_9_합성곱_신경망(Convolution_Neural_Network).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/yoonjong8739/NLP_study/blob/master/_9_%ED%95%A9%EC%84%B1%EA%B3%B1_%EC%8B%A0%EA%B2%BD%EB%A7%9D(Convolution_Neural_Network).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "86l7I5SQ_u9J"
      },
      "source": [
        "# 합성곱 신경망(Convolution Neural Network)\n",
        "\n",
        "- 이미지 인식, 음성 인식 등에 자주 사용되는데,  \n",
        "  특히, 이미지 인식 분야에서 거의 모든 딥러닝 기술에 사용\n",
        "\n",
        "<br>\n",
        "\n",
        "## 컨볼루션 신경망의 등장\n",
        "- 1989년 얀 르쿤(Yann LeCun) 교수의 논문에서 발표\n",
        "\n",
        "  - 필기체 인식에서 의미가 있었지만 범용화하는데에는 무리\n",
        "\n",
        "- 1998년, \"Gradient-Based Learning Applied to Document Recognition\"이라는 논문에서 LeNet-5 제시\n",
        "\n",
        "- 합성곱 층(convolution layer), 풀링 층(pooling layer) 소개\n",
        "  <br>\n",
        "\n",
        "  <img src=\"https://miro.medium.com/max/4308/1*1TI1aGBZ4dybR6__DI9dzA.png\">\n",
        "  \n",
        "  <center>[LeNet-5 구조]</center>\n",
        "\n",
        "  <sub>출처: https://medium.com/@pechyonkin/key-deep-learning-architectures-lenet-5-6fc3c59e6f4</sub>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y46evOQHNukF"
      },
      "source": [
        "## 컨볼루션 신경망 구조 예시\n",
        "\n",
        "<img src=\"https://www.oreilly.com/api/v2/epubs/9781789138900/files/assets/c38754ca-f2ea-425a-b7a6-1fe0f2f5074e.png\" width=\"600\">\n",
        "\n",
        "<sub>출처: https://www.oreilly.com/library/view/neural-network-projects/9781789138900/8e87ad66-6de3-4275-81a4-62b54436bf16.xhtml</sub>\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wtb8UWyTNxp_"
      },
      "source": [
        "## 합성곱 연산\n",
        "- 필터(filter) 연산\n",
        "  - 입력 데이터에 필터를 통한 어떠한 연산을 진행\n",
        "  \n",
        "  - **필터에 대응하는 원소끼리 곱하고, 그 합을 구함**\n",
        "\n",
        "  - 연산이 완료된 결과 데이터를 **특징 맵(feature map)**이라 부름\n",
        "\n",
        "- 필터(filter)\n",
        "  - 커널(kernel)이라고도 칭함\n",
        "  \n",
        "  - 흔히 사진 어플에서 사용하는 '이미지 필터'와 비슷한 개념\n",
        "\n",
        "  - 필터의 사이즈는 \"거의 항상 홀수\"\n",
        "    - 짝수이면 패딩이 비대칭이 되어버림\n",
        "  \n",
        "    - 왼쪽, 오른쪽을 다르게 주어야함\n",
        "  \n",
        "    - 중심위치가 존재, 즉 구별된 하나의 픽셀(중심 픽셀)이 존재\n",
        "\n",
        "  - 필터의 학습 파라미터 개수는 입력 데이터의 크기와 상관없이 일정  \n",
        "    따라서, 과적합을 방지할 수 있음\n",
        "\n",
        "  <br>\n",
        "  <img src=\"https://www.tomasbeuzen.com/deep-learning-with-pytorch/_images/conv-1.gif\">\n",
        "\n",
        "  <sub>출처: http://deeplearning.net/software/theano_versions/dev/tutorial/conv_arithmetic.html</sub>\n",
        "\n",
        "\n",
        "- 연산 시각화\n",
        "  <img src=\"https://www.researchgate.net/profile/Ihab_S_Mohamed/publication/324165524/figure/fig3/AS:611103423860736@1522709818959/An-example-of-convolution-operation-in-2D-2.png\" width=\"500\">\n",
        "\n",
        "  <sub>출처: https://www.researchgate.net/figure/An-example-of-convolution-operation-in-2D-2_fig3_324165524</sub>\n",
        "\n",
        "\n",
        "- 일반적으로, 합성곱 연산을 한 후의 데이터 사이즈는  \n",
        "  ### $\\quad (n-f+1) \\times (n-f+1)$\n",
        "    $n$: 입력 데이터의 크기  \n",
        "    $f$: 필터(커널)의 크기\n",
        "\n",
        "\n",
        "  <img src=\"https://miro.medium.com/max/1400/1*Fw-ehcNBR9byHtho-Rxbtw.gif\" width=\"400\">\n",
        "\n",
        "  <sub>출처: https://towardsdatascience.com/intuitively-understanding-convolutions-for-deep-learning-1f6f42faee1</sub>\n",
        "  \n",
        "  위 예에서 입력 데이터 크기($n$)는 5, 필터의 크기($k$)는 3이므로  \n",
        "  출력 데이터의 크기는 $(5 - 3 + 1) = 3$\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5-ekDsJwN2Y-"
      },
      "source": [
        "## 패딩(padding)과 스트라이드(stride)\n",
