4WEEK_IMDB_DATASET-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d2569cbe",
   "metadata": {},
   "source": [
    "# IMDB DATASET\n",
    "\n",
    "목표: 리뷰 텍스트를 이용하여 긍정적인 영화 리뷰인지 부정적인 영화 리뷰인지 분류한다\n",
    "\n",
    "구성: 2,5000개의 훈련 데이터, 2,5000개의 테스트 데이터(각각은 50% 긍정, 50% 부정으로 이루어짐)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4bba44f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 14, 22, 16, 43, 530, 973, 1622, 1385, 65, 458, 2, 66, 2, 4, 173, 36, 256, 5, 25, 100, 43, 838, 112, 50, 670, 2, 9, 35, 480, 284, 5, 150, 4, 172, 112, 167, 2, 336, 385, 39, 4, 172, 2, 1111, 17, 546, 38, 13, 447, 4, 192, 50, 16, 6, 147, 2, 19, 14, 22, 4, 1920, 2, 469, 4, 22, 71, 87, 12, 16, 43, 530, 38, 76, 15, 13, 1247, 4, 22, 17, 515, 17, 12, 16, 626, 18, 2, 5, 62, 386, 12, 8, 316, 8, 106, 5, 4, 2, 2, 16, 480, 66, 2, 33, 4, 130, 12, 16, 38, 619, 5, 25, 124, 51, 36, 135, 48, 25, 1415, 33, 6, 22, 12, 215, 28, 77, 52, 5, 14, 407, 16, 82, 2, 8, 4, 107, 117, 2, 15, 256, 4, 2, 7, 2, 5, 723, 36, 71, 43, 530, 476, 26, 400, 317, 46, 7, 4, 2, 1029, 13, 104, 88, 4, 381, 15, 297, 98, 32, 2, 56, 26, 141, 6, 194, 2, 18, 4, 226, 22, 21, 134, 476, 26, 480, 5, 144, 30, 2, 18, 51, 36, 28, 224, 92, 25, 104, 4, 226, 65, 16, 38, 1334, 88, 12, 16, 283, 5, 16, 2, 113, 103, 32, 15, 16, 2, 19, 178, 32]\n",
      "[1 0 0 ... 0 1 0]\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import imdb\n",
    "\n",
    "(train_data,train_labels), (test_data,test_labels) = imdb.load_data(num_words=10000)\n",
    "#num_words=10000은 가장 자주 나타나는 단어 1만개를 사용하겠다는 의미\n",
    "#train_data, train_labels는 부정=0, 긍정=1로 표현된 리스트\n",
    "print(train_data[0]) #첫번째 리뷰의 단어 인덱스 리스트\n",
    "print(train_labels) #긍정 또는 부정의 인덱스"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c8762302",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 1. 1. ... 0. 0. 0.]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def vectorize_sequences(sequences, dimension=10000): #one-hot encoding함수 구현\n",
    "    results = np.zeros((len(sequences),dimension)) #행:sequences, 열:dimension 크기의 원소가 0인 배열 생성\n",
    "    for i, sequence in enumerate(sequences): #enumerate는 (리스트의 원소 인덱스, 원소)로 반환한다. -> i:인덱스 sequence:원소 \n",
    "        results[i,sequence] = 1. #results[i]에서 특정 인덱스의 위치를 1로 만든다\n",
    "    return results\n",
    "\n",
    "x_train = vectorize_sequences(train_data)\n",
    "x_test = vectorize_sequences(test_data)\n",
    "\n",
    "print(x_train[0]) #0~2번 인덱스는 각각 패딩, 문서 시작, 사전에 없음을 의미한다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2e0ae388",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1. 0. 0. ... 0. 1. 0.]\n"
     ]
    }
   ],
   "source": [
    "y_train = np.asarray(train_labels).astype('float32') #0 또는 1 ->  0.0 또는 1.0\n",
    "y_test = np.asarray(test_labels).astype('float32')\n",
    "print(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f00e0339",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import models\n",
    "from keras import layers\n",
    "\n",
    "model = models.Sequential()\n",
    "model.add(layers.Dense(16, activation='relu',input_shape=(10000,))) #relu는 음수를 0으로 만든다\n",
    "model.add(layers.Dense(16,activation='relu'))\n",
    "model.add(layers.Dense(1,activation='sigmoid')) #sigmoid는 임의의 값을 0~1사이로 압축하기 때문에 출력을 확률처럼 해석할 수 있다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3cf2d1d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer='rmsprop',\n",
    "             loss='binary_crossentropy',\n",
    "             metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8acd8407",
   "metadata": {},
   "outputs": [],
   "source": [
    "# optimizer의 매개변수를 수정할 때\n",
    "# from tensorflow.keras import optimizers\n",
    "# \n",
    "# model.compile(optimizer=optimizers.RMSprop(lr=0.001), #learning rate 설정\n",
    "#              loss='binary_crossentropy',\n",
    "#              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "14426987",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 손실함수, 측정 함수 custom할 때\n",
    "# from keras import losses\n",
    "# from keras import metrics\n",
    "#\n",
    "# model.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
    "#              loss=losses.binary_crossentropy,\n",
    "#              metrics=[metrics.binary_accuracy])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3b6b96d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_val = x_train[:10000] #처음부터 인덱스9999까지\n",
    "partial_x_train = x_train[10000:] #10000부터 끝까지\n",
    "y_val = y_train[:10000] #검증할 데이터\n",
