy: 0.8980 - val_loss: 0.5604 - val_accuracy: 0.9200\n",
            "Epoch 62/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5903 - accuracy: 0.9006 - val_loss: 0.6050 - val_accuracy: 0.8967\n",
            "Epoch 63/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6303 - accuracy: 0.8726 - val_loss: 0.5739 - val_accuracy: 0.9233\n",
            "Epoch 64/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5897 - accuracy: 0.9137 - val_loss: 0.6669 - val_accuracy: 0.8567\n",
            "Epoch 65/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5929 - accuracy: 0.9040 - val_loss: 0.6530 - val_accuracy: 0.8833\n",
            "Epoch 66/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6250 - accuracy: 0.8838 - val_loss: 0.5642 - val_accuracy: 0.9233\n",
            "Epoch 67/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5725 - accuracy: 0.8982 - val_loss: 0.6049 - val_accuracy: 0.8933\n",
            "Epoch 68/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5871 - accuracy: 0.8872 - val_loss: 0.5457 - val_accuracy: 0.9433\n",
            "Epoch 69/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5967 - accuracy: 0.8972 - val_loss: 0.6053 - val_accuracy: 0.8767\n",
            "Epoch 70/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5810 - accuracy: 0.9027 - val_loss: 0.5821 - val_accuracy: 0.8900\n",
            "Epoch 71/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5715 - accuracy: 0.9092 - val_loss: 0.5745 - val_accuracy: 0.9300\n",
            "Epoch 72/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5641 - accuracy: 0.9151 - val_loss: 0.5784 - val_accuracy: 0.8967\n",
            "Epoch 73/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6684 - accuracy: 0.8425 - val_loss: 0.6135 - val_accuracy: 0.8833\n",
            "Epoch 74/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5834 - accuracy: 0.9046 - val_loss: 0.5861 - val_accuracy: 0.9100\n",
            "Epoch 75/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5677 - accuracy: 0.9046 - val_loss: 0.5975 - val_accuracy: 0.8933\n",
            "Epoch 76/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5845 - accuracy: 0.8968 - val_loss: 0.5927 - val_accuracy: 0.8800\n",
            "Epoch 77/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6018 - accuracy: 0.8907 - val_loss: 0.5668 - val_accuracy: 0.9167\n",
            "Epoch 78/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5730 - accuracy: 0.9041 - val_loss: 0.7347 - val_accuracy: 0.8167\n",
            "Epoch 79/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6346 - accuracy: 0.8690 - val_loss: 0.5502 - val_accuracy: 0.9367\n",
            "Epoch 80/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5875 - accuracy: 0.8929 - val_loss: 0.5869 - val_accuracy: 0.8967\n",
            "Epoch 81/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5759 - accuracy: 0.9086 - val_loss: 0.5574 - val_accuracy: 0.9167\n",
            "Epoch 82/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5681 - accuracy: 0.8969 - val_loss: 0.5382 - val_accuracy: 0.9467\n",
            "Epoch 83/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5956 - accuracy: 0.8940 - val_loss: 1.0420 - val_accuracy: 0.6833\n",
            "Epoch 84/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.7703 - accuracy: 0.7892 - val_loss: 0.5572 - val_accuracy: 0.9567\n",
            "Epoch 85/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5934 - accuracy: 0.8995 - val_loss: 0.5582 - val_accuracy: 0.9300\n",
            "Epoch 86/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6105 - accuracy: 0.8815 - val_loss: 0.9053 - val_accuracy: 0.7033\n",
            "Epoch 87/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6892 - accuracy: 0.8442 - val_loss: 0.7505 - val_accuracy: 0.7800\n",
            "Epoch 88/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6013 - accuracy: 0.8943 - val_loss: 0.5626 - val_accuracy: 0.9167\n",
            "Epoch 89/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5851 - accuracy: 0.8943 - val_loss: 0.5450 - val_accuracy: 0.9367\n",
            "Epoch 90/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5672 - accuracy: 0.9079 - val_loss: 0.6291 - val_accuracy: 0.8400\n",
            "Epoch 91/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6081 - accuracy: 0.8992 - val_loss: 0.7397 - val_accuracy: 0.7967\n",
            "Epoch 92/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6249 - accuracy: 0.8783 - val_loss: 0.5778 - val_accuracy: 0.8900\n",
            "Epoch 93/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.6188 - accuracy: 0.8769 - val_loss: 0.5704 - val_accuracy: 0.9033\n",
            "Epoch 94/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5756 - accuracy: 0.9034 - val_loss: 0.5568 - val_accuracy: 0.9200\n",
            "Epoch 95/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5874 - accuracy: 0.9032 - val_loss: 0.5657 - val_accuracy: 0.8967\n",
            "Epoch 96/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5965 - accuracy: 0.8973 - val_loss: 0.5832 - val_accuracy: 0.9033\n",
            "Epoch 97/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6107 - accuracy: 0.8909 - val_loss: 0.5573 - val_accuracy: 0.9167\n",
            "Epoch 98/100\n",
            "44/44 [==============================] - 0s 7ms/step - loss: 0.5973 - accuracy: 0.8910 - val_loss: 0.7398 - val_accuracy: 0.7733\n",
            "Epoch 99/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.6184 - accuracy: 0.8738 - val_loss: 0.5627 - val_accuracy: 0.9233\n",
            "Epoch 100/100\n",
            "44/44 [==============================] - 0s 8ms/step - loss: 0.5960 - accuracy: 0.8812 - val_loss: 0.6762 - val_accuracy: 0.8133\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "X0AwQp6_zAmB",
        "outputId": "5561998a-ce23-4a78-a73a-e01f34c269a1"
      },
      "source": [
        "plt.plot(hist_2.history['loss'])\n",
        "plt.plot(hist_2.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "x1Nnq_5vz48m",
        "outputId": "326ee22f-6e2a-4cee-ce50-e03bb3d3764f"
      },
      "source": [
        "plt.plot(hist_2.history['accuracy'])\n",
        "plt.plot(hist_2.history['val_accuracy'])\n",
        "plt.title('Model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='lower right')\n",
        "plt.show()"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Wmt33XFz6WP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ddc66581-1a58-435c-ea5a-5d7e549f5502"
      },
      "source": [
        "model.evaluate(X_test, Y_TEST)\n"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10/10 [==============================] - 0s 2ms/step - loss: 0.1770 - accuracy: 0.9033\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.17700804769992828, 0.903333306312561]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gsMAZUE04_g5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "68426118-8862-4e34-f140-55ac1372b9c7"
      },
      "source": [
        "model_2.evaluate(X_test, Y_TEST)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10/10 [==============================] - 0s 4ms/step - loss: 0.6659 - accuracy: 0.8467\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6658517718315125, 0.846666693687439]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KJL5vJyewtr5"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}