        "- 필터(커널) 사이즈과 함께 **입력 이미지와 출력 이미지의 사이즈를 결정**하기 위해 사용\n",
        "\n",
        "- 사용자가 결정할 수 있음\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "alV1bgcjN4Kc"
      },
      "source": [
        "\n",
        "### 패딩\n",
        "- 입력 데이터의 주변을 특정 값으로 채우는 기법\n",
        "  - 주로 0으로 많이 채움\n",
        "\n",
        "  <img src=\"http://deeplearning.net/software/theano_versions/dev/_images/arbitrary_padding_no_strides.gif\" width=\"300\">\n",
        "\n",
        "- 출력 데이터의 크기\n",
        "  ### $\\quad (n+2p-f+1) \\times (n+2p-f+1)$\n",
        "\n",
        "  위 그림에서, 입력 데이터의 크기($n$)는 5, 필터의 크기($f$)는 4, 패딩값($p$)은 2이므로    \n",
        "  출력 데이터의 크기는 ($5 + 2\\times 2 - 4 + 1) = 6$\n",
        "\n",
        "### 'valid' 와 'same'\n",
        "- 'valid'\n",
        "  - 패딩을 주지 않음\n",
        "  - padding=0 (0으로 채워진 테두리가 아니라 패딩을 주지 않는다는 의미)\n",
        "\n",
        "- 'same'\n",
        "  - 패딩을 주어 입력 이미지의 크기와 연산 후의 이미지 크기를 같게함\n",
        "\n",
        "  - 만약, 필터(커널)의 크기가 $k$ 이면,  \n",
        "    패딩의 크기는 $p = \\frac{k-1}{2}$ (단, <u>stride=1)</u>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zlZ7zG6ON85J"
      },
      "source": [
        "\n",
        "\n",
        "### 스트라이드\n",
        "- 필터를 적용하는 간격을 의미\n",
        "\n",
        "- 아래는 그림의 간격 2\n",
        "\n",
        "  <img src=\"https://miro.medium.com/v2/resize:fit:588/1*BMngs93_rm2_BpJFH2mS0Q.gif\">\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LPcsND-0OCNm"
      },
      "source": [
        "## 출력 데이터의 크기\n",
        "\n",
        "## $\\qquad OH = \\frac{H + 2P - FH}{S} + 1 $\n",
        "## $\\qquad OW = \\frac{W + 2P - FW}{S} + 1 $\n",
        "\n",
        "- 입력 크기 : $(H, W)$\n",
        "\n",
        "- 필터 크기 : $(FH, FW)$\n",
        "\n",
        "- 출력 크기 : $(OH, OW)$\n",
        "\n",
        "- 패딩, 스트라이드 : $P, S$\n",
        "\n",
        "- (주의)\n",
        "  - 위 식의 값에서 $\\frac{H + 2P - FH}{S}$ 또는 $\\frac{W + 2P - FW}{S}$가 정수로 나누어 떨어지는 값이어야 함\n",
        "  - 만약, 정수로 나누어 떨어지지 않으면  \n",
        "    패딩, 스트라이드값을 조정하여 정수로 나누어 떨어지게 해야함\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1x4UoMbF8jJ9"
      },
      "source": [
        "## 풀링(Pooling)\n",
        "\n",
        "- 필터(커널) 사이즈 내에서 특정 값을 추출하는 과정"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lDiaO3XF8oC_"
      },
      "source": [
        "### 맥스 풀링(Max Pooling)\n",
        "- 가장 많이 사용되는 방법\n",
        "\n",
        "- 출력 데이터의 사이즈 계산은 컨볼루션 연산과 동일\n",
        "## $\\quad OH = \\frac{H + 2P - FH}{S} + 1 $\n",
        "## $\\quad OW = \\frac{W + 2P - FW}{S} + 1 $\n",
        "\n",
        "- 일반적으로 stride=2, kernel_size=2 를 통해  \n",
        "  **특징맵의 크기를 <u>절반으로 줄이는 역할</u>**\n",
        "\n",
        "- 모델이 물체의 주요한 특징을 학습할 수 있도록 해주며,  \n",
        "  컨볼루션 신경망이 이동 불변성 특성을 가지게 해줌\n",
        "  - 예를 들어, 아래의 그림에서 초록색 사각형 안에 있는  \n",
        "    2와 8의 위치를 바꾼다해도 맥스 풀링 연산은 8을 추출\n",
        "\n",
        "- 모델의 파라미터 개수를 줄여주고, 연산 속도를 빠르게 해줌\n",
        "\n",
        "  <br>\n",
        "\n",
        "  <img src=\"https://cs231n.github.io/assets/cnn/maxpool.jpeg\" width=\"600\">\n",
        "\n",
        "  <sub>출처: https://cs231n.github.io/convolutional-networks/</sub>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CrvtX4LrkhPI"
      },
      "source": [
        "## Conv1D\n",
        "\n",
        "- 텍스트 분류나 시계열 예측 같은 간단한 문제, 오디오 생성, 기계 번역 등의 문제에서 좋은 성능\n",
        "\n",
        "- 타임스텝의 순서에 민감하지 X\n",
        "\n",
        "\n",
        "- 2D Convolution\n",
        "  - 지역적 특징을 인식\n",
        "\n",
        "  <img src=\"https://jiho-ml.com/content/images/2020/07/2D-convolutional-example_2x-700x584.png\" width=\"500\">\n",
        "\n",
        "- 1D Convolution\n",
        "  - 문맥을 인식\n",
        "\n",
