103-example_overfitting_and_underfitting.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TensorFlow2.0教程-过拟合和欠拟合"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.0.0-alpha0\n"
     ]
    }
   ],
   "source": [
    "from __future__ import absolute_import, division, print_function\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f92faa7c978>]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAF5tJREFUeJzt3XuQnXV9x/H3N7vZ3MnFbCBkExPqgsTKJW4RqyItFxNQMqOgybQVEc30grXVaQeGFpXOdEAdbZ1GIa2KpZWI1Gq00VC51BmFmAUEk0Bgs4RkwyWbhIRLEnL79o/z7Hr2cM4+zznnOZfndz6vmZ08l995nt/veU4+5znP5XfM3RERkbCMaXQFREQkfQp3EZEAKdxFRAKkcBcRCZDCXUQkQAp3EZEAKdxFRAKkcBcRCZDCXUQkQO2NWvHMmTN9/vz5jVq9iEgmPfTQQ7vdvTOuXMPCff78+fT29jZq9SIimWRmzyQpp9MyIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBig13M/umme0ys40l5puZfdXM+szsMTNblH41RUSkHEmO3G8DFo8yfwnQHf2tAL5efbVERKQasfe5u/vPzWz+KEWWAv/uud/re9DMppnZbHd/LqU6jvDyoSO89XN3F513ZtdUzp43nQkdbfx6+z5On30CnVPGcfNPn+Cz71/I53+0mRXnncKD/Xv40hVncvDwMVZv2MG7u2dy2klTePL5l1ny1tkA3LdlF92zJtM1fSIAv3p6L9MnjmXjs/u5eOFJPLf/ELtfeY1zT3kDAC+8dIjHBvZz0cITWd+/hynjx7L+6T3MmNTB0rPmvK6uP3r0Wc7r7mTqxLEl23r3puc5a940Zk0ZT++2vRw77gy+8hrvO+Pk4TIPb3+Rce1jeMvJUyvepo2yvn8Pd29+geuWvJn2ttxxxq6XDvFotB0r8ey+g9zww0185cNncu8Tu1h61hx+uvE5eubPYObkcVXXecO2vUydMJZTT5zCozv2McaMt3aN3Pb3PvECp88+gdlTJyRa5pFjx/nvR3Zy+aIuxoyxqutYT8/seZXtew/w7u7YZ2qkztJ4iGkOsCNvfCCa9rpwN7MV5I7umTdvXkUr+9r9W0vOe3RgP48O7B8ef6B/z/Dw53+0GYBVP+8H4OKv/Hx43h2/2j48/PDfX8SMSR1c9a0NTOxoY/ONuS8tH7r1geEyH1i0m+8/vBOAbTddCsDlt/ySHXsPsu2mS/nwqgdH1Kt71hQWnnzC8Pi23a/yyTse4fzTOrntqnOKtuXIseOsuP0hfqdzEvd85nwuv+W36z+zaxpzZ+Q+dD7wtV+OqEeWDG2nk6dN4Op3LQDgilsf4Jk9B+j/x0sqCrrfv+legOEDgM4p4/jT/3iYM7qmsuaad1Vd5yui/bDtpktZuvIXw8P5PnZbL51TxrHh+gsTLfPW/9vKl+5+kjYzPvi2rqrrWE/v+eL9QDbff6Gr6wVVd1/l7j3u3tPZWdkn/e6XX0u5ViMdPXZ8ePjA4WNFy+x66fV12LH3YMllHjwycjmHjubGn9t3qORrjkc/XF5sua8dLV6vrNp34PDw8DN7DgBgKR3AvnzoKAADL5beP7UwWMb7dPcrufbvP3ikVtWRFpRGuO8E5uaNd0XTRESkQdII9zXAR6K7Zs4F9tfqfLuIiCQTe87dzO4AzgdmmtkA8FlgLIC73wKsBS4B+oADwFW1qqyIiCST5G6Z5THzHfiL1GokIiJVy9wTqt7oCoiIZEDmwl1EROIp3EVEAqRwFxEJkMJdRCRACncRkQAp3AtUezeOe7ElFF+qV7i2oqsITCu0UaSWFO4VqDSUhxjJO06pdl1ZkB/kafUp8/p1NO92HGpz89ZQskjh3qTK+QAQESmkcBcRCZDCXUQkQAp3EZEAKdxFRAKUuXBv4pseRESaRubCXURE4incRUQCpHAXEQmQwl1EJEAK95QVu+Bb6iLwaBeHR+t2oBWuKbdCG0VqKXPh3gx9rVTbNUA5/ae0QjcEtepPZuQ6mn87Nn8NJUsyF+7NoJ4fMM3wYVZr9bi9tZk7DhvS/DWULFG4N6lWOGIXkdpRuDdABg4iRSTjFO51UOp0b6WngXVMLyJxFO4iIgFSuIuIBCh74a7z1SIisbIX7iIiEkvhLiISIIW7iEiAEoW7mS02sy1m1mdm1xaZP8/M7jOzR8zsMTO7JP2qiohIUrHhbmZtwEpgCbAQWG5mCwuK/R1wp7ufDSwDvpZ2Reul2geMir28ko7Dyl1HaLLQXYBIM0ty5H4O0Ofu/e5+GFgNLC0o48AJ0fBU4Nn0qhieDPRh1TCtuGnU1YTUQnuCMnOAHXnjA8DbC8p8DrjbzD4JTAIuTKV2TaqeB5WtcABbj87RsrAZ9W1F0pTWBdXlwG3u3gVcAtxuZq9btpmtMLNeM+sdHBxMadVh0tG9iFQjSbjvBObmjXdF0/JdDdwJ4O4PAOOBmYULcvdV7t7j7j2dnZ2V1VhERGIlCfcNQLeZLTCzDnIXTNcUlNkOXABgZqeTC3cdmouINEhsuLv7UeAaYB3wOLm7YjaZ2Y1mdllU7DPAJ8zsUeAO4KOuE4giIg2T5IIq7r4WWFsw7Ya84c3AO9OtmoiIVCpzT6jq64CISLzMhbuIiMRTuIuIBEjhLiISIIV7yordJFTqOsFo1w9Gu9eoFe5DaoEmitSUwr0CxZ4eLeeJ0nIePm2FJ1Xr0bdKFjajtcLOlrrJXLjr9nkRkXiZC/dmoI7D0qWOw3J04CJpUriLiARI4S4iEiCFu4hIgBTuKSnndKnOrIpIrWUu3LMYjKVucKv0xjfdMScicTIX7iIiEk/hLiISIIW7iEiAFO4Fqn2gptirK+lbZtR1ZPHCQ5laoY0itaRwFxEJUObCvVlvFKlVx2GtphU7z2rBJksdZC7c9W1dRCRe5sK9GdS147D6raphark9h5atc/jSahTuIiIBUrhL0HQ+W1qVwl1EJEAKdxGRAGUu3HVhTEQkXubCXURE4incRUQCpHAXEQmQwj1lxa4JlLpOMNqv3Y/WgVm1nZtlQSu0UaSWFO4VKHbvdDm3U6sfmpHqcS+67neXVpMo3M1ssZltMbM+M7u2RJkPmdlmM9tkZt9Jt5oiIlKO9rgCZtYGrAQuAgaADWa2xt0355XpBq4D3unuL5rZrFpVWF/WRUTiJTlyPwfoc/d+dz8MrAaWFpT5BLDS3V8EcPdd6VZTRETKkSTc5wA78sYHomn5TgVONbNfmNmDZra42ILMbIWZ9ZpZ7+DgYGU1bgLqFTJd9dieevhNWk1aF1TbgW7gfGA58K9mNq2wkLuvcvced+/p7OxMadUiIlIoSbjvBObmjXdF0/INAGvc/Yi7Pw08SS7sRUSkAZKE+wag28wWmFkHsAxYU1DmB+SO2jGzmeRO0/SnWE8RESlDbLi7+1HgGmAd8Dhwp7tvMrMbzeyyqNg6YI+ZbQbuA/7G3ffUosKjPfgjIiI5sbdCArj7WmBtwbQb8oYd+HT015LK+cjR55OI1JqeUK2DUk9HWoWPTVpLPLcqItVQuBeo9qi6WJ8olfQtU+46QqNvNyLVyVy4V3q0W2u16lum1bTipmnFNkvtZS7cRUQkXubCXXfLiIjEy1y4i4hIPIW7iEiAFO4VqOfdKq1wGqoeLczCdsxAFSVDFO4iIgFSuIuIBEjhLiISoMyFu05LiojEy1y4i4hIPIV7gWq/GRS746HUnRqjrWu0OydCvqsi7aZlYVtloIqSQQp3EZEAKdwLJOnEqViXu+V1aJa8bLN2lJYmKzGcyrJt6N/m3Y5DNWviKkoGKdxFRAKUvXDXCUoRkVjZC3cREYmlcBcRCZDCvQLqOCxd6jgsJwNVlAxRuIuIBEjhLiISIIW7iEiAMhfu9TzfLSKSVZkLdxERiadwL1CL7wUll1nhylrhropWaKNILWUu3Iv16yLhaMX+VVqxzVJ7mQt3ERGJp3Cvg5IHZhUeselIT0TiJAp3M1tsZlvMrM/Mrh2l3AfNzM2sJ70qjqS7ZURE4sWGu5m1ASuBJcBCYLmZLSxSbgrwKWB92pUUEZHyJDlyPwfoc/d+dz8MrAaWFin3D8DNwKEU6yciIhVIEu5zgB154wPRtGFmtgiY6+7/k2LdmlY9b9NrhZNQ9