Tugas_5_(3model)convolutional_neural_network.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "convolutional_neural_network.ipynb",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/MrezaFd/Tugas-Besar-ML-059-055-/blob/main/Tugas_5_(3model)convolutional_neural_network.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pg3l1ejwCF9P"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_yhI15_dBJI1",
        "outputId": "ef536c61-ab8e-4090-8ce2-0a1551875c49",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 91
        }
      },
      "source": [
        "from google.colab import files #import lib\n",
        "files.upload()"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-f0347755-e115-4187-9bb8-d1828ed964ce\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-f0347755-e115-4187-9bb8-d1828ed964ce\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving kaggle.json to kaggle.json\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'kaggle.json': b'{\"username\":\"rezafdz\",\"key\":\"fd16295d7b817e274541292110c8f784\"}'}"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T-PttF09A1jP"
      },
      "source": [
        "!mkdir -p ~/.kaggle #membuat directory bernama kaggle \n",
        "!cp kaggle.json ~/.kaggle/ #copy file kaggle.json upload tadi ke directory baru\n",
        "!chmod 600 ~/.kaggle/kaggle.json #berikan izin untuk file kaggle.jsonv"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OvexdoKwBZ5V",
        "outputId": "60cce84e-6197-4128-9a28-3d7e4a8fd748",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "#Unduh dataset\n",
        "!kaggle datasets download -d fanconic/skin-cancer-malignant-vs-benign"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading skin-cancer-malignant-vs-benign.zip to /content\n",
            " 98% 317M/325M [00:04<00:00, 83.4MB/s]\n",
            "100% 325M/325M [00:04<00:00, 69.9MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4rtnNrFsEDIf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "07df2e76-19f0-4838-fd45-394aad8b0e37"
      },
      "source": [
        "from zipfile import ZipFile\n",
        "file_name = \"skin-cancer-malignant-vs-benign.zip\"\n",
        "\n",
        "with ZipFile(file_name,'r') as zip :\n",
        "  zip.extractall()\n",
        "  print('Berhasil')"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Berhasil\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "31Pr3YSaDHnn"
      },
      "source": [
        "!mkdir a"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u6Wd9eg8DOao"
      },
      "source": [
        "!mv -t a test train"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y9L_1kqnEEjc"
      },
      "source": [
        "# Definisikan path untuk data train dan data validation\n",
        "import os\n",
        "base_dir = '/content/a'\n",
        "\n",
        "\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "validation_dir = os.path.join(base_dir, 'test')"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9gc0BCSREF0M"
      },
      "source": [
        "import cv2\n",
        "import numpy as np\n",
        "\n",
        "# Gather data train\n",
        "train_data = []\n",
        "train_label = []\n",
        "for r, d, f in os.walk(train_dir):\n",
        "    for file in f:\n",
        "        if \".jpg\" in file:\n",
        "            imagePath = os.path.join(r, file)\n",
        "            image = cv2.imread(imagePath)\n",
        "            image = cv2.resize(image, (150,150))\n",
        "            train_data.append(image)\n",
        "            label = imagePath.split(os.path.sep)[-2]\n",
        "            train_label.append(label)\n",
        "\n",
        "train_data = np.array(train_data)\n",
        "train_label = np.array(train_label)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FuoKFhhVEG_O"
      },
      "source": [
        "# Gather data validation\n",
        "val_data = []\n",
        "val_label = []\n",
        "for r, d, f in os.walk(validation_dir):\n",
        "    for file in f:\n",
        "        if \".jpg\" in file:\n",
        "            imagePath = os.path.join(r, file)\n",
        "            image = cv2.imread(imagePath)\n",
        "            image = cv2.resize(image, (150,150))\n",
        "            val_data.append(image)\n",
        "            label = imagePath.split(os.path.sep)[-2]\n",
        "            val_label.append(label)\n",
        "\n",
        "val_data = np.array(val_data)\n",
        "val_label = np.array(val_label)"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kiXbwrN3ELqv"
      },
      "source": [
        "<h3>Data Preprocessing</h3>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m5VhnvtrEI1I",
        "outputId": "b16afb5f-aaef-43d8-f164-2c0f70ca40f9"
      },
      "source": [
        "# Tampilkan shape dari data train dan data validation\n",
        "print(\"Train Data = \", train_data.shape)\n",
        "print(\"Train Label = \", train_label.shape)\n",
        "print(\"Validation Data = \", val_data.shape)\n",
        "print(\"Validation Label = \", val_label.shape)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Data =  (2637, 150, 150, 3)\n",
            "Train Label =  (2637,)\n",
            "Validation Data =  (660, 150, 150, 3)\n",
            "Validation Label =  (660,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q1LVp6dCENne",
        "outputId": "440a6f9e-1ca5-451b-d406-0c9b31ce4ea9"
      },
      "source": [
        "# Normalisasi dataset\n",
        "print(\"Data sebelum di-normalisasi \", train_data[0][0][0])\n",
        "\n",
        "x_train = train_data.astype('float32') / 255.0\n",
        "x_val = val_data.astype('float32') / 255.0\n",
        "print(\"Data setelah di-normalisasi \", x_train[0][0][0])"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Data sebelum di-normalisasi  [184 171 248]\n",
            "Data setelah di-normalisasi  [0.72156864 0.67058825 0.972549  ]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tc7FivC6EPhm",
        "outputId": "6bb3f064-2b7b-49ff-8e0d-9e5762cedba9"
      },
      "source": [
        "# Transformasi label encoder\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "print(\"Label sebelum di-encoder \", train_label[995:2005])\n",
        "\n",
        "lb = LabelEncoder()\n",
        "y_train = lb.fit_transform(train_label)\n",
        "y_val = lb.fit_transform(val_label)\n",
        "\n",
        "print(\"Label setelah di-encoder \", y_train[995:2005])"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Label sebelum di-encoder  ['benign' 'benign' 'benign' ... 'malignant' 'malignant' 'malignant']\n",
            "Label setelah di-encoder  [0 0 0 ... 1 1 1]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4jVWbOhOEzQJ"
      },
      "source": [
        "<h3>Definisikan Model 1</h3>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import InputLayer, Dense, Conv2D, MaxPool2D, Flatten, BatchNormalization, Dropout\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, EarlyStopping\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "import time"
      ],
      "metadata": {
        "id": "KrR7ChtaPxw6"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Feature Extraction Layer\n",
        "model = Sequential()\n",
        "\n",
        "model.add(InputLayer(input_shape=[150,150,3]))\n",
        "model.add(Conv2D(filters=16, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Conv2D(filters=32, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Conv2D(filters=64, kernel_size=3, strides=1, padding='same', activation='relu'))\n",
        "model.add(MaxPool2D(pool_size=2, padding='same'))\n",
        "model.add(Flatten())\n",
        "# Fully Connected Layer\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dense(1, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "FL0L2f9Aq3vb"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Print model summary\n",
        "print(model.summary())"
      ],
      "metadata": {
        "id": "LlJn1hPgq9RP",
        "outputId": "435cff7b-d7e9-4c4c-be5f-190b903f9389",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 150, 150, 16)      448       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 75, 75, 16)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 75, 75, 32)        4640      \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 38, 38, 32)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 38, 38, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_2 (MaxPooling  (None, 19, 19, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 23104)             0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 128)               2957440   \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 2,981,153\n",
            "Trainable params: 2,981,153\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model.compile(optimizer=Adam(lr=0.001), \n",
        "              loss='binary_crossentropy', \n",
        "              metrics=['acc'])"
      ],
      "metadata": {
        "id": "Ib7grrGrrAK6",
        "outputId": "aadde141-5958-4c0a-aa4c-fad7937bbdfe",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "H = model.fit(x_train, y_train, batch_size=20, epochs=100, validation_data=(x_val, y_val))"
      ],
      "metadata": {
        "id": "52d_HJuCrF6e",
        "outputId": "0ec91464-a6d4-4a53-8f30-9c3c4bdd564b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "132/132 [==============================] - 17s 41ms/step - loss: 0.5511 - acc: 0.6928 - val_loss: 0.4173 - val_acc: 0.7773\n",
            "Epoch 2/100\n",
            "132/132 [==============================] - 4s 30ms/step - loss: 0.4277 - acc: 0.7876 - val_loss: 0.3894 - val_acc: 0.8000\n",
            "Epoch 3/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3823 - acc: 0.8104 - val_loss: 0.3708 - val_acc: 0.7955\n",
            "Epoch 4/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3708 - acc: 0.8184 - val_loss: 0.3568 - val_acc: 0.8197\n",
            "Epoch 5/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3345 - acc: 0.8339 - val_loss: 0.3501 - val_acc: 0.8364\n",
            "Epoch 6/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.3219 - acc: 0.8385 - val_loss: 0.3394 - val_acc: 0.8364\n",
            "Epoch 7/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2988 - acc: 0.8582 - val_loss: 0.3463 - val_acc: 0.8242\n",
            "Epoch 8/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2762 - acc: 0.8726 - val_loss: 0.3760 - val_acc: 0.8364\n",
            "Epoch 9/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2761 - acc: 0.8730 - val_loss: 0.3622 - val_acc: 0.8242\n",
            "Epoch 10/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2499 - acc: 0.8881 - val_loss: 0.3325 - val_acc: 0.8288\n",
            "Epoch 11/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2260 - acc: 0.8984 - val_loss: 0.3551 - val_acc: 0.8303\n",
            "Epoch 12/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.2020 - acc: 0.9090 - val_loss: 0.3549 - val_acc: 0.8364\n",