        "  <img src=\"https://img1.daumcdn.net/thumb/R1280x0/?scode=mtistory2&fname=https%3A%2F%2Fblog.kakaocdn.net%2Fdn%2FF2bI1%2FbtqJSciC5nJ%2FstSwF20OKLLuCSzvVz68fk%2Fimg.png\" width=\"500\">\n",
        "\n",
        "  <sub>출처: https://missinglink.ai/guides/keras/keras-conv1d-working-1d-convolutional-neural-networks-keras/</sub>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aJNAI-aYnDuc"
      },
      "source": [
        "### Conv1D Layer\n",
        "- 입력: `(batch_size, timesteps, channels)`\n",
        "\n",
        "- 출력: `(batch_size, timesteps, filters)`\n",
        "\n",
        "- 필터의 사이즈가 커져도 모델이 급격히 증가하지 않기 때문에 다양한 크기를 사용할 수 있음\n",
        "- 데이터의 품질이 좋으면 굳이 크기를 달리하여 여러 개를 사용하지 않아도 될 수도 있음"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ADJZHPLJqTZa"
      },
      "source": [
        "### MaxPooling1D Layer\n",
        "- 다운 샘플링 효과\n",
        "\n",
        "- 단지 1차원형태로 바뀐 것 뿐"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wirZzwYJrNll"
      },
      "source": [
        "### GlobalMaxPooling Layer\n",
        "- 배치차원을 제외하고 2차원 형태를 1차원 형태로 바꾸어주는 레이어\n",
        "\n",
        "- Flatten layer로 대신 사용가능"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YQY_1IL9nQs2"
      },
      "source": [
        "## IMDB 데이터"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pl4Eu32qnUES"
      },
      "source": [
        "### 데이터 로드 및 전처리"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WcG-Hiltklbe"
      },
      "source": [
        "from keras.datasets import imdb\n",
        "from keras.utils import pad_sequences\n",
        "from keras.models import Sequential\n",
        "from keras.optimizers import RMSprop\n",
        "from keras.layers import Dense, Embedding, Conv1D, MaxPooling1D, GlobalMaxPooling1D"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GGYaNT8hnec5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ca9a52a9-025f-41f7-e5f9-22fcad464246"
      },
      "source": [
        "num_words = 10000\n",
        "max_len = 500\n",
        "batch_size = 32\n",
        "\n",
        "(input_train, y_train), (input_test, y_test) = imdb.load_data(num_words = num_words)\n",
        "print(len(input_train))\n",
        "print(len(input_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "25000\n",
            "25000\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nQBzp3ownpEK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4f966b0b-5a9e-44a2-ff62-f3109c287227"
      },
      "source": [
        "pad_x_train = pad_sequences(input_train, maxlen=max_len)\n",
        "pad_x_test = pad_sequences(input_test, maxlen=max_len)\n",
        "\n",
        "print(pad_x_train.shape)\n",
        "print(pad_x_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(25000, 500)\n",
            "(25000, 500)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j3E0hYiHoBO0"
      },
      "source": [
        "### 모델 구성"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PvKPr7N6nuHr"
      },
      "source": [
        "def build_model():\n",
        "    model = Sequential()\n",
        "\n",
        "    model.add(Embedding(input_dim=num_words, output_dim=32, input_length=max_len))\n",
        "\n",
        "    model.add(Conv1D(filters=32, kernel_size=7, activation='relu'))\n",
        "    model.add(MaxPooling1D(pool_size=7))\n",
        "    model.add(Conv1D(filters=32, kernel_size=5, activation='relu'))\n",
        "    model.add(MaxPooling1D(pool_size=5))\n",
        "\n",
        "    model.add(GlobalMaxPooling1D())  # Flatten()으로 대신 사용\n",
        "\n",
        "    model.add(Dense(1, activation='sigmoid'))\n",
        "\n",
        "    model.compile(optimizer = RMSprop(learning_rate=1e-4),\n",
        "                  loss = 'binary_crossentropy',\n",
        "                  metrics = ['acc'])\n",
        "\n",