dieWdiOOuUoaar6gqqZjQG+DHwmQdkVZtZrZr2Dg4PVrlpEREpIEu47gbl5413RtCFTgN8F7jezbcC5wJpiF1XdfZW797h7T2dnZ+W1FhGRUSUJ9w1At5ktMLMOYBmwZmimu+9395nuPt/d5wMPApe5e28tKqwnF0VE4sWGu7sfBa4B1gGPA3e6+yYzu9HMLqt1BUVEpHztSQq5+1pgbcG0G0qUPb/6amVXsW8W6lumfLq4KFIdPaFagWJPiJbz0Gg5T5i2wsOo9XjiNgvbUf0mSZoU7iIiAVK4i4gESOEuIhKgzIV7K1xMFBGpVubCXURE4incRUQCpHCvgDoOS1ctt+fQsrOwHXVvv6RJ4S4iEiCFuwRNP0korSpz4a6vriIi8TIX7rXmVZ4ALvbhU2qRla6pFT7gdMurSHUU7iIiAcpcuDdr50rqDCwdzbp/a8l0YUBqIHPhLiIi8RTuIiIBUrinpJwLgLpWKCK1lrlwb4U7RUREqpW5cM+iUtfLKr2M1ooXHUWkPAp3EZEAKdwrUM8TQ63wME8tT7UNb78MbMdW2NdSPwp3EZEAKdxFRAKUuXCv9VfXapdf7PWV9C0zWj1CvmNoqG3ptbD5t1W1/RmJFJO5cBcRkXgK9wJJuvkoVqRWfcu0Qrcj+bd2pn+bp434pxkN9S3TCvta6kfhLiISIIW7iEiAFO4iIgHKXLjrvgIRkXiZC3cREYmXKNzNbLGZbTGzPjO7tsj8T5vZZjN7zMzuMbM3pl9VERFJKjbczawNWAksARYCy81sYUGxR4Aedz8DuAv4QtoVFRGR5JIcuZ8D9Ll7v7sfBlYDS/MLuPt97n4gGn0Q6Eq3ms1FHYelq7ZP3PqIf5pZK+xrqZ8k4T4H2JE3PhBNK+Vq4CfFZpjZCjPrNbPewcHB5LUUEZGypHpB1cz+GOgBvlhsvruvcvced+/p7OxMc9UiIpKnPUGZncDcvPGuaNoIZnYhcD3wHnd/LZ3q1V/VHYcVnVZ8oZV2GNUKX9/VmZZIdZIcuW8Aus1sgZl1AMuANfkFzOxs4FbgMnfflX41m185faKYOhEprQU3TQs2WeogNtzd/ShwDbAOeBy40903mdmNZnZZVOyLwGTge2b2azNbU2JxIiJSB0lOy+Dua4G1BdNuyBu+MOV6iYhIFfSEqohIgBTuIiIByly46yYKEZF4mQt3ERGJp3AXEQmQwj0l5fSPogd0wqV9K81C4V6JMv//lnrAqdKHmYJ7BqqGeejZ6TcsE3WU7FC4i4gESOGesmJfyyvpW2a00zxBf/NP+Ug7C5sqC3WU7MlguOu/gohInAyGexMocs5bHYdVwUoMp7joZt7iWaijZI/CXUQkQAp3EZEAKdxFRAKkcBcRCVDmwj3o2wBFRFKSuXAXEZF4CncRkQAp3EVEAqRwFxEJkMK9EqNc1C06q0R5XRuOFNkQaV0494J/m1kW6ijZoXAXEQmQwj0tZXQMoj5ESmvJbdOSjZZaU7iLiARI4S4iEiCFu4hIgBTuIiIBUriLiAQoc+Gue4FFROJlLtxFRCReonA3s8VmtsXM+szs2iLzx5nZd6P5681sftoVFRGR5GLD3czagJXAEmAhsNzMFhYUuxp40d3fBHwFuDntioqISHJJjtzPAfrcvd/dDwOrgaUFZZYC346G7wIuMDM9dyci0iDmMT00mdnlwGJ3/3g0/ifA2939mrwyG6MyA9H41qjM7lLL7enp8d7e3rIr/PFv9/Kzx18o+3VJdU2fwISxbTy16xUAumdNBhgeL1Q4/5TOSfQPvjqizKSONk6eNmF4/MDhY+zcd3DE6wsdd2drtJzuWZNHrH/m5A6mT+wYsd5Sy2lm+W0qth3bKjg+KNxP48eO4dCR4yPWUY387V1s2zvQV+Y+KbYdsiLL779G+ssLunn/mSdX9Foze8jde+LKtVe09AqZ2QpgBcC8efMqWsaHerpSCffJ49o5eOQYx47nPtymTRzLvgNHOKNrKpB7086bMZHuE38bOlMnjGX/wSP0vHE62/ceYNfLrw3PbxtjPPH8y7z5pCn0D75KR9sYDh/Lhcp5p3ZSmFM79x3kzK6pzJk+gVK2Dr7KqSdO5k0F4f5782cML69v8BXGtY8ZrkeWDLXpPad2MmlcGzByO1bi8LHjPLPnAIvmTePh7fv4g9Nm8ZONz3P67BNYMHNiKnU+YXw73SdO5pk9Bzju/rpt37frFU6ZOSnxPpkzfQL3bxnkgjfPYtzYbN3jcOjoMXbsPZjJ918jTZ0wtubrSBLuO4G5eeNd0bRiZQbMrB2YCuwpXJC7rwJWQe7IvZIKX/yWk9h206WVvFREpGUkOUzYAHSb2QIz6wCWAWsKyqwBroyGLwfu9bjzPSIiUjOxR+7uftTMrgHWAW3AN919k5ndCPS6+xrgG8DtZtYH7CX3ASAiIg2S6Jy7u68F1hZMuyFv+BBwRbpVExGRSmXr6o2IiCSicBcRCZDCXUQkQAp3EZEAKdxFRAIU2/1AzVZsNgg8U+HLZwIluzYIlNrcGtTm1lBNm9/o7p1xhRoW7tUws94kfSuERG1uDWpza6hHm3VaRkQkQAp3EZEAZTXcVzW6Ag2gNrcGtbk11LzNmTznLiIio8vqkbuIiIwic+Ee92PdWWFmc83sPjPbbGabzOxT0fQZZva/ZvZU9O/0aLqZ2Vejdj9mZovylnVlVP4pM7uy1DqbhZm1mdkjZvbjaHxB9MPqfdEPrXdE00v+8LqZXRdN32Jm721MS5Ixs2lmdpeZPWFmj5vZO0Lfz2b219H7eqOZ3WFm40Pbz2b2TTPbFf0S3dC01Parmb3NzH4TvearZmX+NJm7Z+aPXJfDW4FTgA7gUWBho+tVYVtmA4ui4SnAk+R+gPwLwLXR9GuBm6PhS4CfAAacC6yPps8A+qN/p0fD0xvdvpi2fxr4DvDjaPxOYFk0fAvwZ9HwnwO3RMPLgO9GwwujfT8OWBC9J9oa3a5R2vtt4OPRcAcwLeT9DMwBngYm5O3fj4a2n4HzgEXAxrxpqe1X4FdRWYteu6Ss+jV6A5W5Md8BrMsbvw64rtH1SqltPwQuArYAs6Nps4Et0fCtwPK88lui+cuBW/OmjyjXbH/kfsnrHuAPgR9Hb9zdQHvhPib3GwLviIbbo3JWuN/zyzXbH7lfJXua6PpW4f4LcT9H4b4jCqz2aD+/N8T9DMwvCPdU9ms074m86SPKJfnL2mmZoTfNkIFoWqZFX0PPBtYDJ7r7c9Gs54ETo+FSbc/aNvkn4G+B49H4G4B97n40Gs+v/3Dbovn7o/JZavMCYBD4VnQq6t/MbBIB72d33wl8CdgOPEduvz1E2Pt5SFr7dU40XDg9sayFe3DMbDLwX8BfuftL+fM895EdzO1MZvY+YJe7P9ToutRRO7mv7l9397OBV8l9XR8W4H6eDiwl98F2MjAJWNzQSjVAo/dr1sI9yY91Z4aZjSUX7P/p7t+PJr9gZrOj+bOBXdH0Um3P0jZ5J3CZmW0DVpM7NfPPwDTL/bA6jKz/cNts5A+vZ6nNA8CAu6+Pxu8iF/Yh7+cLgafdfdDdjwDfJ7fvQ97PQ9Larzuj4cLpiWUt3JP8WHcmRFe+vwE87u5fzpuV/2PjV5I7Fz80/SPRVfdzgf3R1791wMVmNj06Yro4mtZ03P06d+9y9/nk9t297v5HwH3kflgdXt/mYj+8vgZYFt1lsQDoJnfxqem4+/PADjM7LZp0AbCZgPczudMx55rZxOh9PtTmYPdznlT2azTvJTM7N9qGH8lbVjKNviBRwQWMS8jdWbIVuL7R9amiHe8i95XtMeDX0d8l5M413gM8BfwMmBGVN2Bl1O7fAD15y/oY0Bf9XdXotiVs//n89m6ZU8j9p+0DvgeMi6aPj8b7ovmn5L3++mhbbKHMuwga0NazgN5oX/+A3F0RQe9n4PPAE8BG4HZyd7wEtZ+BO8hdUzhC7hva1WnuV6An2n5bgX+h4KJ83J+eUBURCVDWTsuIiEgCCncRkQAp3EVEAqRwFxEJkMJdRCRACncRkQAp3EVEAqRwFxEJ0P8D0v7P5vtsweQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "NUM_WORDS = 10000\n",
    "(train_data, train_labels), (test_data, test_labels) = keras.datasets.imdb.load_data(num_words=NUM_WORDS)\n",
    "\n",
    "def multi_hot_sequences(sequences, dimension):\n",
    "    results = np.zeros((len(sequences), dimension))\n",
    "    for i, word_indices in enumerate(sequences):\n",