            "Epoch 13/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.1809 - acc: 0.9219 - val_loss: 0.4316 - val_acc: 0.8348\n",
            "Epoch 14/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1778 - acc: 0.9261 - val_loss: 0.3514 - val_acc: 0.8409\n",
            "Epoch 15/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1514 - acc: 0.9439 - val_loss: 0.4128 - val_acc: 0.8379\n",
            "Epoch 16/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1281 - acc: 0.9526 - val_loss: 0.4565 - val_acc: 0.8409\n",
            "Epoch 17/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.1232 - acc: 0.9564 - val_loss: 0.4156 - val_acc: 0.8242\n",
            "Epoch 18/100\n",
            "132/132 [==============================] - 4s 27ms/step - loss: 0.1559 - acc: 0.9367 - val_loss: 0.4847 - val_acc: 0.8379\n",
            "Epoch 19/100\n",
            "132/132 [==============================] - 5s 35ms/step - loss: 0.1063 - acc: 0.9613 - val_loss: 0.5242 - val_acc: 0.8470\n",
            "Epoch 20/100\n",
            "132/132 [==============================] - 4s 29ms/step - loss: 0.1012 - acc: 0.9613 - val_loss: 0.5298 - val_acc: 0.8318\n",
            "Epoch 21/100\n",
            "132/132 [==============================] - 4s 29ms/step - loss: 0.0620 - acc: 0.9746 - val_loss: 0.5870 - val_acc: 0.8500\n",
            "Epoch 22/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.0529 - acc: 0.9799 - val_loss: 0.5098 - val_acc: 0.8394\n",
            "Epoch 23/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0740 - acc: 0.9765 - val_loss: 0.5220 - val_acc: 0.8379\n",
            "Epoch 24/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 0.0565 - acc: 0.9810 - val_loss: 0.7637 - val_acc: 0.8333\n",
            "Epoch 25/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 0.0424 - acc: 0.9856 - val_loss: 0.6204 - val_acc: 0.8455\n",
            "Epoch 26/100\n",
            "132/132 [==============================] - 4s 28ms/step - loss: 0.0513 - acc: 0.9863 - val_loss: 0.6836 - val_acc: 0.8470\n",
            "Epoch 27/100\n",
            "132/132 [==============================] - 4s 34ms/step - loss: 0.0238 - acc: 0.9924 - val_loss: 0.7600 - val_acc: 0.8470\n",
            "Epoch 28/100\n",
            "132/132 [==============================] - 5s 36ms/step - loss: 0.0088 - acc: 0.9981 - val_loss: 0.8438 - val_acc: 0.8545\n",
            "Epoch 29/100\n",
            "132/132 [==============================] - 4s 32ms/step - loss: 0.0027 - acc: 1.0000 - val_loss: 0.8853 - val_acc: 0.8455\n",
            "Epoch 30/100\n",
            "132/132 [==============================] - 4s 29ms/step - loss: 0.0019 - acc: 1.0000 - val_loss: 0.9679 - val_acc: 0.8500\n",
            "Epoch 31/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 8.7406e-04 - acc: 1.0000 - val_loss: 0.9608 - val_acc: 0.8485\n",
            "Epoch 32/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 5.9407e-04 - acc: 1.0000 - val_loss: 1.0012 - val_acc: 0.8561\n",
            "Epoch 33/100\n",
            "132/132 [==============================] - 4s 27ms/step - loss: 4.8757e-04 - acc: 1.0000 - val_loss: 1.0394 - val_acc: 0.8545\n",
            "Epoch 34/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.7979e-04 - acc: 1.0000 - val_loss: 1.0562 - val_acc: 0.8545\n",
            "Epoch 35/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.2436e-04 - acc: 1.0000 - val_loss: 1.0956 - val_acc: 0.8576\n",
            "Epoch 36/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.6192e-04 - acc: 1.0000 - val_loss: 1.1012 - val_acc: 0.8530\n",
            "Epoch 37/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.3704e-04 - acc: 1.0000 - val_loss: 1.1184 - val_acc: 0.8545\n",
            "Epoch 38/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.0107e-04 - acc: 1.0000 - val_loss: 1.1480 - val_acc: 0.8515\n",
            "Epoch 39/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.7552e-04 - acc: 1.0000 - val_loss: 1.1717 - val_acc: 0.8515\n",
            "Epoch 40/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 1.5106e-04 - acc: 1.0000 - val_loss: 1.1937 - val_acc: 0.8515\n",
            "Epoch 41/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.3415e-04 - acc: 1.0000 - val_loss: 1.2048 - val_acc: 0.8545\n",
            "Epoch 42/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.1858e-04 - acc: 1.0000 - val_loss: 1.2234 - val_acc: 0.8515\n",
            "Epoch 43/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.0491e-04 - acc: 1.0000 - val_loss: 1.2192 - val_acc: 0.8530\n",
            "Epoch 44/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 9.4263e-05 - acc: 1.0000 - val_loss: 1.2527 - val_acc: 0.8515\n",
            "Epoch 45/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 8.3262e-05 - acc: 1.0000 - val_loss: 1.2578 - val_acc: 0.8530\n",
            "Epoch 46/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 7.3590e-05 - acc: 1.0000 - val_loss: 1.2969 - val_acc: 0.8515\n",
            "Epoch 47/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 6.7076e-05 - acc: 1.0000 - val_loss: 1.2920 - val_acc: 0.8515\n",
            "Epoch 48/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 6.1125e-05 - acc: 1.0000 - val_loss: 1.3063 - val_acc: 0.8515\n",
            "Epoch 49/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 5.4624e-05 - acc: 1.0000 - val_loss: 1.3119 - val_acc: 0.8500\n",
            "Epoch 50/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 4.8974e-05 - acc: 1.0000 - val_loss: 1.3119 - val_acc: 0.8515\n",
            "Epoch 51/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 4.4869e-05 - acc: 1.0000 - val_loss: 1.3382 - val_acc: 0.8515\n",
            "Epoch 52/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 4.0588e-05 - acc: 1.0000 - val_loss: 1.3651 - val_acc: 0.8515\n",
            "Epoch 53/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.6645e-05 - acc: 1.0000 - val_loss: 1.3574 - val_acc: 0.8500\n",
            "Epoch 54/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.3426e-05 - acc: 1.0000 - val_loss: 1.3824 - val_acc: 0.8485\n",
            "Epoch 55/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.0720e-05 - acc: 1.0000 - val_loss: 1.3822 - val_acc: 0.8515\n",
            "Epoch 56/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.7604e-05 - acc: 1.0000 - val_loss: 1.3971 - val_acc: 0.8500\n",
            "Epoch 57/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.5583e-05 - acc: 1.0000 - val_loss: 1.4074 - val_acc: 0.8500\n",
            "Epoch 58/100\n",
            "132/132 [==============================] - 4s 28ms/step - loss: 2.3393e-05 - acc: 1.0000 - val_loss: 1.4186 - val_acc: 0.8500\n",
            "Epoch 59/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 2.1037e-05 - acc: 1.0000 - val_loss: 1.4340 - val_acc: 0.8515\n",
            "Epoch 60/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 1.9342e-05 - acc: 1.0000 - val_loss: 1.4423 - val_acc: 0.8515\n",
            "Epoch 61/100\n",
            "132/132 [==============================] - 3s 23ms/step - loss: 1.7875e-05 - acc: 1.0000 - val_loss: 1.4515 - val_acc: 0.8515\n",
            "Epoch 62/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.6460e-05 - acc: 1.0000 - val_loss: 1.4580 - val_acc: 0.8515\n",
            "Epoch 63/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.4812e-05 - acc: 1.0000 - val_loss: 1.4729 - val_acc: 0.8515\n",
            "Epoch 64/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.3520e-05 - acc: 1.0000 - val_loss: 1.4722 - val_acc: 0.8500\n",
            "Epoch 65/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.2764e-05 - acc: 1.0000 - val_loss: 1.4984 - val_acc: 0.8500\n",
            "Epoch 66/100\n",
            "132/132 [==============================] - 4s 28ms/step - loss: 1.1321e-05 - acc: 1.0000 - val_loss: 1.4991 - val_acc: 0.8500\n",
            "Epoch 67/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.0369e-05 - acc: 1.0000 - val_loss: 1.5110 - val_acc: 0.8530\n",
            "Epoch 68/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 9.7004e-06 - acc: 1.0000 - val_loss: 1.5130 - val_acc: 0.8515\n",
            "Epoch 69/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 8.9984e-06 - acc: 1.0000 - val_loss: 1.5336 - val_acc: 0.8530\n",
            "Epoch 70/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 8.2901e-06 - acc: 1.0000 - val_loss: 1.5324 - val_acc: 0.8515\n",
            "Epoch 71/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 7.5910e-06 - acc: 1.0000 - val_loss: 1.5476 - val_acc: 0.8530\n",
            "Epoch 72/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 7.0326e-06 - acc: 1.0000 - val_loss: 1.5675 - val_acc: 0.8530\n",
            "Epoch 73/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 6.4481e-06 - acc: 1.0000 - val_loss: 1.5682 - val_acc: 0.8515\n",
            "Epoch 74/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 5.8305e-06 - acc: 1.0000 - val_loss: 1.5759 - val_acc: 0.8515\n",
            "Epoch 75/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 5.4708e-06 - acc: 1.0000 - val_loss: 1.5880 - val_acc: 0.8515\n",
            "Epoch 76/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 5.0035e-06 - acc: 1.0000 - val_loss: 1.6000 - val_acc: 0.8530\n",
            "Epoch 77/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 4.6736e-06 - acc: 1.0000 - val_loss: 1.6051 - val_acc: 0.8515\n",
            "Epoch 78/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 4.3311e-06 - acc: 1.0000 - val_loss: 1.6156 - val_acc: 0.8515\n",
            "Epoch 79/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.9207e-06 - acc: 1.0000 - val_loss: 1.6337 - val_acc: 0.8485\n",
            "Epoch 80/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 3.7231e-06 - acc: 1.0000 - val_loss: 1.6364 - val_acc: 0.8500\n",
            "Epoch 81/100\n",
            "132/132 [==============================] - 4s 30ms/step - loss: 3.3621e-06 - acc: 1.0000 - val_loss: 1.6479 - val_acc: 0.8485\n",
            "Epoch 82/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 3.1356e-06 - acc: 1.0000 - val_loss: 1.6584 - val_acc: 0.8470\n",
            "Epoch 83/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.8784e-06 - acc: 1.0000 - val_loss: 1.6648 - val_acc: 0.8485\n",
            "Epoch 84/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 2.6399e-06 - acc: 1.0000 - val_loss: 1.6698 - val_acc: 0.8500\n",
            "Epoch 85/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.4625e-06 - acc: 1.0000 - val_loss: 1.6774 - val_acc: 0.8485\n",
            "Epoch 86/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.2933e-06 - acc: 1.0000 - val_loss: 1.6853 - val_acc: 0.8485\n",
            "Epoch 87/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 2.1187e-06 - acc: 1.0000 - val_loss: 1.6950 - val_acc: 0.8485\n",
            "Epoch 88/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 1.9488e-06 - acc: 1.0000 - val_loss: 1.6988 - val_acc: 0.8500\n",