        "    return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wwWZ47NHolB6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "aa764c96-b57c-47c3-eb6d-3df8c78d88ff"
      },
      "source": [
        "model = build_model()\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding (Embedding)       (None, 500, 32)           320000    \n",
            "                                                                 \n",
            " conv1d (Conv1D)             (None, 494, 32)           7200      \n",
            "                                                                 \n",
            " max_pooling1d (MaxPooling1D  (None, 70, 32)           0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 66, 32)            5152      \n",
            "                                                                 \n",
            " max_pooling1d_1 (MaxPooling  (None, 13, 32)           0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " global_max_pooling1d (Globa  (None, 32)               0         \n",
            " lMaxPooling1D)                                                  \n",
            "                                                                 \n",
            " dense (Dense)               (None, 1)                 33        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 332,385\n",
            "Trainable params: 332,385\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0qUbPGdmpHpQ"
      },
      "source": [
        "### 모델 학습"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xcCgjLHvonWB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e299e4d3-0bf7-43b8-990d-3804a851be0d"
      },
      "source": [
        "history = model.fit(\n",
        "    pad_x_train, y_train,\n",
        "    batch_size = 128, epochs = 30,\n",
        "    validation_split = 0.2\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "157/157 [==============================] - 5s 9ms/step - loss: 0.6929 - acc: 0.5152 - val_loss: 0.6923 - val_acc: 0.5166\n",
            "Epoch 2/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.6907 - acc: 0.5818 - val_loss: 0.6914 - val_acc: 0.5880\n",
            "Epoch 3/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.6881 - acc: 0.6564 - val_loss: 0.6896 - val_acc: 0.5768\n",
            "Epoch 4/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.6836 - acc: 0.7186 - val_loss: 0.6835 - val_acc: 0.7180\n",
            "Epoch 5/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.6730 - acc: 0.7767 - val_loss: 0.6676 - val_acc: 0.7508\n",
            "Epoch 6/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.6461 - acc: 0.7969 - val_loss: 0.6274 - val_acc: 0.7866\n",
            "Epoch 7/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.5883 - acc: 0.8080 - val_loss: 0.5568 - val_acc: 0.8032\n",
            "Epoch 8/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.5053 - acc: 0.8274 - val_loss: 0.4752 - val_acc: 0.8196\n",
            "Epoch 9/30\n",
            "157/157 [==============================] - 1s 8ms/step - loss: 0.4244 - acc: 0.8457 - val_loss: 0.4145 - val_acc: 0.8340\n",
            "Epoch 10/30\n",
            "157/157 [==============================] - 1s 9ms/step - loss: 0.3658 - acc: 0.8589 - val_loss: 0.3791 - val_acc: 0.8374\n",
            "Epoch 11/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.3266 - acc: 0.8717 - val_loss: 0.3585 - val_acc: 0.8458\n",
            "Epoch 12/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2993 - acc: 0.8820 - val_loss: 0.3459 - val_acc: 0.8500\n",
            "Epoch 13/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2787 - acc: 0.8909 - val_loss: 0.3356 - val_acc: 0.8544\n",
            "Epoch 14/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2617 - acc: 0.8963 - val_loss: 0.3306 - val_acc: 0.8562\n",