    "        results[i, word_indices] = 1.0\n",
    "    return results\n",
    "\n",
    "train_data = multi_hot_sequences(train_data, dimension=NUM_WORDS)\n",
    "test_data = multi_hot_sequences(test_data, dimension=NUM_WORDS)\n",
    "plt.plot(train_data[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "防止过度拟合的最简单方法是减小模型的大小，即模型中可学习参数的数量。\n",
    "\n",
    "深度学习模型往往善于适应训练数据，但真正的挑战是概括，而不是适合。\n",
    "\n",
    "另一方面，如果网络具有有限的记忆资源，则将不能容易地学习映射。为了最大限度地减少损失，它必须学习具有更强预测能力的压缩表示。同时，如果您使模型太小，则难以适应训练数据。 “太多容量”和“容量不足”之间存在平衡。\n",
    "\n",
    "要找到合适的模型大小，最好从相对较少的图层和参数开始，然后开始增加图层的大小或添加新图层，直到看到验证损失的收益递减为止。\n",
    "\n",
    "我们将在电影评论分类网络上使用Dense图层作为基线创建一个简单模型，然后创建更小和更大的版本，并进行比较。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.创建一个baseline模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_5\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_15 (Dense)             (None, 16)                160016    \n",
      "_________________________________________________________________\n",
      "dense_16 (Dense)             (None, 16)                272       \n",
      "_________________________________________________________________\n",
      "dense_17 (Dense)             (None, 1)                 17        \n",
      "=================================================================\n",
      "Total params: 160,305\n",
      "Trainable params: 160,305\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "import tensorflow.keras.layers as layers\n",
    "baseline_model = keras.Sequential(\n",
    "[\n",
    "    layers.Dense(16, activation='relu', input_shape=(NUM_WORDS,)),\n",
    "    layers.Dense(16, activation='relu'),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "]\n",
    ")\n",
    "baseline_model.compile(optimizer='adam',\n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['accuracy', 'binary_crossentropy'])\n",
    "baseline_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/20\n",
      "25000/25000 - 3s - loss: 0.4808 - accuracy: 0.8096 - binary_crossentropy: 0.4808 - val_loss: 0.3333 - val_accuracy: 0.8762 - val_binary_crossentropy: 0.3333\n",
      "Epoch 2/20\n",
      "25000/25000 - 2s - loss: 0.2450 - accuracy: 0.9126 - binary_crossentropy: 0.2450 - val_loss: 0.2831 - val_accuracy: 0.8882 - val_binary_crossentropy: 0.2831\n",
      "Epoch 3/20\n",
      "25000/25000 - 2s - loss: 0.1806 - accuracy: 0.9374 - binary_crossentropy: 0.1806 - val_loss: 0.2921 - val_accuracy: 0.8832 - val_binary_crossentropy: 0.2921\n",
      "Epoch 4/20\n",
      "25000/25000 - 2s - loss: 0.1450 - accuracy: 0.9511 - binary_crossentropy: 0.1450 - val_loss: 0.3135 - val_accuracy: 0.8788 - val_binary_crossentropy: 0.3135\n",
      "Epoch 5/20\n",
      "25000/25000 - 2s - loss: 0.1187 - accuracy: 0.9614 - binary_crossentropy: 0.1187 - val_loss: 0.3406 - val_accuracy: 0.8752 - val_binary_crossentropy: 0.3406\n",
      "Epoch 6/20\n",
      "25000/25000 - 2s - loss: 0.1000 - accuracy: 0.9676 - binary_crossentropy: 0.1000 - val_loss: 0.3765 - val_accuracy: 0.8692 - val_binary_crossentropy: 0.3765\n",
      "Epoch 7/20\n",
      "25000/25000 - 2s - loss: 0.0827 - accuracy: 0.9750 - binary_crossentropy: 0.0827 - val_loss: 0.4124 - val_accuracy: 0.8659 - val_binary_crossentropy: 0.4124\n",
      "Epoch 8/20\n",
      "25000/25000 - 2s - loss: 0.0685 - accuracy: 0.9815 - binary_crossentropy: 0.0685 - val_loss: 0.4567 - val_accuracy: 0.8610 - val_binary_crossentropy: 0.4567\n",
      "Epoch 9/20\n",
      "25000/25000 - 2s - loss: 0.0581 - accuracy: 0.9857 - binary_crossentropy: 0.0581 - val_loss: 0.4987 - val_accuracy: 0.8597 - val_binary_crossentropy: 0.4987\n",
      "Epoch 10/20\n",
      "25000/25000 - 2s - loss: 0.0481 - accuracy: 0.9889 - binary_crossentropy: 0.0481 - val_loss: 0.5402 - val_accuracy: 0.8569 - val_binary_crossentropy: 0.5402\n",
      "Epoch 11/20\n",
      "25000/25000 - 2s - loss: 0.0392 - accuracy: 0.9923 - binary_crossentropy: 0.0392 - val_loss: 0.5883 - val_accuracy: 0.8540 - val_binary_crossentropy: 0.5883\n",
      "Epoch 12/20\n",
      "25000/25000 - 2s - loss: 0.0310 - accuracy: 0.9946 - binary_crossentropy: 0.0310 - val_loss: 0.6316 - val_accuracy: 0.8534 - val_binary_crossentropy: 0.6316\n",
      "Epoch 13/20\n",
      "25000/25000 - 2s - loss: 0.0242 - accuracy: 0.9964 - binary_crossentropy: 0.0242 - val_loss: 0.6779 - val_accuracy: 0.8515 - val_binary_crossentropy: 0.6779\n",
      "Epoch 14/20\n",
      "25000/25000 - 2s - loss: 0.0185 - accuracy: 0.9978 - binary_crossentropy: 0.0185 - val_loss: 0.7149 - val_accuracy: 0.8510 - val_binary_crossentropy: 0.7149\n",
      "Epoch 15/20\n",
      "25000/25000 - 2s - loss: 0.0143 - accuracy: 0.9989 - binary_crossentropy: 0.0143 - val_loss: 0.7571 - val_accuracy: 0.8496 - val_binary_crossentropy: 0.7571\n",
      "Epoch 16/20\n",
      "25000/25000 - 2s - loss: 0.0111 - accuracy: 0.9994 - binary_crossentropy: 0.0111 - val_loss: 0.7954 - val_accuracy: 0.8492 - val_binary_crossentropy: 0.7954\n",
      "Epoch 17/20\n",
      "25000/25000 - 2s - loss: 0.0087 - accuracy: 0.9997 - binary_crossentropy: 0.0087 - val_loss: 0.8301 - val_accuracy: 0.8497 - val_binary_crossentropy: 0.8301\n",
      "Epoch 18/20\n",
      "25000/25000 - 3s - loss: 0.0068 - accuracy: 0.9998 - binary_crossentropy: 0.0068 - val_loss: 0.8629 - val_accuracy: 0.8490 - val_binary_crossentropy: 0.8629\n",
      "Epoch 19/20\n",
      "25000/25000 - 3s - loss: 0.0055 - accuracy: 0.9999 - binary_crossentropy: 0.0055 - val_loss: 0.8937 - val_accuracy: 0.8492 - val_binary_crossentropy: 0.8937\n",
      "Epoch 20/20\n",
      "25000/25000 - 3s - loss: 0.0044 - accuracy: 0.9999 - binary_crossentropy: 0.0044 - val_loss: 0.9217 - val_accuracy: 0.8488 - val_binary_crossentropy: 0.9217\n"
     ]
    }
   ],
   "source": [
    "baseline_history = baseline_model.fit(train_data, train_labels,\n",
    "                                     epochs=20, batch_size=512,\n",
    "                                     validation_data=(test_data, test_labels),\n",
    "                                     verbose=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.创建一个小模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_18 (Dense)             (None, 4)                 40004     \n",
      "_________________________________________________________________\n",
      "dense_19 (Dense)             (None, 4)                 20        \n",
      "_________________________________________________________________\n",
      "dense_20 (Dense)             (None, 1)                 5         \n",
      "=================================================================\n",
      "Total params: 40,029\n",
      "Trainable params: 40,029\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "small_model = keras.Sequential(\n",
    "[\n",
    "    layers.Dense(4, activation='relu', input_shape=(NUM_WORDS,)),\n",
    "    layers.Dense(4, activation='relu'),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "]\n",
    ")\n",
    "small_model.compile(optimizer='adam',\n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['accuracy', 'binary_crossentropy'])\n",
    "small_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/20\n",
      "25000/25000 - 3s - loss: 0.6170 - accuracy: 0.6609 - binary_crossentropy: 0.6170 - val_loss: 0.5217 - val_accuracy: 0.8034 - val_binary_crossentropy: 0.5217\n",
      "Epoch 2/20\n",
      "25000/25000 - 2s - loss: 0.4356 - accuracy: 0.8661 - binary_crossentropy: 0.4356 - val_loss: 0.3979 - val_accuracy: 0.8781 - val_binary_crossentropy: 0.3979\n",
      "Epoch 3/20\n",
      "25000/25000 - 2s - loss: 0.3002 - accuracy: 0.9146 - binary_crossentropy: 0.3002 - val_loss: 0.3160 - val_accuracy: 0.8866 - val_binary_crossentropy: 0.3160\n",
      "Epoch 4/20\n",
      "25000/25000 - 2s - loss: 0.2255 - accuracy: 0.9322 - binary_crossentropy: 0.2255 - val_loss: 0.2930 - val_accuracy: 0.8880 - val_binary_crossentropy: 0.2930\n",
      "Epoch 5/20\n",
      "25000/25000 - 2s - loss: 0.1884 - accuracy: 0.9416 - binary_crossentropy: 0.1884 - val_loss: 0.2901 - val_accuracy: 0.8858 - val_binary_crossentropy: 0.2901\n",
      "Epoch 6/20\n",
      "25000/25000 - 2s - loss: 0.1632 - accuracy: 0.9507 - binary_crossentropy: 0.1632 - val_loss: 0.2918 - val_accuracy: 0.8848 - val_binary_crossentropy: 0.2918\n",
      "Epoch 7/20\n",
      "25000/25000 - 2s - loss: 0.1449 - accuracy: 0.9560 - binary_crossentropy: 0.1449 - val_loss: 0.2991 - val_accuracy: 0.8817 - val_binary_crossentropy: 0.2991\n",
      "Epoch 8/20\n",
      "25000/25000 - 2s - loss: 0.1295 - accuracy: 0.9618 - binary_crossentropy: 0.1295 - val_loss: 0.3087 - val_accuracy: 0.8796 - val_binary_crossentropy: 0.3087\n",
      "Epoch 9/20\n",
      "25000/25000 - 2s - loss: 0.1167 - accuracy: 0.9669 - binary_crossentropy: 0.1167 - val_loss: 0.3210 - val_accuracy: 0.8772 - val_binary_crossentropy: 0.3210\n",