            "Epoch 89/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 1.8140e-06 - acc: 1.0000 - val_loss: 1.7215 - val_acc: 0.8470\n",
            "Epoch 90/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 1.6593e-06 - acc: 1.0000 - val_loss: 1.7207 - val_acc: 0.8485\n",
            "Epoch 91/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.5372e-06 - acc: 1.0000 - val_loss: 1.7276 - val_acc: 0.8485\n",
            "Epoch 92/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.4339e-06 - acc: 1.0000 - val_loss: 1.7392 - val_acc: 0.8485\n",
            "Epoch 93/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.3089e-06 - acc: 1.0000 - val_loss: 1.7472 - val_acc: 0.8470\n",
            "Epoch 94/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.2041e-06 - acc: 1.0000 - val_loss: 1.7639 - val_acc: 0.8470\n",
            "Epoch 95/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.1372e-06 - acc: 1.0000 - val_loss: 1.7792 - val_acc: 0.8470\n",
            "Epoch 96/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 1.0391e-06 - acc: 1.0000 - val_loss: 1.7798 - val_acc: 0.8470\n",
            "Epoch 97/100\n",
            "132/132 [==============================] - 3s 24ms/step - loss: 9.6105e-07 - acc: 1.0000 - val_loss: 1.7836 - val_acc: 0.8485\n",
            "Epoch 98/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 8.8503e-07 - acc: 1.0000 - val_loss: 1.7881 - val_acc: 0.8485\n",
            "Epoch 99/100\n",
            "132/132 [==============================] - 4s 28ms/step - loss: 8.3854e-07 - acc: 1.0000 - val_loss: 1.7947 - val_acc: 0.8500\n",
            "Epoch 100/100\n",
            "132/132 [==============================] - 3s 25ms/step - loss: 7.6314e-07 - acc: 1.0000 - val_loss: 1.8117 - val_acc: 0.8470\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "EVALUASI MODEL 1\n"
      ],
      "metadata": {
        "id": "GKFD0Xbi8YMi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "%matplotlib inline\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(np.arange(0, 100), H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Loss Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "xsJHYZG48Ym6",
        "outputId": "bcc70edc-d063-47f4-e71d-aea513928b4f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEaCAYAAADg2nttAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXhU5fXA8e+dmeyTbWayEAgEwhqQzQCyFIFEwAWlSKG1WBX8VcVqxWoFtEJVEEUUbaFuCHVp1aJo0aISdogoGMMOJuyQkG2yJ5Nk5t7fH5HRSCAhJjPJzPk8Dw+Zu8w9JzeTk/u+976vommahhBCCHEJOncHIIQQovWTYiGEEKJBUiyEEEI0SIqFEEKIBkmxEEII0SApFkIIIRokxUKINmLUqFHceeed7g5DeCkpFsIr3X777SQnJ7s7DKdVq1ahKIrzX1RUFDfccAP79u37We9rMBhYtWpV8wQpvJoUCyFaCb1eT3Z2NtnZ2Xz00Ufk5uYybtw4iouL3R2aEFIshKjPkSNHuP766zEajRiNRiZMmEBmZqZzfUlJCXfccQfR0dH4+fkRGxvLgw8+6Fy/fft2hg8fTnBwMMHBwfTr14/PP/+8weNGR0cTHR3N0KFDeeGFF8jOzmbnzp31bltTU8Ps2bNp3749vr6+JCQk8K9//cu5Pi4uDofDwR133OG8YhGiqaRYCPETlZWVjB07FpvNxpYtW9iyZQtlZWWMHz+e6upqAB577DHS0tL4+OOPycjI4L333qNXr14A2O12brzxRoYMGUJaWhppaWnMnz+fwMDAy4ojICAAqC0K9Zk7dy6vvfYaS5cuZf/+/UybNo1p06axYcMGAHbt2oVer2fp0qXOKxYhmsrg7gCEaG3+9a9/kZeXxzfffIPFYgHg3XffJS4ujnfffZff/e53nDx5kgEDBjBkyBAAOnbsyLBhwwAoLS2lsLCQG2+8kW7dugE4/2+svLw85s2bR0hICIMHD75gfUVFBS+99BIvvPACv/rVr4Da4rFr1y4WLFhAUlISERERAISGhhIdHd20b4YQ35MrCyF+4sCBAyQkJDgLBUBUVBQ9evTgwIEDAMycOZPVq1fTp08f/vjHP7Ju3TpUVQUgPDycO++8k3HjxnHttdeyaNEijhw50uBxHQ6Hs9krMjKSzMxMVq9eTWRk5AXbZmZmUl1dzciRI+ssv/rqq50xCtGcpFgI0QTjxo3j1KlTPProo9hsNqZNm8aYMWNwOBwAvPbaa3zzzTdcc801bNmyhT59+vDKK69c8j31ej3p6ens2bOHkpISDh06xDXXXOOKdIRokBQLIX6id+/eHDx4kPz8fOeynJwcjhw5Qp8+fZzLTCYTv/nNb3jllVf49NNP2bJlCwcPHnSu79OnDw8++CDr1q1jxowZvPrqqw0eu2vXrsTHxxMcHNzgdn5+fmzdurXO8vOF6TxfX19nARPi55A+C+G1ysrKSE9Pr7PM39+fW265hSeeeIKpU6eyePFiNE3joYceon379kydOhWARx99lCuvvJLevXuj0+l45513MBqNdOzYkczMTF577TUmTJhAbGwsWVlZbNu2jYEDBzZb7IGBgdx///385S9/ISIign79+rF69Wo+/vhj1q9f79yuc+fObNq0iWuvvRZfX986TWtCXA4pFsJrffXVVwwYMKDOsh49enD48GG++OILZs2a5ewTGDVqFJ999hm+vr5AbVF5/PHHOXHiBHq9nv79+7Nu3TpCQ0OpqKggIyODX//61+Tl5WE2m7n++ut57rnnmjX+BQsWoNPpeOCBB8jLy6Nr1668/fbbJCUlObdZsmQJs2bNIi4ujpqaGmSuM9FUisyUJ4QQoiHSZyGEEKJBUiyEEEI0SIqFEEKIBkmxEEII0SApFkIIIRrk0bfOZmVlNWk/i8VS54Esb+CNOYN35u2NOYN35n25OcfExFx0nVxZCCGEaJAUCyGEEA2SYiGEEKJBHt1n8VOapmGz2VBV9ZKzhuXk5FBVVeXCyNyvsTlrmoZOp8Pf319mXhPCi3hVsbDZbPj4+GAwXDptg8GAXq93UVStw+XkbLfbsdlszpnchBCez6uaoVRVbbBQiIYZDAbnRD9CCO/gVcVCmk2aj3wvhfAuXlUshBDCkx3IqeDDAwUt8t7SJiOEEG1cYaWdVWm5bD5RQrTRh+t6hONvaN5rAbmycKHi4mJWrVp12fvdeuutFBcXX/Z+DzzwAJ988sll7yeEaP1Kqhx8daaUN77JYebaY2w/Vcqvept56frOzV4oQK4sXKqkpIQ333yT22+/vc5yu91+yY73t956q4UjE0K0NpqmYVc1rJV2cspqOFdWw9mSas4UV3GmpJpzZTUAGHQKA9oFccfASNqH+LZYPF5bLNR3X0M7fbz+dYrSpOknldjO6H79fxddv3DhQk6ePMk111yDj48Pfn5+hIaGkpmZyfbt25k+fTpZWVlUVVUxY8YMpk2bBsCQIUNYt24d5eXlTJs2jcGDB7N7926io6N54403GnUL67Zt23jyySdxOBz069ePp59+Gj8/PxYuXMgXX3yBwWBg5MiRPP7446xdu5YXXngBnU5HSEgIH3744WV/L4QQl8ehamw5UcJ7+/LJK6/BUc+vIB+dQkyIL/Emf5LjQ0mIDKSb2R9ffcs3ErmkWCxfvpy0tDRCQ0NZsmTJBev/+9//sm3bNqD29tYzZ86wYsUKjEYj9957L/7+/uh0OvR6PYsWLXJFyC1i7ty5HDlyhPXr15Oamsrvfvc7Nm7cSMeOHYHa+ZLDw8OprKzk+uuv57rrrsNkMtV5j+PHj7Ns2TIWL17MXXfdxf/+9z9uvvnmSx7XZrMxa9Ys3nvvPeLj47n//vt58803ufnmm1m3bh1bt27Fx8eHgoLajrGlS5fyzjvv0K5duyY1fwkhLk9aVhmrvs3jZFEVXU3+jOgUgkEHep1CmL+BaKMP0UZfzIEG9Dr33InokmIxatQoxo8fz7Jly+pdf+ONN3LjjTcCsHv3bj799FOMRqNz/bx58wgJCWnWmC51BWAwGLDb7c16vPr079/fWSgA3njjDdatWwfUjph7/PjxC4pFbGwsffr0AaBv376cPn26weMcPXqUjh07Eh8fD8CvfvUr/vnPf3LHHXfg5+fHn/70J8aNG8fo0aMBSExMZNasWUyYMIFrr722WXIVQlwou7Sa13fnsDurnGijDw8Nj2F4p2B0rfDWdJcUi4SEBHJzcxu17Y4dOxg+fHgLR9Q6BAYGOr9OTU1l27ZtrF27loCAACZPnlzv8Bt+fn7Or/V6PTabrcnHNxgMfPrpp2zfvp3//e9/vP766/znP//hmWeeIS0tjQ0bNnDttdeybt26C4qWEKJpNE3jZFEVW06U8N/DhRh0CrcPiOCGHiZ89K2vSJzXqvosqqqqSE9PZ8aMGXWWL1iwAIBrrrmG5OTki+6fkpJCSkoKAIsWLcJisdRZn5OT0+gnuFviSe/Q0FDKy8udQ2soiuI8Tnl5OWFhYQQHB5ORkUFaWhp6vR6DwYCiKOj1eudwHOf30el06HS6i8Z6vumuR48enDlzhtOnT9O5c2fWrFnDsGHDqKqqoqKignHjxjF06FAGDx6MwWDgxIkTDB48mMGDB7N582Zyc3OJjIys895+fn4XfH/bIoPB4BF5XA5vzBncm7emaRw4V8qnB3P48kQheWXVAIztEcHMEXFEGP0aeIemac6cW1Wx+Oabb+jRo0edJqgnn3wSk8lEcXExTz31FDExMSQkJNS7f3Jycp1i8tNJP6qqqho1/lFLNUOFhISQmJjIyJEj8ff3x2KxOI8zcuRI/vnPfzJ8+HDi4+MZOHAgDocDu92Opmk4HA4cDgeAcx9VVVFV9aKxqqqKw+HAYDCwZMkSZsyY4ezg/u1vf0tRURHTp0+nqqoKTdN4/PHHsdvtzJ8/n+PHj6NpGiNGjKBHjx4XHKOqqsojJpKRCXG8hzvyziuv4ZusMj7LKOJ4YRX+Bh0D2gUxtbeJgTFBmAN9wFZKvq20RY7fnJMfKVpTbvtpgtzcXJ555pl6O7jPW7x4MUOHDmXEiBH1rn///ffx9/d39m805Kcz5VVUVNRp+rkYV/VZtCaXm3Njv5etnTf+4vTGnKFl8z6UV8G674pQlNo7lmocGofyK8n5/vbWzuF+jO8Wxsi4EAJ9XDdIaXMWi1ZzZVFRUcHBgwe57777nMtsNhuaphEQEIDNZmPv3r1MnjzZjVEKIURdn2UU8truHAIMOgJ89NSoGgrQ3eLPhB7h9IkKJC7Mr82Pp+aSYrF06VIOHjxIaWkpd999N1OmTHH+FTt27FgAvv76a/r164e/v79zv+LiYp577jkAHA4HI0aMoH///q4IuU2ZO3cuu3btqrPszjvvZOrUqW6KSAjPpmkaRTYH/96bz+eZRQxsF8Sfhsdg9PPcqQ1c1gzlDtIM1XjSDOU9vDFn+Hl5q5rGodxKtp0s4WBuJefKqqn6/qm5SQkmpvWLcNvzD5fikc1QQgjRmlTWqOzPqeCbrDK+PlNGQaUdX73CFVGB9G0XSDtj7ZPUPSO8YxIwKRZCCEHt1cMxaxXp58rZk13OwbxK7KqGn16hf7sgbu8UwqD2RgJ8vHP8VSkWQgiv41BrH4w7nF/JMauNk0VVnCquwmavbVqKC/Pjhh7hDIwJIiEiAB8XjL3U2kmxEEJ4hSq7g60nSthyvJj9uZXY7LVTAwf76ekU5kdSfBjdzf70iw4iPEB+Nf6UfEdasW7dupGRkVHvutOnT3PbbbexceNGF0clRNtyzGrj88witp/MoKzagSXQwOjOIfSMCKBXRACRQT5t/rZWV5BiIYTwGGVVDvIqaiistHOurIZNx4r5rsCGr15hdDcLw9v7c0VUYKscqK+189pi8fruHI4X1j8In9LE+Sw6h/tzZ2LURdcvXLiQmJgY5+RHS5YsQa/Xk5qaSnFxMXa7nT//+c+MGzfuso5rs9mYM2cOe/fuRa/XM2/ePIYPH86RI0d48MEHqa6uRtM0Xn31VaKjo7nrrrvIzs5GVVX++Mc/ctNNN112rkK0JlV2lXf35fPRISvqjz66HUJ8ufPKSEZ3DiWufZRX3jLcXLy2WLjDjTfeyLx585zFYu3atbzzzjvMmDGD4OBgrFYrEyZMYOzYsZd1Wbxq1SoURWHDhg1kZmbym9/8hm3btvHWW28xY8YMJk2aRHV1NQ6Hg40bNxIdHe2cfa+kpKQlUhXCZfbllLPsq3Nkl9aQ1CWUxPZBhPsbCAuonQdCmpiah9cWi0tdAbTUQ3l9+vQhPz+fc+fOUVBQQGhoKJGRkcyfP5+vvvoKRVE4d+4ceXl5F4zyeim7du3ijjvuAKBr16506NCBY8eOceWVV/LSSy+RnZ3NtddeS5cuXejZsydPPPEECxYsIDk5mSFDhjR7nkK0tLIqB9tOlrDhWDEZBTaijT48mRRL3+ggd4fmsby2WLjLDTfcwKeffkpubi433ngjH374IQUFBaxbtw4fHx+GDBlS7zwWTfHLX/6SAQMGsGHDBm699VaeeeYZRowYwWeffcbGjRt59tlnGTFiBLNmzWqW4wnRUnLLavjydCknimycKKy9zdWuQqcwP6YPjGR8tzD8DHJ7a0uSYuFiN954Iw8//DBWq5UPPviAtWvXYrFY8PHxYceOHZw5c+ay33Pw4MGsWbOGESNGcPToUc6ePUt8fDwnT56kU6dOzJgxg7Nnz3Lo0CG6du1KWFgYN998MyEhIfz73/9ugSyF+Pk0TWNfTgWfflfI12fKUDUIDzAQF+bHTe2CGNEphM7hbX+AvrZCioWL9ejRg/LycqKjo4mKimLSpEncdtttJCUl0bdvX7p27XrZ73nbbbcxZ84ckpKS0Ov1vPDCC/j5+bF27Vo++OADDAYDkZGR3HfffezZs4ennnoKRVHw8fHh6aefboEshWg6m11ly/ESPj1SyMniKoL99ExKMDOuaxiRRh93h+e1ZCDBeshAgg2TgQTbrtaYc0mVgz3Z5aSfK2fn6VLKqlU6h9c+Rf2LTiHN0sTUGvNuaTKQoBCizaqocbDim1z2ZJdjVzXsqkZZtYoGGH11DGxn5NruYfSKCJAmplZEikUrd+jQIe6///46y/z8/Pjkk0/cFJEQTXcot4LnU7PJr6hhWMdggnz0GHQQ5m+gX7sgupr8W+VQ38LLikVbbHHr1asX69evd3cYF2iL30vhPmXVDlbvL+Djw1Yig3x4+ppOXjO0t6fwqmKh0+mw2+0YDF6VdrOz2+3odHKboqhfWZWD0moHfgYdPjqFjceK+c/+fMqqVZLiQ5lxZaRL56EWzcOrfmv6+/tjs9moqqq6ZFuon59fsz3r0FY0NmdN09DpdHWmvxXivM3Hi/nH1znOEV3PG9AuiN/1j6CLSX5u2iqXFIvly5eTlpZGaGgoS5YsuWD9gQMHePbZZ51PLQ8ZMoTJkycDkJ6ezsqVK1FVlaSkJCZOnNjkOBRFISCg4UtfuWtCiMtTWaPyyq5zbDpeQkJEANd0DaPKrlLlUIk3+XNFlDxZ3da5pFiMGjWK8ePHs2zZsotu06tXL2bPnl1nmaqqrFixgsceewyz2cycOXNITEykQ4cOLR2yEOIizj8st+FYMbllNZRVOyiotFNZo/LrK8xM6WORTmoP5JJikZCQQG5u7mXvl5mZ6Xx4DWDYsGHs2rVLioUQLqJqGnvOVVBss2NXNUpsDjYcK+ZMSTXBvjo6hfkRE+JLz4gARnUOpXdk23/2RtSv1fRZfPfddzz88MOEh4dz6623Ehsbi9VqxWw2O7cxm80XnQwIICUlhZSUFAAWLVqExWJpUiwGg6HJ+7ZV3pgzeGfejc3Zrmo8k5LB/w7V/UOvZ6SRuUM6kdzdgp+h7XRUy7n+me/VLO/yM3Xu3Jnly5fj7+9PWloaixcv5qWXXrrs90lOTiY5Odn5uqlt8N7Yfu+NOYN35t2YnGscKs/tyGLn6TKmXmHm6rhQfHQKvgaFMP/aXxulRYWUuiLgZiLnumGt/gnuHw8bMXDgQFasWEFJSQkmk4mCggLnuoKCAkwmkztCFMKjnSi0se1k7a9+H73CvnPl7M+t5M4rI5nQUz5zopUUi6KiIkJDQ1EUhczMTFRVJTg4mKCgILKzs8nNzcVkMpGamnrB08xCiKY7V1rNv/fms+VECefvJlc18NUr/HFoO8Z0CXVvgKLVcEmxWLp0KQcPHqS0tJS7776bKVOmOAetGzt2LDt37uSLL75Ar9fj6+vLAw88gKIo6PV6pk+fzoIFC1BVldGjRxMbG+uKkIXwSDa7ypcnrHyVmcehvAoO5VWi1yn8MsHEpAQzwX56HKqGBhjkjibxI1416mxjSdum9/CGvMurHXyTVU7qqRK+ySqn2qGhU6BLuD9XRAUyoWc45kDPH/rbG871T3lcn4UQonkVVNSw/WQpu8+WcSC3Asf3Ewclx4cytncH2vlW4y8zy4nLIMVCCA9SXu3ggwMFrD1SSLVDo1OoHxN7mRjU3kiPiAB0ioLFEuZ1f2GLn0+KhRAeoLDSzoajxXx02EpplYNRcSFMvcJCTIivu0MTHkKKhRBtlEPV2HOunPVHi/nqdCkODfq3C+I2GbBPtAApFkK0MaeKq9hwtJgtJ0oorLQT7KdnQk8T18SH0iHUz93hCQ8lxUKINiK7tJp/7c1n24kS9Dq4MsbIqM4hDGpvxEcvndWiZUmxEKIVstlV3tmTR0FF7fNINarGN2fLnM9ETOxlItRfPr7CdeSnTYhWpshm56nNZzhqtRETXNtBrQFju4Yx5QoLpgD52ArXk586Idwou7SaN9PzMAcaGBAdhDnQwNNbz2KttDN7ZHuGdAh2d4hCAFIshHCb08VVPL7hNBU1Kg5VY+3hQgCC/fQ8ldyRHpaGZ3UUwlWkWAjhBsesNuZtPI1egWfHdSLa6MOhvEoyrTaGdwymXbA8HyFaFykWQrjY4bxKnth8mgCDjieTOjofnOvfLoj+7WSuatE6SbEQwoXSs8tZuOUMpkADT4zpSKTR8wfwE55BioUQLpJ6qoQlO7KIDfVj/uhYwuSuJtGGyE+rEM2osNJO6qlSMq02jhbYyCqtpnZ2CLCr0NMSwF9Gd8Do23bmrhYCpFgI0STWSjsnCm3EhvphCTRQXOVgzUEr//uudrTXUH89XU3+DIgJck4iFOij4/oe4TI0uGiTpFgIcZmq7CrzN5zmZHEVUFsE7KqGXdW4Oi6Eyb3NtA/xRVFkpjnhOaRYCHGZ/vltLieLq7h7UBQAJ4tqi8aEnibay5DgwkO5pFgsX76ctLQ0QkNDWbJkyQXrt23bxscff4ymaQQEBHDnnXcSFxcHwL333ou/vz86nQ69Xs+iRYtcEbIQ9fr6TCmfflfEhJ7hXNs93N3hCOEyLikWo0aNYvz48Sxbtqze9ZGRkcyfPx+j0ci3337Lq6++ysKFC53r582bR0hIiCtCFeKiCipqeGnnOTqH+3Fb/wh3hyOES7mkWCQkJJCbm3vR9T169HB+3a1bNwoKClwRlhANSs8uZ8uJEk4U2jhVXI1OgT8Nj5EhwYXXaXV9Fhs3bmTAgAF1li1YsACAa665huTkZHeEJbzQgZwKntx8mkAfPV1M/kyIDmJox2BiZYIh4YVaVbHYv38/mzZt4oknnnAue/LJJzGZTBQXF/PUU08RExNDQkJCvfunpKSQkpICwKJFi7BYLE2Kw2AwNHnftsobc4aL532mqJJntmcSExrAK1P6EeJBc0fIufYezZlzq/kEnDx5kldeeYU5c+YQHPzDsMwmkwmA0NBQBg0aRGZm5kWLRXJycp0rj/z8/CbFYrFYmrxvW+WNOcMPeZ8otFFpV4kJ9kWvU3jk85OoqsrcX7SjuqyI/DJ3R9p8vP1ce5PLzTkmJuai61pFscjPz+e5557jD3/4Q51gbTab8w4pm83G3r17mTx5shsjFZ6mvNrOy1+fY11GkXNZ7UN0Gn8d01FGfxXiey4pFkuXLuXgwYOUlpZy9913M2XKFOz22ukix44dy+rVqykrK+P1118HcN4iW1xczHPPPQeAw+FgxIgR9O/f3xUhCw+naRppWeW8+t/j5JRWMaFnOP2jg8gqreZcaTVXxhjpExXo7jCFaDUUTdM0dwfRUrKyspq0n1yueq5im52Nx4pZf7SYsyXVdAwPYOagCHpFeE9h8JZz/VPemLfHNUMJ4QpbT5Tw953ZVDk0ekUEcPNV0Uy8sgulRVZ3hyZEqyfFQni8GofGqm9z+eRIIQkRAdwzOJqOYbW3v/oZdJS6OT4h2gIpFsKjFVXaWbTtLIfyKrmxZzi3DYh0jgIrhGg8KRbCY50otPHU5jOUVDl4aHgMv4iTIWOEaCopFsIjfX2mlCU7sgny0fH02E7Em/zdHZIQbZoUC+FxNh8vZmlqNvEmf+Ze3R5zoMxzLcTPJcVCeJQvT5fy4pfZ9IkK5C+jOuAns9IJ0SzkkyQ8xrfZ5Ty3PYuu319RSKEQovnIlYVo81RNY31mMa9/k0NsqC/zRscS6KN3d1hCeBQpFqJNO2a18Y+vz/FdgY0+UYE8PCIGo58UCiGamxQL0aZU1qhsOl5MRoGNo1Ybp4qqCPHXM2tYO66OC0FR5BkKIVqCFAvRZmiaxvOpWXx9poxQfz1dTf4M7xjM9T3CMfrK1YQQLUmKhWgzPsso4uszZdwxMIKbeprkKkIIF5LbRUSbcLq4ijfScunfLogbpVAI4XJSLESrV+NQWbIjC3+Djj8ObYdOCoUQLifNUKJVO1lUxctfn+N4YRWPXd0BU4D8yArhDvLJE61SRY2Df+3N59MjhQT56LjvqmgGdTC6OywhvJYUC9HqOFSNhVvOsj+ngrFdw5jWP4IQeXZCCLeSYiFanf8cKGBfTgX3XRVNcnyYu8MRQuDCYrF8+XLS0tIIDQ1lyZIlF6zXNI2VK1fy7bff4ufnx8yZM+nSpQsAmzdv5sMPPwRg0qRJjBo1ylVhCxfbn1PBe/vyGRUXQlKXUHeHI4T4nsvuhho1ahRz58696Ppvv/2Wc+fO8dJLL/H73/+e119/HYCysjJWr17NwoULWbhwIatXr6asrMxVYQsXKrbZWbIji2ijL3cNjpLbY4VoRVxWLBISEjAaL95BuXv3bkaOHImiKHTv3p3y8nIKCwtJT0+nb9++GI1GjEYjffv2JT093VVhCxfRNI2/7cymtMrBwyNiZCBAIVqZVtNnYbVasVgsztdmsxmr1YrVasVsNjuXm0wmrFZrve+RkpJCSkoKAIsWLarzfpfDYDA0ed+2yt05f7zvHLvOlnP/yM4M7t7eZcd1d97u4I05g3fm3Zw5N7pY7N+/n8jISCIjIyksLOSdd95Bp9Nxyy23EBbWOjohk5OTSU5Odr7Oz89v0vtYLJYm79tWuTPn7NJqXtxynL7RgYzu4OvSOORcew9vzPtyc46JibnoukY3Q61YsQKdrnbzN998E4fDgaIovPLKK40O5FJMJlOdpAoKCjCZTJhMJgoKCpzLrVYrJpOpWY4p3COjoJIvT5VSYrPjUDVeSM3CoFfk6WwhWrFGX1mcbyZyOBzs2bOH5cuXYzAYuOuuu5olkMTERD777DOGDx9ORkYGgYGBhIeH079/f/797387O7X37NnDLbfc0izHFK7nUDUWbDlLYaUdAEuggfwKO38aHoNF5soWotVqdLEICAigqKiI06dP06FDB/z9/bHb7djt9kbtv3TpUg4ePEhpaSl33303U6ZMce47duxYBgwYQFpaGvfffz++vr7MnDkTAKPRyM0338ycOXMAmDx58iU7ykXrdjCvgsJKO7f0taBXFPblVjC8YzAj40LcHZoQ4hIaXSzGjx/PnDlzsNvt3H777QAcPnyY9u0b1xn5wAMPXHK9oijceeed9a4bM2YMY8aMaWyoohXbfrIUP73CTb1M+Bt0TMbc8E5CCLdrdLGYOHEigwcPRqfTER0dDdT2M9x9990tFpzwLA5VI/VUKYM6GPE3yIDHQrQll3Xr7I97yvfv349OpyMhIaHZgxKeaW9OBSVVDn7RSZqchGhrGv3n3bx58+0qq5QAACAASURBVDh8+DAAH330ES+++CIvvviicxgOIRqy/WQJgT46BsYEuTsUIcRlanSxOH36NN27dwdgw4YNzJs3jwULFrB+/foWC054jhqHxpenSxncwYivXpqghGhrGt0MpWkaAOfOnQOgQ4cOAJSXl7dAWMLTpGeXU16tShOUEG1Uo4tFjx49eOONNygsLGTQoEFAbeEIDg5useCE59h+sgSjr45+0dIEJURb1Oj2gHvvvZfAwEA6derElClTAMjKyuK6665rseCEZzhRaGP7qRJGdArBRy9PaAvRFjX6yiI4OPiCJ6cHDhzY7AEJz1Lj0Fj6ZTZGXz2/7etdg7gJ4UkaXSzsdjsffvghW7dupbCwkPDwcEaOHMmkSZMwGFrN4LWilXl/fz7HC6uYe3V7Qvzl50SItqrRn963336bo0eP8n//939ERESQl5fHBx98QEVFhfOJbiF+7Lv8SlYfKGBMl1CGdJC+LSHaskYXi507d7J48WJnh3ZMTAydO3fm4YcflmIhLnB+MiNTgIE7r4x0dzhCiJ+p0R3c52+dFaIxMgpsnCqu5jd9LQT5yqx3QrR1jb6yGDp0KM888wyTJ092TqjxwQcfMHTo0JaMT7RRO06VYtDBVbHS/CSEJ2h0sZg2bRoffPABK1asoLCwEJPJxLBhwxo9RLnwHpqmsf1kCQPaBWGUqwohPEKji4XBYGDq1KlMnTrVuay6uppbb72VadOmtUhwom36rsBGfoWdaf0i3B2KEKKZ/KxBehSZAlPUY/vJEgw6hcEdZJIqITyFjOgmmpWqaew4VcrAmCDp2BbCgzTYDLV///6LrpP+CvFTR/IqKaiw87v+0rEthCdpsFj84x//uOR6i6VxQzikp6ezcuVKVFUlKSmJiRMn1lm/atUqDhw4ANT2hRQXF7Nq1SoApk6dSseOHZ3He+SRRxp1TNG8yqodDXZYbz9Vio80QQnhcRosFsuWLfvZB1FVlRUrVvDYY49hNpuZM2cOiYmJzmHOgToP9q1bt47jx487X/v6+rJ48eKfHYdouuOFNh5cd4LF4+Loavavd5vz06Ze2T6IQB9pghLCk7ikzyIzM5Po6GiioqIwGAwMGzaMXbt2XXT7HTt2MGLECFeEJhrpZFEVqgYHcisuus3+3AqslXZGdJQ5K4TwNC4Z2c1qtWI2m52vzWYzGRkZ9W6bl5dHbm4uffr0cS6rqalh9uzZ6PV6brrpJgYPHlzvvikpKaSkpACwaNGiRjeR/ZTBYGjyvm1VQzmXH6sE4Ey5dtHtUr8pwOir57r+nfAztI0rCznX3sMb827OnFvdMKA7duzgqquuQqf74aJn+fLlmEwmcnJyeOKJJ+jYsSPR0dEX7JucnExycrLzdX5+fpNiOP+EujdpKOeTecUAHMwurne7yhqVTRn5XN05hNKiQkpbLNLmJefae3hj3pebc0xMzEXXuaQZymQyUVBQ4HxdUFCAyWSqd9vU1FSGDx9+wf4AUVFRJCQkcOLEiRaLVdQvv6IGgLMl1VTWqBes//J0KVUOjTGdQ10dmhDCBVxSLOLj48nOziY3Nxe73U5qaiqJiYkXbHf27FnKy8vp3r27c1lZWRk1NbW/qEpKSjhy5EidjnHhGnnlNfgbFDRqZ777qU3Hiok2+tAzIsD1wQkhWpxLmqH0ej3Tp09nwYIFqKrK6NGjiY2N5b333iM+Pt5ZOHbs2MGwYcPqPBl+9uxZXn31VXQ6HaqqMnHiRCkWbpBXbmdgjJHUU6UcLbTRKzLwR+tq2JdTwa/7WuSpfiE8lMv6LAYOHHjBNKw/HmcKcM7t/WM9evRgyZIlLRqbuLTyageVdpXuZn8O5lZw1Fr3ymLz8WI0YHRnuQtKCE8lw32IBuWV1zYDRgb5EG/y56i1yrlO0zQ2HS+hd2QAUUZfd4UohGhhUixEg/Iraod1sXxfLE4XV1Flr+3kPpBbydmSasZ0kY5tITyZFAvRoPNXFpZAA11M/qganCiqvbr4+LCVED89v+gkTVBCeLJW95yFaH3yymsw6CA8wEBXU+1QH8esNoy+enadKWPKFWb8DPJ3hxCeTIqFaFBehR1zoA86RcESaCDYT89Rq42TRVXodQrXdQt3d4hCiBYmxUI0KL+8hojA2h8VRVGID/djX07tOFCjOocQFiA/RkJ4Omk7EA3Kr6jBEujjfB1v8udcWQ3VDo0be9b/JL4QwrNIsRCX5FA18ivsWILqFguA/u2C6BTm567QhBAuJMVCXFKhzY6qQUTQD01NvSMDiTL6MKWP+RJ7CiE8iTQ2i0s6f9tsxI+aocICDLx6U7y7QhJCuIFcWYhLyiuvfSAv4kfNUEII7yPFQvDVmVL++OF+ahzaBevOD01uCZKLUCG8mRQLL+dQNVam5bL7dBFpWWUXrM8vryHIRydzagvh5aRYeLltJ0vILq1Br1PYdLzkgvV5P7kTSgjhnaRY/ISmaWg11e4OwyUcqsb7+wuIC/NjUt927DpbRlmVo842eT96IE8I4b2kWPyIVlOD+tBtlH/4lrtDcYkdp0o5W1LNlCvMXNcrEruqsf1U3auL/PIa6dwWQkix+DHFxwdCwqg5tNfdobQ4VdN4b18+HUN9GRobTLeIIDqG+rL5R01RNrtKabUqzVBCCCkWP6XE96TmuwNoqqPhjdsYTdPIK6/hy1OlvPx1DmdKqpnSx4JOUVAUhVGdQzmUV0l2aW0zXP6PhiYXQng3l/0WSE9PZ+XKlaiqSlJSEhMnTqyzfvPmzbz11luYTLVjDY0fP56kpCTnug8//BCASZMmMWrUqJYLtGsvtC2foZw9BbGdW+44bvCPr3P4PLMIAJ0CQzoYGdYx2Ln+6s4hvJWex5bjJVzfI5wvvt9WmqGEEC4pFqqqsmLFCh577DHMZjNz5swhMTGRDh061Nlu2LBhzJgxo86ysrIyVq9ezaJFiwCYPXs2iYmJGI3GFolVie+FBmhHD6F4ULHYeqKEzzOLGNc1jKT4UOLC/C6Yg8IS6MMV0YH897CVDw4WUO3QuDImiG5mfzdFLYRoLVzSDJWZmUl0dDRRUVEYDAaGDRvGrl27GrVveno6ffv2xWg0YjQa6du3L+np6S0XrCUKXbgZjh5uuWO4WG5ZDS9/fY4elgDuGhRFD0vARScrur57OA5NY2RcCC9eF8fjo2Px1UtrpRDeziVXFlarFbP5h0HnzGYzGRkZF2z31VdfcejQIdq1a8dtt92GxWK5YF+TyYTVaq33OCkpKaSkpACwaNEiLBZLk+It7tmXmmNHmrx/a2JXNf6yaR+aovDUDb2JCq3/KsFgMGCxWLjBYuGGAZ5zRdWQ83l7E2/MGbwz7+bMudX0XF555ZUMHz4cHx8f1q9fz7Jly5g3b95lvUdycjLJycnO1/n5+U2KJaBHH2xfbiIv8zuUsLY9X8PqAwXszSph1rB2+NaUkZ9/4VPaABaLpcnfr7bMG/P2xpzBO/O+3JxjYmIuus4l7Qsmk4mCggLn64KCAmdH9nnBwcH4+NR2pCYlJXHs2LF697VarRfs29x8el5R+0Ubb4qqcah8dMhKYkwQozqHujscIUQb5pJiER8fT3Z2Nrm5udjtdlJTU0lMTKyzTWFhofPr3bt3Ozu/+/fvz549eygrK6OsrIw9e/bQv3//Fo3Xp0sPMPigHT3UosdpaamnSimtcjBBZrMTQvxMLmmG0uv1TJ8+nQULFqCqKqNHjyY2Npb33nuP+Ph4EhMTWbduHbt370av12M0Gpk5cyYARqORm2++mTlz5gAwefLkFrsT6jzFxwfiuqG1oSuLlKNFmAN9GNAuyLnss4wi2gX70Dc60I2RCSE8gaJp2oXjUnuIrKysJu1nsVjIffk5tJT/onvp3yi+rXvq0P05FTyacgp/g8LS6zrTLtiXk0VV3P/pcW4fEMEvExqe0c4b23PBO/P2xpzBO/Nuc30WbZHStRc47HDyqLtDuaQqu8rfv8omMsgHvaLw4pfZOFSNzzIK8dEpJHWRvgohxM8nxeJi4nsCoGW6t9+iskbloc9O1DvXBMC7+/LJLq3hD1dF8/tBURzKq+TdfflsOlbC8I7BhPi3mhvehBBtmBSLn9h6ooTc0iqU4FDo2AVt62doVVVui2fn6VIyCmysPVx4wbqjVhsfHbKSHB9Kv+ggro4LYVjHYN7fX0ClXWV8tzA3RCyE8ERSLH6kpMrBy7vOce8H+8grr0E3ZQbk56B9+p7bYtp0vBiA9HPlFNnszuWaprH8q3OE+um5Y0AkAIqicM/gaML99cSF+dEzIsAtMQshPI8Uix8J8dMzb3QsxZU1PJpyirz2PVGGjkH7Yg3a2VMujye/ooa95yoYGhuMqsGOk6XOdd9ml5NptTGtfwRGvx+mPA3x07P0us78NSkWRVFcHrMQwjNJsfiJHpYAXvxlH8qqHTyacpKc634H/oGoby9HU1WXxrLleAkacNuACDqF+bH1xA9zTaw5aMUcYODquAs7sMMCDIRJX4UQohlJsahHr+hgnkzqSEWNyqwt+awfdy9a5kG0T95zWf+FpmlsOl5Mr4gA2gX7MjIuhMP5leSUVZNZYGNvTgUTeobjo5erByFEy5NicRHxJn+evzaOrmZ//mE18ddhD5K9/nPUP9+O+u5raHnnGv1eW44X8+fPT7DxWDEOtXGPtRy1VnG6uJrR3w/T8YtOtfNObDtRyppDBQT66BgnHdhCCBeRYnEJUUZfnkiK5e5BUXwX2I77hj7C8oF3cO7LnajPPIJWVP/otz/25alSln6Zzamial78MpuZa4+RcrSIhp6F3HS8GB+dwvDvJyeKMvrSKyKAdRmFpJ4qZXy3MAJ99Jd8DyGEaC7SsN0AnaJwbfdwBncw8sGBAj7P1LFpyMMk5h+k9ztr6f2rm+lsCUKvu7A5KC2rjOd2nKWbOYD5YzqwL6eC9/cV8Led58gureHW/hHObYttdj4+ZMVmV1EUhS3HixncwVin8/oXnUJ4dXcOBh3c0CPcJfkLIQRIsWg0c6APvx8Uzc29zaw5aGWnvgdf2fWw/gyh/npGxYUwpkso7UP8OGq1sS+nnPf3FxAb6sfjozsQ6KNnSIdgBrc38o+vc1h9oIAAHx2Te5s5arXx9JYzWCvtBProUKktUtd1r1sQhncK5o20HK6OC8UcKFOdCiFcR4rFZTIH+nBnYhR3JkaR+85KDhw8zs5Bk/j0OwcfHy7EoFOwf98v0cMSwKNXt8fo+8PVgaIo3DUoikq7ylvpeWSXVrP1RAkhfnqeHVfbR3IxYf4GloyPI8ro2+J5CiHEj0mx+Bkipk5j5OK5jPz0cUq69WP7gInkhrenV2QgCZGBF719Va9T+OPQdlTZVVKOFtM7MoA//6J9o253jQuX+bCFEK4nxeJnUAw+6B58Em37ekJS/st17/8VYjqiu2cOin/IJfc16BQeHhFDenYFA2KCMNTT5yGEEK2F3A31Myl+/uiSJqBb8DLK7/8MJYWoTz+EduDbBvf10esY1MEohUII0epJsWgmik6PbtAIdHOXQLgF9cW/on7xUYO3yAohRFsgxaKZKRHR6GY/CwOGoP3nDbQVz6NVu2/UWiGEaA5SLFqA4h+A7q5HUG66Be3rrbUP8BXkujssIYRoMpd1cKenp7Ny5UpUVSUpKYmJEyfWWf/JJ5+wYcMG9Ho9ISEh3HPPPURE1D60NnXqVDp27AjUThP4yCOPuCrsJlN0OpQbfo0WG4+6YgnqnN+DXg+KAsEh6Gb8CaV7b3eHKYQQjeKSYqGqKitWrOCxxx7DbDYzZ84cEhMT6dChg3ObuLg4Fi1ahJ+fH1988QVvv/02s2bNAsDX15fFixe7ItRmp/QbhO7R59G+3AgOB2gaWvpXqEvnobtnNsoVie4OUQghGuSSZqjMzEyio6OJiorCYDAwbNgwdu3aVWebPn364OfnB0C3bt2wWhsed6mtUKJi0E2chu7m29BNvh3dI4ugXSzqsgWoX21xd3hCCNEglxQLq9WK2Wx2vjabzZcsBhs3bqR///7O1zU1NcyePZtHH32Ur7/+ukVjdQUlOBTdQwsgvifaiudRv1gjd00JIVq1VvdQ3tatWzl27Bjz5893Llu+fDkmk4mcnByeeOIJOnbsSHR09AX7pqSkkJKSAsCiRYuwWCxNisFgMDR538uhPfF3il96kqr/rMS/qIDg3z+EYnDPKXFVzq2NN+btjTmDd+bdnDm75DeTyWSioKDA+bqgoACTyXTBdnv37mXNmjXMnz8fHx+fOvsDREVFkZCQwIkTJ+otFsnJySQnJztf5+fnNylei8XS5H0vl3bb/ShhFir/9z6Vp0+gu3s2SpDRJcf+MVfm3Jp4Y97emDN4Z96Xm3NMTMxF17mkGSo+Pp7s7Gxyc3Ox2+2kpqaSmFi3Y/f48eO89tpr/PnPfyY09IepQsvKyqipqQGgpKSEI0eO1OkYb+sUnQ7dL6eh3PFHyDiIuuhhtNwsd4clhBB1uOTKQq/XM336dBYsWICqqowePZrY2Fjee+894uPjSUxM5O2338Zms/H8888DP9wie/bsWV599VV0Oh2qqjJx4kSPKhbn6YYloVmiUJc/jfr0w+jumSu31gohWg1F8+Ce1ayspv2F7s7LVS03C/WlJ6EgB+XXv0cZOQ5Fafmxo7zxEh28M29vzBm8M+821wwlGk+JjEE3ZzH0uALt7eVory9Bs1W4OywhhJdrdXdDCVCCjOjun4e2bjXax/9CO3kU5apREG5GCbdAt94oPjJTnhDCdaRYtFKKTody/RS0rgmo/3wJ7eN3ANAAuvRAd/88t9w1JYTwTlIsWjmlRx/0C19Fq6mGIivad/vR3l6OungOullPoISGN/wmQgjxM0mfRRuh+PjWDn8+PBndfY9Dfg7qs7PR8nMuuo+maagf/wv1/RUujFQI4YmkWLRBSkJ/dLOegLIS1IUPoWUerHc7bf3HaJ+8W/v/oT0ujlII4UmkWLRRSnxPdLMXQ0Ag6nOPoe7YUGe9lvYl2uqVMHAomCNR31+BpjrcFK0Qoq2TPos2TGnXAd3c51BfeRZt1Ys4dm5C6dgFwi1oa96EuG7oZjwIe3fVbrM9BWXkOHeHLYRog+TKoo1TgoJr74y6YSpUlKFt/BTtvdchJBzdHx5D8fWDK4dD1wS0j95Gq5RnNoQQl0+uLDyAYjCg3PRbuOm3aA4H5GVDeATK9/ODKIqCbuoM1AV/Qnv3NRhzPbSLrS0kQgjRCFIsPIyi10P0hWNnKXHdUEZdi7Z5HVrqBlB00POKH64+hBDiEqRYeBHllrtRrrkJTp9AO34E7fM1aP9ZifLbu90dmhCilZNi4UUURYHIGIiMQblyGKqq1t5W22cgJF3n7vCEEK2YdHB7MeWXv4PYzqirXsRhzauzTtM01K2f4Zh/H+qOFJn2VQgvJ8XCiyk+Puj+72GorqJ48V/QjuxHUx1olRVorz2H9tZyKClCW/US6nNz0bJPuztkIYSbSDOUl1PadUC59V5q3lwGz82F4FDw8YXCApRf3ooyfhLajg1oq1eh/vV+lOHXoFz3KxRzBACaNQ+OHYFe/VCCgt2cjRCipUixEOiuGo0p6XoKtnxR++R3fg66GQ86Z+pTfjEWrf8QtP/+C23berQdKSiDRqCdOwsnMmrfJKYjuj89iRIiAxsK4YmkWAgAdAGBKIkjUBJH1LteCQ5F+e09aOMno637D1rqRmjfCWXS78AUgfbm31EXP1pbMMLMLo5eCNHSpFiIy6KYI1CmzUT77T11pnvVwi2oLz2BunguuvseR4lu78YohRDNzWXFIj09nZUrV6KqKklJSUycOLHO+pqaGv7+979z7NgxgoODeeCBB4iMjARgzZo1bNy4EZ1Oxx133EH//v1dFba4iJ/OC650741u1l9RX/wr6vw/oIy4BuWGX6OEmdwUoRCiObnkbihVVVmxYgVz587lhRdeYMeOHZw5c6bONhs3biQoKIi//e1vXH/99bzzTu3McGfOnCE1NZXnn3+eRx99lBUrVqCqqivCFpdJie+J7snlKCPHoW1fj/ro71Fffgb1izVoGQfRrHlopcVotgo0u93d4QohLoNLriwyMzOJjo4mKioKgGHDhrFr1y46dPhhWIrdu3fzq1/9CoCrrrqKN954A03T2LVrF8OGDcPHx4fIyEiio6PJzMyke/furghdXCYlNBzllrvRkm9C+99/0A7vhW92UO9TGjpd7Z1XekPt1zodKAqggML3///ka+eBLvJ1nWB+vLz+bfINBhyOeoZuv8hb/owNG+9i+TSTfL2+/pw9XL7BgMMb/kgxBqP/86Jmf1uXFAur1YrZ/EOnp9lsJiMj46Lb6PV6AgMDKS0txWq10q1bN+d2JpMJq9Va73FSUlJISUkBYNGiRVgslibFazAYmrxvW9XsOVsskHAFAI7CAuyZh3AUWaGmBq26Cq2mCqqq0Gqqa68yVAeoGmgqaFptcdG02n9otZOPOx8M/FHp0S7y9Y9dZLmmaeh0unquVBv5AGJLPKfogocfdYqC6qkPWV4iL0WnYFA9NO8f0QUZCfn+s9ycn2uP6uBOTk4mOTnZ+To/P79J72OxWJq8b1vV4jl37tly7/0zyLn2Ht6U9/k8LzfnmJiYi65zSZ+FyWSioKDA+bqgoACTyXTRbRwOBxUVFQQHB1+wr9VqvWBfIYQQLcslxSI+Pp7s7Gxyc3Ox2+2kpqaSmJhYZ5srr7ySzZs3A7Bz50569+6NoigkJiaSmppKTU0Nubm5ZGdn07VrV1eELYQQ4nsuaYbS6/VMnz6dBQsWoKoqo0ePJjY2lvfee4/4+HgSExMZM2YMf//737nvvvswGo088MADAMTGxjJ06FAefPBBdDodM2bMQKeTIa2EEMKVFM2DhxPNyspq0n7e1LZ5njfmDN6ZtzfmDN6Zd5vrsxBCCNG2SbEQQgjRICkWQgghGiTFQgghRIM8uoNbCCFE85Ari3rMnj3b3SG4nDfmDN6ZtzfmDN6Zd3PmLMVCCCFEg6RYCCGEaJB+/vz5890dRGvUpUsXd4fgct6YM3hn3t6YM3hn3s2Vs3RwCyGEaJA0QwkhhGiQFAshhBAN8qjJj36u9PR0Vq5ciaqqJCUlMXHiRHeH1CLy8/NZtmwZRUVFKIpCcnIy1113HWVlZbzwwgvk5eURERHBrFmzMBqN7g63WamqyuzZszGZTMyePZvc3FyWLl1KaWkpXbp04b777sNg8KyPRXl5OS+//DKnT59GURTuueceYmJiPPpcf/LJJ2zcuBFFUYiNjWXmzJkUFRV53Llevnw5aWlphIaGsmTJEoCLfo41TWPlypV8++23+Pn5MXPmzMvrz9CEpmma5nA4tD/84Q/auXPntJqaGu2hhx7STp8+7e6wWoTVatWOHj2qaZqmVVRUaPfff792+vRp7a233tLWrFmjaZqmrVmzRnvrrbfcGWaLWLt2rbZ06VLt6aef1jRN05YsWaJt375d0zRNe+WVV7TPP//cneG1iL/97W9aSkqKpmmaVlNTo5WVlXn0uS4oKNBmzpypVVVVaZpWe443bdrkkef6wIED2tGjR7UHH3zQuexi5/abb77RFixYoKmqqh05ckSbM2fOZR1LmqG+l5mZSXR0NFFRURgMBoYNG8auXbvcHVaLCA8Pd/5FERAQQPv27bFarezatYurr74agKuvvtrj8i8oKCAtLY2kpCSgdg7uAwcOcNVVVwEwatQoj8u5oqKCQ4cOMWbMGKB2TuagoCCPP9eqqlJdXY3D4aC6upqwsDCPPNcJCQkXXBFe7Nzu3r2bkSNHoigK3bt3p7y8nMLCwkYfq21fgzUjq9WK2Wx2vjabzWRkZLgxItfIzc3l+PHjdO3aleLiYsLDwwEICwujuLjYzdE1r1WrVjFt2jQqKysBKC0tJTAwEL1eD9RO7Wu1Wt0ZYrPLzc0lJCSE5cuXc/LkSbp06cLtt9/u0efaZDIxYcIE7rnnHnx9fenXrx9dunTx+HN93sXOrdVqxWKxOLczm81YrVbntg2RKwsvZrPZWLJkCbfffjuBgYF11imKgqIoboqs+X3zzTeEhoZ63X32DoeD48ePM3bsWJ599ln8/Pz46KOP6mzjaee6rKyMXbt2sWzZMl555RVsNhvp6enuDsstmvPcypXF90wmEwUFBc7XBQUFmEwmN0bUsux2O0uWLOEXv/gFQ4YMASA0NJTCwkLCw8MpLCwkJCTEzVE2nyNHjrB7926+/fZbqqurqaysZNWqVVRUVOBwONDr9VitVo8752azGbPZTLdu3QC46qqr+Oijjzz6XO/bt4/IyEhnTkOGDOHIkSMef67Pu9i5NZlMdWbNu9zfcXJl8b34+Hiys7PJzc3FbreTmppKYmKiu8NqEZqm8fLLL9O+fXtuuOEG5/LExES2bNkCwJYtWxg0aJC7Qmx2t9xyCy+//DLLli3jgQceoE+fPtx///307t2bnTt3ArB582aPO+dhYWGYzWbnFMP79u2jQ4cOHn2uLRYLGRkZVFVVoWmaM2dPP9fnXezcJiYmsnXrVjRN47vvviMwMLDRTVAgT3DXkZaWxj//+U9UVWX06NFMmjTJ3SG1iMOHD/P444/TsWNH5yXqb37zG7p168YLL7xAfn6+R95Oed6BAwdYu3Yts2fPJicnh6VLl1JWVkbnzp2577778PHxcXeIzerEiRO8/PLL2O12IiMjmTlzJpqmefS5fv/990lNTUWv1xMXF8fdd9+N1Wr1uHO9dOlSDh48SGlpKaGhoUyZMoVBgwbVe241TWPFihXs2bMHX19fZs6cSXx8fKOPJcVCCCFEg6QZSgghRIOkWAghhGiQFAshhBANkmIhhBCiQVIshBBCNEiKhRCtwJQpUzh37py7wxDiouQJbiF+4t5776WoqAid7oe/pUaN+S7NWQAAA7JJREFUGsWMGTPcGFX9Pv/8cwoKCrjllluYN28e06dPp1OnTu4OS3ggKRZC1OORRx6hb9++7g6jQceOHWPgwIGoqsrZs2fp0KGDu0MSHkqKhRCXYfPmzWzYsIG4uDi2bt1KeHg4M2bM4IorrgBqR/Z87bXXOHz4MEajkZtuuonk5GSgdtjsjz76iE2bNlFcXEy7du14+OGHnSOB7t27l4ULF1JSUsKIESOYMWNGg4PAHTt2jMmTJ5OVlUVERIRzVFUhmpsUCyEuU0ZGBkOGDGHFihV8/fXXPPfccyxbtgyj0ciLL75IbGwsr7zyCllZWTz55JNER0fTp08fPvnkE3bs2MH/t3f/IK0kARzHvwoxSCJGjYQoSARFUBGFVIJVSkVtBIsUAVE7FXFRawNBbKy1sRKsLaxCsEplsLTwX5AQhBCRCFlNyF5x3IJc3tsX5e7B3e9TDeyyM7PNb2d2Z3Z3d5dgMEg2m8XtdtvXzWQyJBIJyuUy29vbhMNhxsfH/1Z/pVJheXkZy7IwTRPDMKhWq9RqNWKxGLOzs//ZrWrk91FYiNRxcHDw6Sk9Go3aI4T29namp6dpampicnKS8/NzMpkMw8PD3NzcsLOzQ0tLC6FQiEgkwuXlJaOjoySTSaLRKD09PQCEQqFPdc7Pz+PxePB4PIyMjPD4+Fg3LFwuFycnJySTSZ6enojFYsTjcRYXFxkYGPjnbor8ryksROowDOOH7yw6Ozs/TQ91d3dTLBZ5eXnB6/XS2tpqH/P7/dzd3QF/bgkdCAR+WKfP57PLbrcb0zTrnnd4eMj19TXv7++4XC5SqRSmaXJ7e0swGCSRSDTUV5FfobAQaVCxWMSyLDswCoUC4XCYjo4O3t7eKJfLdmAUCgX7nwFdXV08Pz/T19f3rfo3Njao1WqsrKxwdHTE1dUV6XSatbW173VM5Ce0zkKkQa+vr1xcXFCtVkmn0+RyOSYmJvD7/QwNDXF6esrHxwfZbJZUKsXU1BQAkUiEs7Mz8vk8lmWRzWYplUpfakMulyMQCNDc3MzDw0NDW02LfIVGFiJ17O/vf1pnMTY2hmEYAAwODpLP51laWsLn87G5uUlbWxsA6+vrHB8fs7q6itfrZWFhwZ7OmpmZoVKpEI/HKZVK9Pb2srW19aX23d/f09/fb5fn5ua+010RR/qfhUgD/vp0dm9v73c3ReRfpWkoERFxpLAQERFHmoYSERFHGlmIiIgjhYWIiDhSWIiIiCOFhYiIOFJYiIiIoz8A9Chq/gb5vNwAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H.history[\"acc\"], label=\"train_acc\")\n",
        "plt.plot(np.arange(0, 100), H.history[\"val_acc\"], label=\"val_acc\")\n",
        "plt.title(\"Accuracy Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Acc\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "QZc_3ezr8h78",
        "outputId": "585725a7-ba52-48ef-a5be-0c46e92c223f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "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": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "target = ['bening']\n",
        "pred = model.predict(x_val)\n",
        "labels = (pred > 0.5).astype(np.int)\n",
        "\n",
        "\n",
        "print(classification_report(y_val, labels))"
      ],
      "metadata": {
        "id": "lMgsf3tu8rRm",
        "outputId": "a3a19aec-d057-4ff4-f837-6004705bc87d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.87      0.84      0.86       360\n",
            "           1       0.82      0.85      0.83       300\n",
            "\n",
            "    accuracy                           0.85       660\n",
            "   macro avg       0.85      0.85      0.85       660\n",
            "weighted avg       0.85      0.85      0.85       660\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "MODEL 2"
      ],
      "metadata": {
        "id": "owNe7lyhBQtR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import InputLayer, Dense, Conv2D, MaxPool2D, Flatten, BatchNormalization, Dropout\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, EarlyStopping\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "import time"
      ],
      "metadata": {
        "id": "vTtaFU4xUUJ4"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model2 = Sequential()\n",
        "\n",
        "model2.add(InputLayer(input_shape=[150,150,3]))\n",
        "model2.add(Conv2D(filters=64, kernel_size=3, activation='relu'))\n",
        "model2.add(MaxPool2D(pool_size=2))\n",
        "model2.add(Conv2D(filters=64, kernel_size=3, activation='relu'))\n",
        "model2.add(MaxPool2D(pool_size=2))\n",
        "model2.add(Conv2D(filters=128, kernel_size=3, activation='relu'))\n",
        "model2.add(MaxPool2D(pool_size=2))\n",
        "model2.add(Conv2D(filters=128, kernel_size=3, activation='relu'))\n",
        "model2.add(MaxPool2D(pool_size=2))\n",
        "\n",
        "model2.add(Dropout(0.5))\n",
        "model2.add(BatchNormalization())\n",
        "model2.add(Flatten())\n",
        "\n",
        "model2.add(Dense(512, activation='relu'))\n",
        "model2.add(Dense(128, activation='relu'))\n",
        "model2.add(Dense(1, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "vU1s5UmCBTrf"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(model2.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pk15u4DBBUBc",
        "outputId": "aa5010bc-ca4f-4935-a5db-ae89cd1918bf"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_3 (Conv2D)           (None, 148, 148, 64)      1792      \n",
            "                                                                 \n",
            " max_pooling2d_3 (MaxPooling  (None, 74, 74, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 72, 72, 64)        36928     \n",
            "                                                                 \n",
            " max_pooling2d_4 (MaxPooling  (None, 36, 36, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 34, 34, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_5 (MaxPooling  (None, 17, 17, 128)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_6 (Conv2D)           (None, 15, 15, 128)       147584    \n",
            "                                                                 \n",
            " max_pooling2d_6 (MaxPooling  (None, 7, 7, 128)        0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 7, 7, 128)         0         \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 7, 7, 128)        512       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 6272)              0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 512)               3211776   \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 128)               65664     \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3,538,241\n",
            "Trainable params: 3,537,985\n",
            "Non-trainable params: 256\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model2.compile(optimizer=Adam(lr=0.001), \n",
        "              loss='binary_crossentropy', \n",
        "              metrics=['acc'])"
      ],
      "metadata": {
        "id": "InaBEKPbtOJH",
        "outputId": "8eb1e323-0a73-495e-819c-3376d6ed255e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "H2 = model2.fit(x_train, y_train, batch_size=20, epochs=100, validation_data=(x_val, y_val))"
      ],
      "metadata": {
        "id": "XPEdzxNotW7J",
        "outputId": "f19f1bcc-4fca-43f0-cd3b-7717afe55774",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "132/132 [==============================] - 10s 62ms/step - loss: 0.5651 - acc: 0.7084 - val_loss: 0.6872 - val_acc: 0.7545\n",
            "Epoch 2/100\n",
            "132/132 [==============================] - 7s 56ms/step - loss: 0.4969 - acc: 0.7615 - val_loss: 0.6552 - val_acc: 0.6439\n",
            "Epoch 3/100\n",
            "132/132 [==============================] - 7s 56ms/step - loss: 0.4794 - acc: 0.7573 - val_loss: 0.4537 - val_acc: 0.7758\n",
            "Epoch 4/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.4787 - acc: 0.7569 - val_loss: 0.5585 - val_acc: 0.6712\n",
            "Epoch 5/100\n",
            "132/132 [==============================] - 7s 56ms/step - loss: 0.4561 - acc: 0.7728 - val_loss: 0.4456 - val_acc: 0.7697\n",
            "Epoch 6/100\n",
            "132/132 [==============================] - 9s 65ms/step - loss: 0.4230 - acc: 0.7907 - val_loss: 1.1420 - val_acc: 0.5894\n",
            "Epoch 7/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.4076 - acc: 0.7899 - val_loss: 0.6125 - val_acc: 0.7167\n",
            "Epoch 8/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.4053 - acc: 0.8111 - val_loss: 1.8756 - val_acc: 0.4879\n",
            "Epoch 9/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.3924 - acc: 0.8096 - val_loss: 0.3817 - val_acc: 0.8152\n",
            "Epoch 10/100\n",
            "132/132 [==============================] - 7s 56ms/step - loss: 0.3815 - acc: 0.8153 - val_loss: 0.4691 - val_acc: 0.7530\n",
            "Epoch 11/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.3792 - acc: 0.8165 - val_loss: 0.9531 - val_acc: 0.5652\n",
            "Epoch 12/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.3615 - acc: 0.8229 - val_loss: 0.4517 - val_acc: 0.7788\n",
            "Epoch 13/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.3798 - acc: 0.8127 - val_loss: 1.7592 - val_acc: 0.6091\n",
            "Epoch 14/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.3697 - acc: 0.8108 - val_loss: 1.0409 - val_acc: 0.5864\n",
            "Epoch 15/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.3611 - acc: 0.8172 - val_loss: 0.3581 - val_acc: 0.8212\n",
            "Epoch 16/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.3535 - acc: 0.8271 - val_loss: 0.3784 - val_acc: 0.8379\n",
            "Epoch 17/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.3376 - acc: 0.8290 - val_loss: 0.3927 - val_acc: 0.8182\n",
            "Epoch 18/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.3304 - acc: 0.8343 - val_loss: 0.3899 - val_acc: 0.7924\n",
            "Epoch 19/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.3324 - acc: 0.8366 - val_loss: 0.4010 - val_acc: 0.8333\n",