            "Epoch 15/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2467 - acc: 0.9039 - val_loss: 0.3275 - val_acc: 0.8590\n",
            "Epoch 16/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2341 - acc: 0.9105 - val_loss: 0.3293 - val_acc: 0.8590\n",
            "Epoch 17/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2229 - acc: 0.9160 - val_loss: 0.3256 - val_acc: 0.8642\n",
            "Epoch 18/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2124 - acc: 0.9219 - val_loss: 0.3251 - val_acc: 0.8638\n",
            "Epoch 19/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.2027 - acc: 0.9255 - val_loss: 0.3258 - val_acc: 0.8642\n",
            "Epoch 20/30\n",
            "157/157 [==============================] - 1s 9ms/step - loss: 0.1936 - acc: 0.9303 - val_loss: 0.3278 - val_acc: 0.8660\n",
            "Epoch 21/30\n",
            "157/157 [==============================] - 1s 9ms/step - loss: 0.1852 - acc: 0.9342 - val_loss: 0.3303 - val_acc: 0.8662\n",
            "Epoch 22/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1776 - acc: 0.9377 - val_loss: 0.3324 - val_acc: 0.8648\n",
            "Epoch 23/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1700 - acc: 0.9406 - val_loss: 0.3353 - val_acc: 0.8648\n",
            "Epoch 24/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1632 - acc: 0.9435 - val_loss: 0.3399 - val_acc: 0.8630\n",
            "Epoch 25/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1569 - acc: 0.9466 - val_loss: 0.3458 - val_acc: 0.8624\n",
            "Epoch 26/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1511 - acc: 0.9486 - val_loss: 0.3474 - val_acc: 0.8636\n",
            "Epoch 27/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1452 - acc: 0.9508 - val_loss: 0.3579 - val_acc: 0.8626\n",
            "Epoch 28/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1397 - acc: 0.9534 - val_loss: 0.3578 - val_acc: 0.8592\n",
            "Epoch 29/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1345 - acc: 0.9549 - val_loss: 0.3636 - val_acc: 0.8610\n",
            "Epoch 30/30\n",
            "157/157 [==============================] - 1s 7ms/step - loss: 0.1294 - acc: 0.9575 - val_loss: 0.3672 - val_acc: 0.8612\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n2yEkawIph4C"
      },
      "source": [
        "### 시각화"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BfKC0UOkpPmP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 723
        },
        "outputId": "66e1c6da-e725-4149-ebce-afd198d23969"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('ggplot')\n",
        "\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "acc = history.history['acc']\n",
        "val_acc = history.history['val_acc']\n",
        "\n",
        "epochs = range(1, len(loss)+1)\n",
        "\n",
        "plt.figure(figsize=(12,8))\n",
        "plt.subplot(1,2,1)\n",
        "plt.plot(epochs, loss, 'r--', label='train loss')\n",
        "plt.plot(epochs, val_loss, 'b:', label='validation loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.title('train loss vs validation loss')\n",
        "plt.legend()\n",
        "plt.grid(True)\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "plt.plot(epochs, acc, 'r--', label='train accuracy')\n",
        "plt.plot(epochs, val_acc, 'b:', label='validation accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.title('train accuracy vs validation accuracy')\n",
        "plt.legend()\n",
        "plt.grid(True)\n",
        "\n",
        "plt.show();"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x800 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.evaluate(pad_x_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CRg32S1b6uh7",