      "Epoch 10/20\n",
      "25000/25000 - 2s - loss: 0.1059 - accuracy: 0.9709 - binary_crossentropy: 0.1059 - val_loss: 0.3325 - val_accuracy: 0.8748 - val_binary_crossentropy: 0.3325\n",
      "Epoch 11/20\n",
      "25000/25000 - 2s - loss: 0.0961 - accuracy: 0.9742 - binary_crossentropy: 0.0961 - val_loss: 0.3468 - val_accuracy: 0.8727 - val_binary_crossentropy: 0.3468\n",
      "Epoch 12/20\n",
      "25000/25000 - 2s - loss: 0.0877 - accuracy: 0.9780 - binary_crossentropy: 0.0877 - val_loss: 0.3602 - val_accuracy: 0.8715 - val_binary_crossentropy: 0.3602\n",
      "Epoch 13/20\n",
      "25000/25000 - 2s - loss: 0.0800 - accuracy: 0.9807 - binary_crossentropy: 0.0800 - val_loss: 0.3759 - val_accuracy: 0.8693 - val_binary_crossentropy: 0.3759\n",
      "Epoch 14/20\n",
      "25000/25000 - 2s - loss: 0.0727 - accuracy: 0.9837 - binary_crossentropy: 0.0727 - val_loss: 0.3923 - val_accuracy: 0.8690 - val_binary_crossentropy: 0.3923\n",
      "Epoch 15/20\n",
      "25000/25000 - 2s - loss: 0.0667 - accuracy: 0.9858 - binary_crossentropy: 0.0667 - val_loss: 0.4089 - val_accuracy: 0.8672 - val_binary_crossentropy: 0.4089\n",
      "Epoch 16/20\n",
      "25000/25000 - 2s - loss: 0.0610 - accuracy: 0.9876 - binary_crossentropy: 0.0610 - val_loss: 0.4255 - val_accuracy: 0.8646 - val_binary_crossentropy: 0.4255\n",
      "Epoch 17/20\n",
      "25000/25000 - 2s - loss: 0.0555 - accuracy: 0.9899 - binary_crossentropy: 0.0555 - val_loss: 0.4422 - val_accuracy: 0.8651 - val_binary_crossentropy: 0.4422\n",
      "Epoch 18/20\n",
      "25000/25000 - 2s - loss: 0.0509 - accuracy: 0.9912 - binary_crossentropy: 0.0509 - val_loss: 0.4614 - val_accuracy: 0.8626 - val_binary_crossentropy: 0.4614\n",
      "Epoch 19/20\n",
      "25000/25000 - 2s - loss: 0.0466 - accuracy: 0.9925 - binary_crossentropy: 0.0466 - val_loss: 0.4780 - val_accuracy: 0.8622 - val_binary_crossentropy: 0.4780\n",
      "Epoch 20/20\n",
      "25000/25000 - 2s - loss: 0.0426 - accuracy: 0.9936 - binary_crossentropy: 0.0426 - val_loss: 0.4976 - val_accuracy: 0.8608 - val_binary_crossentropy: 0.4976\n"
     ]
    }
   ],
   "source": [
    "small_history = small_model.fit(train_data, train_labels,\n",
    "                                     epochs=20, batch_size=512,\n",
    "                                     validation_data=(test_data, test_labels),\n",
    "                                     verbose=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.创建一个大模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_7\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_21 (Dense)             (None, 512)               5120512   \n",
      "_________________________________________________________________\n",
      "dense_22 (Dense)             (None, 512)               262656    \n",
      "_________________________________________________________________\n",
      "dense_23 (Dense)             (None, 1)                 513       \n",
      "=================================================================\n",
      "Total params: 5,383,681\n",
      "Trainable params: 5,383,681\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "big_model = keras.Sequential(\n",
    "[\n",
    "    layers.Dense(512, activation='relu', input_shape=(NUM_WORDS,)),\n",
    "    layers.Dense(512, activation='relu'),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "]\n",
    ")\n",
    "big_model.compile(optimizer='adam',\n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['accuracy', 'binary_crossentropy'])\n",
    "big_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/20\n",
      "25000/25000 - 7s - loss: 0.3523 - accuracy: 0.8466 - binary_crossentropy: 0.3523 - val_loss: 0.2936 - val_accuracy: 0.8808 - val_binary_crossentropy: 0.2936\n",
      "Epoch 2/20\n",
      "25000/25000 - 7s - loss: 0.1419 - accuracy: 0.9489 - binary_crossentropy: 0.1419 - val_loss: 0.3197 - val_accuracy: 0.8742 - val_binary_crossentropy: 0.3197\n",
      "Epoch 3/20\n",
      "25000/25000 - 7s - loss: 0.0421 - accuracy: 0.9892 - binary_crossentropy: 0.0421 - val_loss: 0.4467 - val_accuracy: 0.8694 - val_binary_crossentropy: 0.4467\n",
      "Epoch 4/20\n",
      "25000/25000 - 7s - loss: 0.0054 - accuracy: 0.9994 - binary_crossentropy: 0.0054 - val_loss: 0.5972 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.5972\n",
      "Epoch 5/20\n",