            "Epoch 20/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.3253 - acc: 0.8441 - val_loss: 0.3414 - val_acc: 0.8455\n",
            "Epoch 21/100\n",
            "132/132 [==============================] - 7s 56ms/step - loss: 0.3133 - acc: 0.8479 - val_loss: 0.3846 - val_acc: 0.8000\n",
            "Epoch 22/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.3199 - acc: 0.8464 - val_loss: 0.3473 - val_acc: 0.8258\n",
            "Epoch 23/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2993 - acc: 0.8574 - val_loss: 0.3625 - val_acc: 0.8091\n",
            "Epoch 24/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2953 - acc: 0.8582 - val_loss: 0.3797 - val_acc: 0.8439\n",
            "Epoch 25/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.2938 - acc: 0.8654 - val_loss: 0.3500 - val_acc: 0.8303\n",
            "Epoch 26/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2739 - acc: 0.8692 - val_loss: 0.3785 - val_acc: 0.8152\n",
            "Epoch 27/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2829 - acc: 0.8623 - val_loss: 0.4645 - val_acc: 0.7364\n",
            "Epoch 28/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2890 - acc: 0.8635 - val_loss: 0.3733 - val_acc: 0.8258\n",
            "Epoch 29/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2784 - acc: 0.8718 - val_loss: 1.1194 - val_acc: 0.6424\n",
            "Epoch 30/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2863 - acc: 0.8658 - val_loss: 0.3357 - val_acc: 0.8348\n",
            "Epoch 31/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2514 - acc: 0.8832 - val_loss: 0.4541 - val_acc: 0.8121\n",
            "Epoch 32/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2470 - acc: 0.8809 - val_loss: 0.3822 - val_acc: 0.8212\n",
            "Epoch 33/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2349 - acc: 0.8866 - val_loss: 0.3565 - val_acc: 0.8364\n",
            "Epoch 34/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2193 - acc: 0.8972 - val_loss: 0.4231 - val_acc: 0.8318\n",
            "Epoch 35/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.2053 - acc: 0.9056 - val_loss: 0.4431 - val_acc: 0.7970\n",
            "Epoch 36/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.2293 - acc: 0.8912 - val_loss: 0.3581 - val_acc: 0.8515\n",
            "Epoch 37/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.2017 - acc: 0.9090 - val_loss: 0.3780 - val_acc: 0.8318\n",
            "Epoch 38/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.2022 - acc: 0.9075 - val_loss: 0.3986 - val_acc: 0.8258\n",
            "Epoch 39/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1895 - acc: 0.9139 - val_loss: 0.3979 - val_acc: 0.8394\n",
            "Epoch 40/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1983 - acc: 0.9075 - val_loss: 0.4645 - val_acc: 0.8227\n",
            "Epoch 41/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1721 - acc: 0.9253 - val_loss: 0.4251 - val_acc: 0.8242\n",
            "Epoch 42/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1792 - acc: 0.9219 - val_loss: 0.5327 - val_acc: 0.8197\n",
            "Epoch 43/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1685 - acc: 0.9230 - val_loss: 0.3653 - val_acc: 0.8409\n",
            "Epoch 44/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1519 - acc: 0.9344 - val_loss: 0.4919 - val_acc: 0.8227\n",
            "Epoch 45/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.1625 - acc: 0.9276 - val_loss: 0.4376 - val_acc: 0.8470\n",
            "Epoch 46/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.1495 - acc: 0.9393 - val_loss: 0.5253 - val_acc: 0.8121\n",
            "Epoch 47/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1467 - acc: 0.9370 - val_loss: 0.4340 - val_acc: 0.8485\n",
            "Epoch 48/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1049 - acc: 0.9545 - val_loss: 0.5856 - val_acc: 0.8197\n",
            "Epoch 49/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1620 - acc: 0.9317 - val_loss: 0.4860 - val_acc: 0.8197\n",
            "Epoch 50/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1110 - acc: 0.9515 - val_loss: 0.5468 - val_acc: 0.8242\n",
            "Epoch 51/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1238 - acc: 0.9458 - val_loss: 0.4870 - val_acc: 0.8394\n",
            "Epoch 52/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1196 - acc: 0.9484 - val_loss: 0.5284 - val_acc: 0.8455\n",
            "Epoch 53/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1104 - acc: 0.9568 - val_loss: 0.4604 - val_acc: 0.8318\n",
            "Epoch 54/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1341 - acc: 0.9473 - val_loss: 0.5138 - val_acc: 0.8303\n",
            "Epoch 55/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1127 - acc: 0.9530 - val_loss: 0.6713 - val_acc: 0.8015\n",
            "Epoch 56/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0901 - acc: 0.9632 - val_loss: 0.5104 - val_acc: 0.8394\n",
            "Epoch 57/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.1034 - acc: 0.9575 - val_loss: 0.5704 - val_acc: 0.8424\n",
            "Epoch 58/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0963 - acc: 0.9583 - val_loss: 0.8076 - val_acc: 0.7712\n",
            "Epoch 59/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.1104 - acc: 0.9556 - val_loss: 0.5844 - val_acc: 0.8273\n",
            "Epoch 60/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0917 - acc: 0.9613 - val_loss: 0.6547 - val_acc: 0.8000\n",
            "Epoch 61/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0912 - acc: 0.9655 - val_loss: 0.5979 - val_acc: 0.8348\n",
            "Epoch 62/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0928 - acc: 0.9644 - val_loss: 0.6194 - val_acc: 0.8348\n",
            "Epoch 63/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0786 - acc: 0.9716 - val_loss: 0.6550 - val_acc: 0.8333\n",
            "Epoch 64/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0909 - acc: 0.9655 - val_loss: 1.8299 - val_acc: 0.7364\n",
            "Epoch 65/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0858 - acc: 0.9655 - val_loss: 0.6202 - val_acc: 0.8242\n",
            "Epoch 66/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.0813 - acc: 0.9689 - val_loss: 0.7995 - val_acc: 0.8015\n",
            "Epoch 67/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0660 - acc: 0.9738 - val_loss: 0.7871 - val_acc: 0.8121\n",
            "Epoch 68/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0613 - acc: 0.9761 - val_loss: 0.7656 - val_acc: 0.8212\n",
            "Epoch 69/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0966 - acc: 0.9636 - val_loss: 0.6104 - val_acc: 0.8394\n",
            "Epoch 70/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0680 - acc: 0.9761 - val_loss: 0.5735 - val_acc: 0.8348\n",
            "Epoch 71/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0588 - acc: 0.9772 - val_loss: 0.7422 - val_acc: 0.8152\n",
            "Epoch 72/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0548 - acc: 0.9807 - val_loss: 0.6371 - val_acc: 0.8273\n",
            "Epoch 73/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0498 - acc: 0.9818 - val_loss: 0.7782 - val_acc: 0.8182\n",
            "Epoch 74/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0499 - acc: 0.9795 - val_loss: 0.6007 - val_acc: 0.8348\n",
            "Epoch 75/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0577 - acc: 0.9784 - val_loss: 0.7793 - val_acc: 0.8136\n",
            "Epoch 76/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0537 - acc: 0.9818 - val_loss: 0.7583 - val_acc: 0.8212\n",
            "Epoch 77/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0508 - acc: 0.9826 - val_loss: 0.6853 - val_acc: 0.8500\n",
            "Epoch 78/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0373 - acc: 0.9841 - val_loss: 0.7180 - val_acc: 0.8455\n",
            "Epoch 79/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0667 - acc: 0.9776 - val_loss: 0.7006 - val_acc: 0.8439\n",
            "Epoch 80/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0648 - acc: 0.9784 - val_loss: 0.7017 - val_acc: 0.8045\n",
            "Epoch 81/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0573 - acc: 0.9772 - val_loss: 0.6611 - val_acc: 0.8364\n",
            "Epoch 82/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0472 - acc: 0.9810 - val_loss: 0.6446 - val_acc: 0.8288\n",
            "Epoch 83/100\n",
            "132/132 [==============================] - 8s 61ms/step - loss: 0.0532 - acc: 0.9769 - val_loss: 0.8249 - val_acc: 0.8076\n",
            "Epoch 84/100\n",
            "132/132 [==============================] - 8s 61ms/step - loss: 0.0401 - acc: 0.9871 - val_loss: 0.7404 - val_acc: 0.8364\n",
            "Epoch 85/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0491 - acc: 0.9826 - val_loss: 0.7585 - val_acc: 0.8106\n",
            "Epoch 86/100\n",
            "132/132 [==============================] - 7s 53ms/step - loss: 0.0264 - acc: 0.9890 - val_loss: 0.6924 - val_acc: 0.8348\n",
            "Epoch 87/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0430 - acc: 0.9856 - val_loss: 0.6598 - val_acc: 0.8364\n",
            "Epoch 88/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0358 - acc: 0.9871 - val_loss: 0.7247 - val_acc: 0.8364\n",
            "Epoch 89/100\n",
            "132/132 [==============================] - 7s 52ms/step - loss: 0.0759 - acc: 0.9761 - val_loss: 0.8204 - val_acc: 0.8015\n",
            "Epoch 90/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0597 - acc: 0.9799 - val_loss: 0.6495 - val_acc: 0.8364\n",
            "Epoch 91/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0281 - acc: 0.9890 - val_loss: 1.0308 - val_acc: 0.8091\n",
            "Epoch 92/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0632 - acc: 0.9799 - val_loss: 0.7256 - val_acc: 0.8212\n",
            "Epoch 93/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0677 - acc: 0.9735 - val_loss: 0.6496 - val_acc: 0.8333\n",
            "Epoch 94/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0458 - acc: 0.9837 - val_loss: 0.7812 - val_acc: 0.8152\n",
            "Epoch 95/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0433 - acc: 0.9856 - val_loss: 0.7174 - val_acc: 0.8121\n",
            "Epoch 96/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0349 - acc: 0.9871 - val_loss: 0.6738 - val_acc: 0.8379\n",
            "Epoch 97/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0221 - acc: 0.9917 - val_loss: 0.7768 - val_acc: 0.8500\n",
            "Epoch 98/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0199 - acc: 0.9936 - val_loss: 0.8425 - val_acc: 0.8106\n",
            "Epoch 99/100\n",
            "132/132 [==============================] - 7s 55ms/step - loss: 0.0311 - acc: 0.9894 - val_loss: 0.8494 - val_acc: 0.8242\n",