        "outputId": "20d02109-f788-43f2-9596-d30dcc0b0b63"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "782/782 [==============================] - 2s 3ms/step - loss: 0.3895 - acc: 0.8472\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.3895144462585449, 0.8471999764442444]"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Oz_Iu_VHV7Z5"
      },
      "source": [
        "## Reuters 데이터"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4_Q6p0nhV7Z6"
      },
      "source": [
        "### 데이터 로드 및 전처리"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "waEmkCowV7Z7"
      },
      "source": [
        "from keras.datasets import reuters"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dggcmeJ4V7Z_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "30530794-8381-4303-a777-553399a2b53b"
      },
      "source": [
        "num_words = 10000\n",
        "max_len = 500\n",
        "batch_size = 32\n",
        "\n",
        "(input_train, y_train), (input_test, y_test) = reuters.load_data(num_words = num_words)\n",
        "\n",
        "print(len(input_train))\n",
        "print(len(input_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "8982\n",
            "2246\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EBOJIgbgV7aB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cdf9528c-6114-424d-e839-1eb46b5588ab"
      },
      "source": [
        "pad_x_train = pad_sequences(input_train, maxlen=max_len)\n",
        "pad_x_test = pad_sequences(input_test, maxlen=max_len)\n",
        "\n",
        "print(pad_x_train.shape)\n",
        "print(pad_x_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(8982, 500)\n",
            "(2246, 500)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xehgNec0V7aE"
      },
      "source": [
        "### 모델 구성"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RTV2qF8_V7aE"
      },
      "source": [
        "def build_model2():\n",
        "    model = Sequential()\n",
        "\n",
        "    model.add(Embedding(input_dim=num_words, output_dim=64, input_length=max_len))\n",
        "\n",
        "    model.add(Conv1D(filters=64, kernel_size=7, activation='relu'))\n",
        "    model.add(MaxPooling1D(pool_size=7))\n",
        "    model.add(Conv1D(filters=64, kernel_size=5, activation='relu'))\n",
        "    model.add(MaxPooling1D(pool_size=5))\n",
        "\n",
        "    model.add(GlobalMaxPooling1D())  # Flatten()으로 대신 사용\n",
        "\n",
        "    model.add(Dense(46, activation='softmax'))\n",
        "\n",
        "    model.compile(optimizer = 'adam',\n",
        "                  loss = 'sparse_categorical_crossentropy',\n",
        "                  metrics = ['acc'])\n",
        "\n",
        "    return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "48n2W3xpV7aG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "206e55aa-1704-4638-aa10-534cda33aee8"
      },
      "source": [
        "model = build_model2()\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding_1 (Embedding)     (None, 500, 64)           640000    \n",
            "                                                                 \n",
            " conv1d_2 (Conv1D)           (None, 494, 64)           28736     \n",
            "                                                                 \n",
            " max_pooling1d_2 (MaxPooling  (None, 70, 64)           0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " conv1d_3 (Conv1D)           (None, 66, 64)            20544     \n",
            "                                                                 \n",
            " max_pooling1d_3 (MaxPooling  (None, 13, 64)           0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " global_max_pooling1d_1 (Glo  (None, 64)               0         \n",