      "25000/25000 - 7s - loss: 0.0010 - accuracy: 1.0000 - binary_crossentropy: 0.0010 - val_loss: 0.6727 - val_accuracy: 0.8707 - val_binary_crossentropy: 0.6727\n",
      "Epoch 6/20\n",
      "25000/25000 - 6s - loss: 2.6156e-04 - accuracy: 1.0000 - binary_crossentropy: 2.6156e-04 - val_loss: 0.7189 - val_accuracy: 0.8710 - val_binary_crossentropy: 0.7189\n",
      "Epoch 7/20\n",
      "25000/25000 - 7s - loss: 5.5455e-04 - accuracy: 1.0000 - binary_crossentropy: 5.5455e-04 - val_loss: 0.7534 - val_accuracy: 0.8698 - val_binary_crossentropy: 0.7534\n",
      "Epoch 8/20\n",
      "25000/25000 - 6s - loss: 1.0939e-04 - accuracy: 1.0000 - binary_crossentropy: 1.0939e-04 - val_loss: 0.7652 - val_accuracy: 0.8708 - val_binary_crossentropy: 0.7653\n",
      "Epoch 9/20\n",
      "25000/25000 - 6s - loss: 7.9712e-05 - accuracy: 1.0000 - binary_crossentropy: 7.9712e-05 - val_loss: 0.7890 - val_accuracy: 0.8711 - val_binary_crossentropy: 0.7890\n",
      "Epoch 10/20\n",
      "25000/25000 - 6s - loss: 6.1083e-05 - accuracy: 1.0000 - binary_crossentropy: 6.1083e-05 - val_loss: 0.8069 - val_accuracy: 0.8706 - val_binary_crossentropy: 0.8069\n",
      "Epoch 11/20\n",
      "25000/25000 - 6s - loss: 4.8950e-05 - accuracy: 1.0000 - binary_crossentropy: 4.8950e-05 - val_loss: 0.8233 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.8233\n",
      "Epoch 12/20\n",
      "25000/25000 - 6s - loss: 4.0149e-05 - accuracy: 1.0000 - binary_crossentropy: 4.0149e-05 - val_loss: 0.8387 - val_accuracy: 0.8706 - val_binary_crossentropy: 0.8387\n",
      "Epoch 13/20\n",
      "25000/25000 - 6s - loss: 3.3564e-05 - accuracy: 1.0000 - binary_crossentropy: 3.3564e-05 - val_loss: 0.8517 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.8517\n",
      "Epoch 14/20\n",
      "25000/25000 - 7s - loss: 2.8433e-05 - accuracy: 1.0000 - binary_crossentropy: 2.8433e-05 - val_loss: 0.8650 - val_accuracy: 0.8706 - val_binary_crossentropy: 0.8650\n",
      "Epoch 15/20\n",
      "25000/25000 - 7s - loss: 2.4336e-05 - accuracy: 1.0000 - binary_crossentropy: 2.4336e-05 - val_loss: 0.8768 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.8768\n",
      "Epoch 16/20\n",
      "25000/25000 - 7s - loss: 2.1041e-05 - accuracy: 1.0000 - binary_crossentropy: 2.1041e-05 - val_loss: 0.8884 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.8884\n",
      "Epoch 17/20\n",
      "25000/25000 - 7s - loss: 1.8337e-05 - accuracy: 1.0000 - binary_crossentropy: 1.8337e-05 - val_loss: 0.8994 - val_accuracy: 0.8703 - val_binary_crossentropy: 0.8994\n",
      "Epoch 18/20\n",
      "25000/25000 - 6s - loss: 1.6106e-05 - accuracy: 1.0000 - binary_crossentropy: 1.6106e-05 - val_loss: 0.9094 - val_accuracy: 0.8703 - val_binary_crossentropy: 0.9094\n",
      "Epoch 19/20\n",
      "25000/25000 - 6s - loss: 1.4224e-05 - accuracy: 1.0000 - binary_crossentropy: 1.4224e-05 - val_loss: 0.9193 - val_accuracy: 0.8703 - val_binary_crossentropy: 0.9193\n",
      "Epoch 20/20\n",
      "25000/25000 - 6s - loss: 1.2638e-05 - accuracy: 1.0000 - binary_crossentropy: 1.2638e-05 - val_loss: 0.9282 - val_accuracy: 0.8704 - val_binary_crossentropy: 0.9282\n"
     ]
    }
   ],
   "source": [
    "big_history = big_model.fit(train_data, train_labels,\n",
    "                                     epochs=20, batch_size=512,\n",
    "                                     validation_data=(test_data, test_labels),\n",
    "                                     verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_history(histories, key='binary_crossentropy'):\n",
    "  plt.figure(figsize=(16,10))\n",
    "    \n",
    "  for name, history in histories:\n",
    "    val = plt.plot(history.epoch, history.history['val_'+key],\n",
    "                   '--', label=name.title()+' Val')\n",
    "    plt.plot(history.epoch, history.history[key], color=val[0].get_color(),\n",
    "             label=name.title()+' Train')\n",
    "\n",
    "  plt.xlabel('Epochs')\n",
    "  plt.ylabel(key.replace('_',' ').title())\n",
    "  plt.legend()\n",
    "\n",
    "  plt.xlim([0,max(history.epoch)])\n",
    "\n",
    "\n",
    "plot_history([('baseline', baseline_history),\n",
    "              ('small', small_history),\n",
    "              ('big', big_history)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "请注意，较大的网络在仅仅一个时期之后几乎立即开始过度拟合，并且更过拟合更严重。 网络容量越大，能够越快地对训练数据进行建模（导致训练损失低），但过度拟合的可能性越大（导致训练和验证损失之间的差异很大）。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.添加l2正则"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_9\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_27 (Dense)             (None, 16)                160016    \n",