            "Epoch 100/100\n",
            "132/132 [==============================] - 7s 54ms/step - loss: 0.0393 - acc: 0.9879 - val_loss: 0.7044 - val_acc: 0.8288\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "%matplotlib inline\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(np.arange(0, 100), H.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Loss Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "YJFERsM789F4",
        "outputId": "2330d0bc-66f8-4097-8e3b-b591d5e222b6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H2.history[\"acc\"], label=\"train_acc\")\n",
        "plt.plot(np.arange(0, 100), H2.history[\"val_acc\"], label=\"val_acc\")\n",
        "plt.title(\"Accuracy Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Acc\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "b5nUdjJMD2xN",
        "outputId": "35183999-fcd3-4b61-f8d2-6c8c39f3c7f1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "target = ['bening']\n",
        "pred = model2.predict(x_val)\n",
        "labels = (pred > 0.5).astype(np.int)\n",
        "\n",
        "\n",
        "print(classification_report(y_val, labels))"
      ],
      "metadata": {
        "id": "WtCg-RtTD20E",
        "outputId": "ac1e15c4-04c6-442c-f8af-4ade53b679f3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.80      0.84       360\n",
            "           1       0.78      0.87      0.82       300\n",
            "\n",
            "    accuracy                           0.83       660\n",
            "   macro avg       0.83      0.83      0.83       660\n",
            "weighted avg       0.83      0.83      0.83       660\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "MODEL 3"
      ],
      "metadata": {
        "id": "lr-bsJGtBUO6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model3 = Sequential()\n",
        "\n",
        "model3.add(InputLayer(input_shape=[150,150,3]))\n",
        "model3.add(Conv2D(filters=64, kernel_size=3, activation='relu'))\n",
        "model3.add(MaxPool2D(pool_size=2))\n",
        "model3.add(BatchNormalization())\n",
        "\n",
        "model3.add(Conv2D(filters=64, kernel_size=3, activation='relu'))\n",
        "model3.add(MaxPool2D(pool_size=2))\n",
        "model3.add(BatchNormalization())\n",
        "\n",
        "model3.add(Conv2D(filters=128, kernel_size=3, activation='relu'))\n",
        "model3.add(MaxPool2D(pool_size=2))\n",
        "model3.add(BatchNormalization())\n",
        "\n",
        "model3.add(Conv2D(filters=128, kernel_size=3, activation='relu'))\n",
        "model3.add(MaxPool2D(pool_size=2))\n",
        "model3.add(BatchNormalization())\n",
        "model3.add(Dropout(0.5))\n",
        "model3.add(Flatten())\n",
        "\n",
        "model3.add(Dense(512, activation='relu'))\n",
        "model3.add(Dropout(0.5))\n",
        "model3.add(Dense(128, activation='relu'))\n",
        "model3.add(Dense(1, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "P5ACkSoe4UGW"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "SqeleUZs4UJR"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(model3.summary())"
      ],
      "metadata": {
        "outputId": "78c9c48e-cecc-4b25-87f3-815aa1a7786d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ze5eV5gZHaJB"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_7 (Conv2D)           (None, 148, 148, 64)      1792      \n",
            "                                                                 \n",
            " max_pooling2d_7 (MaxPooling  (None, 74, 74, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 74, 74, 64)       256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " conv2d_8 (Conv2D)           (None, 72, 72, 64)        36928     \n",
            "                                                                 \n",
            " max_pooling2d_8 (MaxPooling  (None, 36, 36, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " batch_normalization_2 (Batc  (None, 36, 36, 64)       256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " conv2d_9 (Conv2D)           (None, 34, 34, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_9 (MaxPooling  (None, 17, 17, 128)      0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " batch_normalization_3 (Batc  (None, 17, 17, 128)      512       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " conv2d_10 (Conv2D)          (None, 15, 15, 128)       147584    \n",
            "                                                                 \n",
            " max_pooling2d_10 (MaxPoolin  (None, 7, 7, 128)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " batch_normalization_4 (Batc  (None, 7, 7, 128)        512       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 7, 7, 128)         0         \n",
            "                                                                 \n",
            " flatten_2 (Flatten)         (None, 6272)              0         \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 512)               3211776   \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 512)               0         \n",
            "                                                                 \n",
            " dense_6 (Dense)             (None, 128)               65664     \n",
            "                                                                 \n",
            " dense_7 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3,539,265\n",
            "Trainable params: 3,538,497\n",
            "Non-trainable params: 768\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n",
        "model3.compile(optimizer=Adam(lr=0.001), \n",
        "              loss='binary_crossentropy', \n",
        "              metrics=['acc'])"
      ],
      "metadata": {
        "id": "ZKufQHoN4UMO",
        "outputId": "4152abd0-3bf3-4db8-edb5-f57b1b8d11fd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "H3 = model3.fit(x_train, y_train, batch_size=20, epochs=100, validation_data=(x_val, y_val))"
      ],
      "metadata": {
        "id": "dOxZh9L_4UPs",
        "outputId": "f2a28e92-8177-4029-c580-d8d690442d52",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "132/132 [==============================] - 9s 61ms/step - loss: 0.8941 - acc: 0.7304 - val_loss: 1.0227 - val_acc: 0.4545\n",
            "Epoch 2/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.5375 - acc: 0.7649 - val_loss: 1.2962 - val_acc: 0.4758\n",
            "Epoch 3/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.4385 - acc: 0.7903 - val_loss: 0.5297 - val_acc: 0.7576\n",
            "Epoch 4/100\n",
            "132/132 [==============================] - 10s 73ms/step - loss: 0.4326 - acc: 0.8024 - val_loss: 0.5314 - val_acc: 0.7606\n",
            "Epoch 5/100\n",
            "132/132 [==============================] - 9s 65ms/step - loss: 0.4027 - acc: 0.8024 - val_loss: 0.5372 - val_acc: 0.7561\n",
            "Epoch 6/100\n",
            "132/132 [==============================] - 9s 65ms/step - loss: 0.3762 - acc: 0.8309 - val_loss: 0.3673 - val_acc: 0.8364\n",
            "Epoch 7/100\n",
            "132/132 [==============================] - 8s 64ms/step - loss: 0.3446 - acc: 0.8362 - val_loss: 0.4807 - val_acc: 0.7545\n",
            "Epoch 8/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.3402 - acc: 0.8358 - val_loss: 0.4629 - val_acc: 0.7818\n",
            "Epoch 9/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.3396 - acc: 0.8411 - val_loss: 0.3430 - val_acc: 0.8455\n",
            "Epoch 10/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.3236 - acc: 0.8513 - val_loss: 0.5235 - val_acc: 0.7909\n",
            "Epoch 11/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.3259 - acc: 0.8453 - val_loss: 0.3618 - val_acc: 0.8318\n",
            "Epoch 12/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.3086 - acc: 0.8449 - val_loss: 0.9652 - val_acc: 0.6288\n",
            "Epoch 13/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.3027 - acc: 0.8699 - val_loss: 0.3807 - val_acc: 0.8348\n",
            "Epoch 14/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2958 - acc: 0.8551 - val_loss: 0.4068 - val_acc: 0.8045\n",
            "Epoch 15/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2942 - acc: 0.8559 - val_loss: 0.3542 - val_acc: 0.8485\n",
            "Epoch 16/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2730 - acc: 0.8711 - val_loss: 0.4361 - val_acc: 0.8242\n",
            "Epoch 17/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2754 - acc: 0.8749 - val_loss: 0.4528 - val_acc: 0.8000\n",
            "Epoch 18/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2725 - acc: 0.8741 - val_loss: 0.3388 - val_acc: 0.8561\n",
            "Epoch 19/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2577 - acc: 0.8821 - val_loss: 0.3670 - val_acc: 0.8303\n",
            "Epoch 20/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2542 - acc: 0.8870 - val_loss: 0.3387 - val_acc: 0.8364\n",
            "Epoch 21/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2411 - acc: 0.8915 - val_loss: 0.4455 - val_acc: 0.8258\n",
            "Epoch 22/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2425 - acc: 0.8931 - val_loss: 0.5974 - val_acc: 0.7970\n",
            "Epoch 23/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2226 - acc: 0.8965 - val_loss: 0.4548 - val_acc: 0.7667\n",
            "Epoch 24/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2287 - acc: 0.8999 - val_loss: 0.4833 - val_acc: 0.8121\n",
            "Epoch 25/100\n",
            "132/132 [==============================] - 8s 61ms/step - loss: 0.2171 - acc: 0.9082 - val_loss: 0.3605 - val_acc: 0.8545\n",
            "Epoch 26/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2144 - acc: 0.9094 - val_loss: 0.3465 - val_acc: 0.8500\n",
            "Epoch 27/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.2122 - acc: 0.9063 - val_loss: 0.6020 - val_acc: 0.7879\n",
            "Epoch 28/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.2048 - acc: 0.9060 - val_loss: 0.4047 - val_acc: 0.8470\n",
            "Epoch 29/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.1915 - acc: 0.9101 - val_loss: 0.9444 - val_acc: 0.7485\n",
            "Epoch 30/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1947 - acc: 0.9116 - val_loss: 0.3532 - val_acc: 0.8576\n",
            "Epoch 31/100\n",
            "132/132 [==============================] - 8s 62ms/step - loss: 0.1838 - acc: 0.9181 - val_loss: 0.4369 - val_acc: 0.8167\n",
            "Epoch 32/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1735 - acc: 0.9215 - val_loss: 0.5861 - val_acc: 0.8030\n",
            "Epoch 33/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1625 - acc: 0.9302 - val_loss: 0.6020 - val_acc: 0.8030\n",
            "Epoch 34/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1400 - acc: 0.9416 - val_loss: 0.4092 - val_acc: 0.8561\n",
            "Epoch 35/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1549 - acc: 0.9336 - val_loss: 0.5096 - val_acc: 0.8394\n",