            " balMaxPooling1D)                                                \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 46)                2990      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 692,270\n",
            "Trainable params: 692,270\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WeRmu5WMV7aI"
      },
      "source": [
        "### 모델 학습"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BGVdYyIrV7aJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "72a9d3ee-651a-43e5-cc57-706e1720b0fd"
      },
      "source": [
        "history2 = model.fit(\n",
        "    pad_x_train, y_train,\n",
        "    batch_size = 128, epochs = 30,\n",
        "    validation_split = 0.2\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "57/57 [==============================] - 16s 252ms/step - loss: 2.6841 - acc: 0.3663 - val_loss: 2.0261 - val_acc: 0.4975\n",
            "Epoch 2/30\n",
            "57/57 [==============================] - 12s 207ms/step - loss: 1.8510 - acc: 0.5243 - val_loss: 1.7103 - val_acc: 0.5409\n",
            "Epoch 3/30\n",
            "57/57 [==============================] - 12s 204ms/step - loss: 1.6011 - acc: 0.5873 - val_loss: 1.5662 - val_acc: 0.6227\n",
            "Epoch 4/30\n",
            "57/57 [==============================] - 9s 159ms/step - loss: 1.4075 - acc: 0.6586 - val_loss: 1.4360 - val_acc: 0.6578\n",
            "Epoch 5/30\n",
            "57/57 [==============================] - 9s 154ms/step - loss: 1.1940 - acc: 0.7129 - val_loss: 1.3272 - val_acc: 0.6939\n",
            "Epoch 6/30\n",
            "57/57 [==============================] - 7s 134ms/step - loss: 0.9878 - acc: 0.7552 - val_loss: 1.2658 - val_acc: 0.7117\n",
            "Epoch 7/30\n",
            "57/57 [==============================] - 8s 146ms/step - loss: 0.8082 - acc: 0.7916 - val_loss: 1.2309 - val_acc: 0.7268\n",
            "Epoch 8/30\n",
            "57/57 [==============================] - 9s 157ms/step - loss: 0.6615 - acc: 0.8308 - val_loss: 1.2617 - val_acc: 0.7340\n",
            "Epoch 9/30\n",
            "57/57 [==============================] - 5s 84ms/step - loss: 0.5384 - acc: 0.8657 - val_loss: 1.3028 - val_acc: 0.7340\n",
            "Epoch 10/30\n",
            "57/57 [==============================] - 8s 142ms/step - loss: 0.4440 - acc: 0.8913 - val_loss: 1.3789 - val_acc: 0.7340\n",
            "Epoch 11/30\n",
            "57/57 [==============================] - 7s 122ms/step - loss: 0.3592 - acc: 0.9134 - val_loss: 1.4408 - val_acc: 0.7290\n",
            "Epoch 12/30\n",
            "57/57 [==============================] - 6s 103ms/step - loss: 0.3022 - acc: 0.9296 - val_loss: 1.4707 - val_acc: 0.7390\n",
            "Epoch 13/30\n",
            "57/57 [==============================] - 5s 92ms/step - loss: 0.2525 - acc: 0.9424 - val_loss: 1.5267 - val_acc: 0.7368\n",
            "Epoch 14/30\n",
            "57/57 [==============================] - 5s 93ms/step - loss: 0.2153 - acc: 0.9505 - val_loss: 1.6236 - val_acc: 0.7301\n",
            "Epoch 15/30\n",
            "57/57 [==============================] - 4s 68ms/step - loss: 0.1937 - acc: 0.9574 - val_loss: 1.6123 - val_acc: 0.7284\n",
            "Epoch 16/30\n",
            "57/57 [==============================] - 4s 72ms/step - loss: 0.1790 - acc: 0.9569 - val_loss: 1.6542 - val_acc: 0.7340\n",
            "Epoch 17/30\n",
            "57/57 [==============================] - 4s 73ms/step - loss: 0.1625 - acc: 0.9585 - val_loss: 1.6934 - val_acc: 0.7329\n",
            "Epoch 18/30\n",
            "57/57 [==============================] - 4s 68ms/step - loss: 0.1424 - acc: 0.9621 - val_loss: 1.7649 - val_acc: 0.7279\n",
            "Epoch 19/30\n",
            "57/57 [==============================] - 4s 71ms/step - loss: 0.1527 - acc: 0.9594 - val_loss: 1.7466 - val_acc: 0.7346\n",