      "_________________________________________________________________\n",
      "dense_28 (Dense)             (None, 16)                272       \n",
      "_________________________________________________________________\n",
      "dense_29 (Dense)             (None, 1)                 17        \n",
      "=================================================================\n",
      "Total params: 160,305\n",
      "Trainable params: 160,305\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/20\n",
      "25000/25000 - 3s - loss: 0.5264 - accuracy: 0.8019 - binary_crossentropy: 0.4874 - val_loss: 0.3828 - val_accuracy: 0.8769 - val_binary_crossentropy: 0.3415\n",
      "Epoch 2/20\n",
      "25000/25000 - 3s - loss: 0.3070 - accuracy: 0.9073 - binary_crossentropy: 0.2607 - val_loss: 0.3356 - val_accuracy: 0.8876 - val_binary_crossentropy: 0.2855\n",
      "Epoch 3/20\n",
      "25000/25000 - 3s - loss: 0.2552 - accuracy: 0.9295 - binary_crossentropy: 0.2023 - val_loss: 0.3374 - val_accuracy: 0.8860 - val_binary_crossentropy: 0.2825\n",
      "Epoch 4/20\n",
      "25000/25000 - 3s - loss: 0.2346 - accuracy: 0.9377 - binary_crossentropy: 0.1777 - val_loss: 0.3500 - val_accuracy: 0.8826 - val_binary_crossentropy: 0.2916\n",
      "Epoch 5/20\n",
      "25000/25000 - 3s - loss: 0.2190 - accuracy: 0.9456 - binary_crossentropy: 0.1591 - val_loss: 0.3635 - val_accuracy: 0.8796 - val_binary_crossentropy: 0.3027\n",
      "Epoch 6/20\n",
      "25000/25000 - 3s - loss: 0.2041 - accuracy: 0.9526 - binary_crossentropy: 0.1426 - val_loss: 0.3764 - val_accuracy: 0.8769 - val_binary_crossentropy: 0.3144\n",
      "Epoch 7/20\n",
      "25000/25000 - 3s - loss: 0.2001 - accuracy: 0.9542 - binary_crossentropy: 0.1368 - val_loss: 0.3963 - val_accuracy: 0.8728 - val_binary_crossentropy: 0.3322\n",
      "Epoch 8/20\n",
      "25000/25000 - 3s - loss: 0.1891 - accuracy: 0.9577 - binary_crossentropy: 0.1240 - val_loss: 0.4038 - val_accuracy: 0.8714 - val_binary_crossentropy: 0.3383\n",
      "Epoch 9/20\n",
      "25000/25000 - 3s - loss: 0.1804 - accuracy: 0.9632 - binary_crossentropy: 0.1143 - val_loss: 0.4181 - val_accuracy: 0.8702 - val_binary_crossentropy: 0.3515\n",
      "Epoch 10/20\n",
      "25000/25000 - 3s - loss: 0.1729 - accuracy: 0.9662 - binary_crossentropy: 0.1057 - val_loss: 0.4310 - val_accuracy: 0.8684 - val_binary_crossentropy: 0.3635\n",
      "Epoch 11/20\n",
      "25000/25000 - 3s - loss: 0.1717 - accuracy: 0.9672 - binary_crossentropy: 0.1029 - val_loss: 0.4478 - val_accuracy: 0.8651 - val_binary_crossentropy: 0.3781\n",
      "Epoch 12/20\n",
      "25000/25000 - 3s - loss: 0.1612 - accuracy: 0.9714 - binary_crossentropy: 0.0910 - val_loss: 0.4653 - val_accuracy: 0.8650 - val_binary_crossentropy: 0.3949\n",
      "Epoch 13/20\n",
      "25000/25000 - 3s - loss: 0.1552 - accuracy: 0.9749 - binary_crossentropy: 0.0844 - val_loss: 0.4799 - val_accuracy: 0.8628 - val_binary_crossentropy: 0.4088\n",
      "Epoch 14/20\n",
      "25000/25000 - 3s - loss: 0.1500 - accuracy: 0.9770 - binary_crossentropy: 0.0786 - val_loss: 0.4904 - val_accuracy: 0.8616 - val_binary_crossentropy: 0.4186\n",
      "Epoch 15/20\n",
      "25000/25000 - 3s - loss: 0.1436 - accuracy: 0.9801 - binary_crossentropy: 0.0716 - val_loss: 0.5095 - val_accuracy: 0.8602 - val_binary_crossentropy: 0.4373\n",
      "Epoch 16/20\n",
      "25000/25000 - 3s - loss: 0.1401 - accuracy: 0.9820 - binary_crossentropy: 0.0676 - val_loss: 0.5226 - val_accuracy: 0.8580 - val_binary_crossentropy: 0.4496\n",
      "Epoch 17/20\n",
      "25000/25000 - 3s - loss: 0.1368 - accuracy: 0.9820 - binary_crossentropy: 0.0634 - val_loss: 0.5375 - val_accuracy: 0.8596 - val_binary_crossentropy: 0.4639\n",
      "Epoch 18/20\n",
      "25000/25000 - 3s - loss: 0.1315 - accuracy: 0.9844 - binary_crossentropy: 0.0580 - val_loss: 0.5472 - val_accuracy: 0.8600 - val_binary_crossentropy: 0.4735\n",
      "Epoch 19/20\n",
      "25000/25000 - 3s - loss: 0.1314 - accuracy: 0.9842 - binary_crossentropy: 0.0572 - val_loss: 0.5676 - val_accuracy: 0.8578 - val_binary_crossentropy: 0.4927\n",
      "Epoch 20/20\n",
      "25000/25000 - 3s - loss: 0.1278 - accuracy: 0.9856 - binary_crossentropy: 0.0530 - val_loss: 0.5750 - val_accuracy: 0.8580 - val_binary_crossentropy: 0.5001\n"