    "partial_y_train = y_train[10000:] #훈련시킬 데이터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "fea5a347",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "in user code:\n\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1021, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1010, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1000, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 859, in train_step\n        y_pred = self(x, training=True)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\", line 67, in error_handler\n        raise e.with_traceback(filtered_tb) from None\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\input_spec.py\", line 264, in assert_input_compatibility\n        raise ValueError(f'Input {input_index} of layer \"{layer_name}\" is '\n\n    ValueError: Input 0 of layer \"sequential_3\" is incompatible with the layer: expected shape=(None, 2000), found shape=(None, 3000)\n",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_50008/2992057672.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m history = model.fit(partial_x_train, #fit 메서드는 History 객체를 반환한다. 객체는 훈련하는 동안 발생한 모든 정보를 딕셔너리에 보관\n\u001b[0m\u001b[0;32m      2\u001b[0m                    \u001b[0mpartial_y_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m                    \u001b[0mepochs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m20\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;31m#20번 반복\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m                    \u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m512\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;31m#512개 뽑아서\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m                    validation_data=(x_val,y_val)) #검증 데이터 추가\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     65\u001b[0m     \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint: disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 67\u001b[1;33m       \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     68\u001b[0m     \u001b[1;32mfinally\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     69\u001b[0m       \u001b[1;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\framework\\func_graph.py\u001b[0m in \u001b[0;36mautograph_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m   1145\u001b[0m           \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint:disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1146\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"ag_error_metadata\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1147\u001b[1;33m               \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1148\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1149\u001b[0m               \u001b[1;32mraise\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: in user code:\n\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1021, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1010, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1000, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 859, in train_step\n        y_pred = self(x, training=True)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\", line 67, in error_handler\n        raise e.with_traceback(filtered_tb) from None\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\input_spec.py\", line 264, in assert_input_compatibility\n        raise ValueError(f'Input {input_index} of layer \"{layer_name}\" is '\n\n    ValueError: Input 0 of layer \"sequential_3\" is incompatible with the layer: expected shape=(None, 2000), found shape=(None, 3000)\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(partial_x_train, #fit 메서드는 History 객체를 반환한다. 객체는 훈련하는 동안 발생한 모든 정보를 딕셔너리에 보관\n",
    "                   partial_y_train,\n",
    "                   epochs=20, #20번 반복\n",
    "                   batch_size=512, #512개 뽑아서\n",
    "                   validation_data=(x_val,y_val)) #검증 데이터 추가"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "dc2e21da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "history_dict = history.history\n",
    "loss_values = history_dict['loss'] \n",
    "val_loss_values = history_dict['val_loss']\n",
    "\n",
    "epochs = range(1, len(loss_values) + 1)\n",
    "\n",
    "plt.plot(epochs, loss_values, 'bo', label='Training loss') \n",
    "plt.plot(epochs, val_loss_values, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss') \n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss') \n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "723e2cb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf() # 생성한 그래프 초기화\n",
    "acc = history_dict['accuracy'] \n",
    "val_acc = history_dict['val_accuracy']\n",
    "plt.plot(epochs, acc, 'bo', label='Training acc') \n",
    "plt.plot(epochs, val_acc, 'b', label='Validation acc') \n",