            "Epoch 20/30\n",
            "57/57 [==============================] - 3s 55ms/step - loss: 0.1426 - acc: 0.9630 - val_loss: 1.6881 - val_acc: 0.7407\n",
            "Epoch 21/30\n",
            "57/57 [==============================] - 2s 39ms/step - loss: 0.1311 - acc: 0.9658 - val_loss: 1.7104 - val_acc: 0.7401\n",
            "Epoch 22/30\n",
            "57/57 [==============================] - 4s 75ms/step - loss: 0.1247 - acc: 0.9617 - val_loss: 1.7345 - val_acc: 0.7396\n",
            "Epoch 23/30\n",
            "57/57 [==============================] - 3s 55ms/step - loss: 0.1193 - acc: 0.9656 - val_loss: 1.7539 - val_acc: 0.7368\n",
            "Epoch 24/30\n",
            "57/57 [==============================] - 3s 46ms/step - loss: 0.1209 - acc: 0.9646 - val_loss: 1.7453 - val_acc: 0.7307\n",
            "Epoch 25/30\n",
            "57/57 [==============================] - 2s 36ms/step - loss: 0.1147 - acc: 0.9646 - val_loss: 1.7147 - val_acc: 0.7457\n",
            "Epoch 26/30\n",
            "57/57 [==============================] - 2s 42ms/step - loss: 0.1096 - acc: 0.9640 - val_loss: 1.7488 - val_acc: 0.7279\n",
            "Epoch 27/30\n",
            "57/57 [==============================] - 2s 31ms/step - loss: 0.1038 - acc: 0.9665 - val_loss: 1.7891 - val_acc: 0.7362\n",
            "Epoch 28/30\n",
            "57/57 [==============================] - 3s 50ms/step - loss: 0.1114 - acc: 0.9642 - val_loss: 1.7474 - val_acc: 0.7423\n",
            "Epoch 29/30\n",
            "57/57 [==============================] - 2s 36ms/step - loss: 0.1040 - acc: 0.9641 - val_loss: 1.7896 - val_acc: 0.7385\n",
            "Epoch 30/30\n",
            "57/57 [==============================] - 1s 18ms/step - loss: 0.1061 - acc: 0.9634 - val_loss: 1.7326 - val_acc: 0.7368\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ao8NNj2SV7aL"
      },
      "source": [
        "### 시각화"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D_CisVtyV7aL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 723
        },
        "outputId": "afb7722c-cf22-417d-ba53-4abc857aba59"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('ggplot')\n",
        "\n",
        "loss = history2.history['loss']\n",
        "val_loss = history2.history['val_loss']\n",
        "acc = history2.history['acc']\n",
        "val_acc = history2.history['val_acc']\n",
        "\n",
        "epochs = range(1, len(loss)+1)\n",
        "\n",
        "plt.figure(figsize=(12,8))\n",
        "plt.subplot(1,2,1)\n",
        "plt.plot(epochs, loss, 'r--', label='train loss')\n",
        "plt.plot(epochs, val_loss, 'b:', label='validation loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.title('train loss vs validation loss')\n",
        "plt.legend()\n",
        "plt.grid(True)\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "plt.plot(epochs, acc, 'r--', label='train accuracy')\n",
        "plt.plot(epochs, val_acc, 'b:', label='validation accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.title('train accuracy vs validation accuracy')\n",
        "plt.legend()\n",
        "plt.grid(True)\n",
        "\n",
        "plt.show();"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x800 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NCsNsoiUV7aP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cd4f48aa-0d7d-4178-a9e9-f8a73ea5e1da"
      },
      "source": [
        "model.evaluate(pad_x_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "71/71 [==============================] - 1s 9ms/step - loss: 1.6510 - acc: 0.7400\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.6510415077209473, 0.7399821877479553]"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "6mYksskt_h-f"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}