            "Epoch 36/100\n",
            "132/132 [==============================] - 8s 62ms/step - loss: 0.1439 - acc: 0.9397 - val_loss: 0.3936 - val_acc: 0.8500\n",
            "Epoch 37/100\n",
            "132/132 [==============================] - 9s 71ms/step - loss: 0.1405 - acc: 0.9397 - val_loss: 1.0997 - val_acc: 0.7530\n",
            "Epoch 38/100\n",
            "132/132 [==============================] - 9s 66ms/step - loss: 0.1282 - acc: 0.9431 - val_loss: 0.3844 - val_acc: 0.8379\n",
            "Epoch 39/100\n",
            "132/132 [==============================] - 10s 76ms/step - loss: 0.1350 - acc: 0.9439 - val_loss: 0.6949 - val_acc: 0.8333\n",
            "Epoch 40/100\n",
            "132/132 [==============================] - 9s 65ms/step - loss: 0.1486 - acc: 0.9427 - val_loss: 0.6602 - val_acc: 0.8303\n",
            "Epoch 41/100\n",
            "132/132 [==============================] - 9s 67ms/step - loss: 0.1310 - acc: 0.9484 - val_loss: 0.8121 - val_acc: 0.7985\n",
            "Epoch 42/100\n",
            "132/132 [==============================] - 9s 67ms/step - loss: 0.1298 - acc: 0.9477 - val_loss: 0.6315 - val_acc: 0.8273\n",
            "Epoch 43/100\n",
            "132/132 [==============================] - 8s 64ms/step - loss: 0.1285 - acc: 0.9499 - val_loss: 0.4089 - val_acc: 0.8303\n",
            "Epoch 44/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.1308 - acc: 0.9473 - val_loss: 0.4397 - val_acc: 0.8258\n",
            "Epoch 45/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1018 - acc: 0.9553 - val_loss: 0.5097 - val_acc: 0.8394\n",
            "Epoch 46/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1049 - acc: 0.9568 - val_loss: 0.8458 - val_acc: 0.7712\n",
            "Epoch 47/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.1078 - acc: 0.9575 - val_loss: 0.4733 - val_acc: 0.8561\n",
            "Epoch 48/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0844 - acc: 0.9659 - val_loss: 0.8532 - val_acc: 0.8318\n",
            "Epoch 49/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0910 - acc: 0.9666 - val_loss: 0.5066 - val_acc: 0.8303\n",
            "Epoch 50/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0900 - acc: 0.9655 - val_loss: 0.5201 - val_acc: 0.8455\n",
            "Epoch 51/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0843 - acc: 0.9697 - val_loss: 0.5463 - val_acc: 0.8439\n",
            "Epoch 52/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0846 - acc: 0.9662 - val_loss: 0.6157 - val_acc: 0.8318\n",
            "Epoch 53/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0777 - acc: 0.9685 - val_loss: 1.2561 - val_acc: 0.7379\n",
            "Epoch 54/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0885 - acc: 0.9670 - val_loss: 0.5354 - val_acc: 0.8061\n",
            "Epoch 55/100\n",
            "132/132 [==============================] - 8s 61ms/step - loss: 0.0781 - acc: 0.9704 - val_loss: 0.7729 - val_acc: 0.8197\n",
            "Epoch 56/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0481 - acc: 0.9829 - val_loss: 0.5962 - val_acc: 0.8485\n",
            "Epoch 57/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0739 - acc: 0.9742 - val_loss: 2.6462 - val_acc: 0.7727\n",
            "Epoch 58/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0642 - acc: 0.9772 - val_loss: 0.6852 - val_acc: 0.8394\n",
            "Epoch 59/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0566 - acc: 0.9791 - val_loss: 0.5988 - val_acc: 0.8424\n",
            "Epoch 60/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0628 - acc: 0.9784 - val_loss: 0.6280 - val_acc: 0.8091\n",
            "Epoch 61/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0633 - acc: 0.9772 - val_loss: 0.6799 - val_acc: 0.8258\n",
            "Epoch 62/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0643 - acc: 0.9772 - val_loss: 0.7375 - val_acc: 0.8197\n",
            "Epoch 63/100\n",
            "132/132 [==============================] - 8s 62ms/step - loss: 0.0570 - acc: 0.9784 - val_loss: 0.7466 - val_acc: 0.8409\n",
            "Epoch 64/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0470 - acc: 0.9856 - val_loss: 0.6073 - val_acc: 0.8515\n",
            "Epoch 65/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0563 - acc: 0.9772 - val_loss: 0.8142 - val_acc: 0.7985\n",
            "Epoch 66/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0699 - acc: 0.9784 - val_loss: 0.7102 - val_acc: 0.8182\n",
            "Epoch 67/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0504 - acc: 0.9826 - val_loss: 0.6394 - val_acc: 0.8273\n",
            "Epoch 68/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0488 - acc: 0.9810 - val_loss: 0.6282 - val_acc: 0.8485\n",
            "Epoch 69/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0540 - acc: 0.9818 - val_loss: 0.6028 - val_acc: 0.8470\n",
            "Epoch 70/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0445 - acc: 0.9845 - val_loss: 0.8434 - val_acc: 0.8167\n",
            "Epoch 71/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0459 - acc: 0.9822 - val_loss: 0.7480 - val_acc: 0.8288\n",
            "Epoch 72/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0638 - acc: 0.9784 - val_loss: 0.7124 - val_acc: 0.8318\n",
            "Epoch 73/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0347 - acc: 0.9860 - val_loss: 0.7436 - val_acc: 0.8364\n",
            "Epoch 74/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0774 - acc: 0.9750 - val_loss: 0.8861 - val_acc: 0.8076\n",
            "Epoch 75/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0439 - acc: 0.9822 - val_loss: 0.6496 - val_acc: 0.8424\n",
            "Epoch 76/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0437 - acc: 0.9833 - val_loss: 0.9045 - val_acc: 0.8318\n",
            "Epoch 77/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0369 - acc: 0.9882 - val_loss: 0.8128 - val_acc: 0.8364\n",
            "Epoch 78/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0326 - acc: 0.9863 - val_loss: 0.6672 - val_acc: 0.8061\n",
            "Epoch 79/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0376 - acc: 0.9871 - val_loss: 0.7328 - val_acc: 0.8364\n",
            "Epoch 80/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0342 - acc: 0.9909 - val_loss: 0.7315 - val_acc: 0.8470\n",
            "Epoch 81/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0495 - acc: 0.9837 - val_loss: 0.5915 - val_acc: 0.8439\n",
            "Epoch 82/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0434 - acc: 0.9856 - val_loss: 0.8051 - val_acc: 0.8212\n",
            "Epoch 83/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0369 - acc: 0.9848 - val_loss: 0.8866 - val_acc: 0.8227\n",
            "Epoch 84/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0417 - acc: 0.9856 - val_loss: 0.6829 - val_acc: 0.8394\n",
            "Epoch 85/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0348 - acc: 0.9879 - val_loss: 0.8965 - val_acc: 0.8379\n",
            "Epoch 86/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0403 - acc: 0.9867 - val_loss: 0.9181 - val_acc: 0.7591\n",
            "Epoch 87/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0236 - acc: 0.9920 - val_loss: 0.7526 - val_acc: 0.8500\n",
            "Epoch 88/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0108 - acc: 0.9966 - val_loss: 0.8867 - val_acc: 0.8424\n",
            "Epoch 89/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0410 - acc: 0.9875 - val_loss: 1.2278 - val_acc: 0.7955\n",
            "Epoch 90/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0380 - acc: 0.9898 - val_loss: 0.7647 - val_acc: 0.8318\n",
            "Epoch 91/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0246 - acc: 0.9924 - val_loss: 0.9574 - val_acc: 0.8394\n",
            "Epoch 92/100\n",
            "132/132 [==============================] - 8s 58ms/step - loss: 0.0371 - acc: 0.9890 - val_loss: 0.7684 - val_acc: 0.8485\n",
            "Epoch 93/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0308 - acc: 0.9898 - val_loss: 0.8497 - val_acc: 0.8364\n",
            "Epoch 94/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0456 - acc: 0.9803 - val_loss: 0.9276 - val_acc: 0.8409\n",
            "Epoch 95/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0393 - acc: 0.9875 - val_loss: 0.7220 - val_acc: 0.8439\n",
            "Epoch 96/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0230 - acc: 0.9913 - val_loss: 0.7444 - val_acc: 0.8348\n",
            "Epoch 97/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0185 - acc: 0.9932 - val_loss: 0.8804 - val_acc: 0.8485\n",
            "Epoch 98/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0344 - acc: 0.9867 - val_loss: 1.0086 - val_acc: 0.8136\n",
            "Epoch 99/100\n",
            "132/132 [==============================] - 8s 60ms/step - loss: 0.0206 - acc: 0.9939 - val_loss: 0.8433 - val_acc: 0.8470\n",
            "Epoch 100/100\n",
            "132/132 [==============================] - 8s 59ms/step - loss: 0.0281 - acc: 0.9924 - val_loss: 0.7136 - val_acc: 0.8561\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%matplotlib inline\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H3.history[\"loss\"], label=\"train_loss\")\n",
        "plt.plot(np.arange(0, 100), H3.history[\"val_loss\"], label=\"val_loss\")\n",
        "plt.title(\"Loss Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "GtaENB2IHvJm",
        "outputId": "58d29605-ce76-4da8-d80e-f7effa397bbb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.figure()\n",
        "plt.plot(np.arange(0, 100), H3.history[\"acc\"], label=\"train_acc\")\n",
        "plt.plot(np.arange(0, 100), H3.history[\"val_acc\"], label=\"val_acc\")\n",
        "plt.title(\"Accuracy Plot\")\n",
        "plt.xlabel(\"Epoch #\")\n",
        "plt.ylabel(\"Acc\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "oYS9HIb_HvSG",
        "outputId": "c0f1f69a-f851-4044-a7e4-9aee379c442f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        }
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "target = ['bening']\n",
        "pred = model3.predict(x_val)\n",
        "labels = (pred > 0.5).astype(np.int)\n",
        "\n",
        "\n",
        "print(classification_report(y_val, labels))"
      ],
      "metadata": {
        "id": "_l9O8yUeHvgh",
        "outputId": "df979ea1-0a5a-4a17-ccbd-62903aca9efd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.90      0.83      0.86       360\n",
            "           1       0.81      0.89      0.85       300\n",
            "\n",
            "    accuracy                           0.86       660\n",
            "   macro avg       0.86      0.86      0.86       660\n",
            "weighted avg       0.86      0.86      0.86       660\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "0hZorYCq8GvN"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}