    "plt.title('Training and validation accuracy') \n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Acc') \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "2264169d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "in user code:\n\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1021, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1010, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1000, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 859, in train_step\n        y_pred = self(x, training=True)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\", line 67, in error_handler\n        raise e.with_traceback(filtered_tb) from None\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\input_spec.py\", line 264, in assert_input_compatibility\n        raise ValueError(f'Input {input_index} of layer \"{layer_name}\" is '\n\n    ValueError: Input 0 of layer \"sequential_2\" is incompatible with the layer: expected shape=(None, 10000), found shape=(None, 3000)\n",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_50008/814389236.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     10\u001b[0m               metrics=['accuracy'])\n\u001b[0;32m     11\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 12\u001b[1;33m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m512\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     13\u001b[0m \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mresults\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     65\u001b[0m     \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint: disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 67\u001b[1;33m       \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     68\u001b[0m     \u001b[1;32mfinally\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     69\u001b[0m       \u001b[1;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\framework\\func_graph.py\u001b[0m in \u001b[0;36mautograph_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m   1145\u001b[0m           \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint:disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1146\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"ag_error_metadata\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1147\u001b[1;33m               \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1148\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1149\u001b[0m               \u001b[1;32mraise\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: in user code:\n\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1021, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1010, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 1000, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 859, in train_step\n        y_pred = self(x, training=True)\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\", line 67, in error_handler\n        raise e.with_traceback(filtered_tb) from None\n    File \"C:\\Users\\jw101\\anaconda3\\lib\\site-packages\\keras\\engine\\input_spec.py\", line 264, in assert_input_compatibility\n        raise ValueError(f'Input {input_index} of layer \"{layer_name}\" is '\n\n    ValueError: Input 0 of layer \"sequential_2\" is incompatible with the layer: expected shape=(None, 10000), found shape=(None, 3000)\n"
     ]
    }
   ],
   "source": [
    "#4번째 epochs에서부터 overfitting되었음 \n",
    "#따라서 epochs를 4까지로 제한하고 훈련 재시작\n",
    "model = models.Sequential()\n",
    "model.add(layers.Dense(16, activation='relu', input_shape=(10000,))) \n",
    "model.add(layers.Dense(16, activation='relu')) \n",
    "model.add(layers.Dense(1, activation='sigmoid'))\n",
    "\n",
    "model.compile(optimizer='rmsprop',\n",
    "              loss='binary_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "model.fit(x_train, y_train, epochs=4, batch_size=512) \n",
    "results = model.evaluate(x_test, y_test)\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99e74fe4",
   "metadata": {},
   "source": [
    "활성화 함수{ => 이전 레이어의 값을 다음 레이어로 전달하는 역할\n",
    "\n",
    "    softmax:다중분류, 확률을 리턴,총합이 1이다\n",
    "    \n",
    "    relu:음수를 0으로 만든다\n",
    "    \n",
    "    sigmoid: 이진분류, 절댓값이 커질수록 미분계수가 작아져 죄종 계층에 미치는 영향이 줄어든다.=>vanishing gradient problem\n",
    "    \n",
    "             0~1로 압축하기 때문에 확률로 해석이 가능하다.\n",
    "}\n",
    "손실 함수{ => 목표값과 출력값의 차\n",
    "\n",
    "    binary_crossentropy:sigmoid 활성화 함수 다음에 연계되어 사용\n",
    "    \n",
    "    categorical_crossentropy:softmax활성화 함수 다음에 연계되어 사용\n",
    "    \n",
    "    mse:평균 제곱 오차->예측과 타깃 사이의 거리의 제곱 \n",
    "    \n",
    "    mae:평균 절대 오차->예측과 타깃 사이 거리의 절댓값\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1c5184aa",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}