     ]
    }
   ],
   "source": [
    "l2_model = keras.Sequential(\n",
    "[\n",
    "    layers.Dense(16, kernel_regularizer=keras.regularizers.l2(0.001), \n",
    "                 activation='relu', input_shape=(NUM_WORDS,)),\n",
    "    layers.Dense(16, kernel_regularizer=keras.regularizers.l2(0.001), \n",
    "                 activation='relu'),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "]\n",
    ")\n",
    "l2_model.compile(optimizer='adam',\n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['accuracy', 'binary_crossentropy'])\n",
    "l2_model.summary()\n",
    "l2_history = l2_model.fit(train_data, train_labels,\n",
    "                                     epochs=20, batch_size=512,\n",
    "                                     validation_data=(test_data, test_labels),\n",
    "                                     verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history([('baseline', baseline_history),\n",
    "              ('l2', l2_history)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.添加dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_10\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_30 (Dense)             (None, 16)                160016    \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 16)                0         \n",
      "_________________________________________________________________\n",
      "dense_31 (Dense)             (None, 16)                272       \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 16)                0         \n",
      "_________________________________________________________________\n",
      "dense_32 (Dense)             (None, 1)                 17        \n",
      "=================================================================\n",
      "Total params: 160,305\n",
      "Trainable params: 160,305\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Train on 25000 samples, validate on 25000 samples\n",
      "Epoch 1/20\n",
      "25000/25000 - 4s - loss: 0.6364 - accuracy: 0.6512 - binary_crossentropy: 0.6364 - val_loss: 0.5510 - val_accuracy: 0.8113 - val_binary_crossentropy: 0.5510\n",
      "Epoch 2/20\n",
      "25000/25000 - 3s - loss: 0.5050 - accuracy: 0.8296 - binary_crossentropy: 0.5050 - val_loss: 0.4348 - val_accuracy: 0.8736 - val_binary_crossentropy: 0.4348\n",
      "Epoch 3/20\n",
      "25000/25000 - 3s - loss: 0.4169 - accuracy: 0.8725 - binary_crossentropy: 0.4169 - val_loss: 0.3729 - val_accuracy: 0.8830 - val_binary_crossentropy: 0.3729\n",
      "Epoch 4/20\n",
      "25000/25000 - 3s - loss: 0.3538 - accuracy: 0.8985 - binary_crossentropy: 0.3538 - val_loss: 0.3368 - val_accuracy: 0.8841 - val_binary_crossentropy: 0.3368\n",
      "Epoch 5/20\n",
      "25000/25000 - 3s - loss: 0.3099 - accuracy: 0.9104 - binary_crossentropy: 0.3099 - val_loss: 0.3291 - val_accuracy: 0.8842 - val_binary_crossentropy: 0.3291\n",
      "Epoch 6/20\n",
      "25000/25000 - 3s - loss: 0.2699 - accuracy: 0.9236 - binary_crossentropy: 0.2699 - val_loss: 0.3193 - val_accuracy: 0.8820 - val_binary_crossentropy: 0.3193\n",
      "Epoch 7/20\n",
      "25000/25000 - 3s - loss: 0.2419 - accuracy: 0.9328 - binary_crossentropy: 0.2419 - val_loss: 0.3168 - val_accuracy: 0.8808 - val_binary_crossentropy: 0.3168\n",
      "Epoch 8/20\n",
      "25000/25000 - 3s - loss: 0.2203 - accuracy: 0.9379 - binary_crossentropy: 0.2203 - val_loss: 0.3318 - val_accuracy: 0.8807 - val_binary_crossentropy: 0.3318\n",
      "Epoch 9/20\n",
      "25000/25000 - 3s - loss: 0.1985 - accuracy: 0.9446 - binary_crossentropy: 0.1985 - val_loss: 0.3471 - val_accuracy: 0.8792 - val_binary_crossentropy: 0.3471\n",
      "Epoch 10/20\n",
      "25000/25000 - 3s - loss: 0.1839 - accuracy: 0.9478 - binary_crossentropy: 0.1839 - val_loss: 0.3624 - val_accuracy: 0.8792 - val_binary_crossentropy: 0.3624\n",
      "Epoch 11/20\n",
      "25000/25000 - 3s - loss: 0.1620 - accuracy: 0.9548 - binary_crossentropy: 0.1620 - val_loss: 0.3806 - val_accuracy: 0.8781 - val_binary_crossentropy: 0.3806\n",
      "Epoch 12/20\n",
      "25000/25000 - 3s - loss: 0.1503 - accuracy: 0.9580 - binary_crossentropy: 0.1503 - val_loss: 0.3941 - val_accuracy: 0.8764 - val_binary_crossentropy: 0.3941\n",
      "Epoch 13/20\n",
      "25000/25000 - 3s - loss: 0.1440 - accuracy: 0.9598 - binary_crossentropy: 0.1440 - val_loss: 0.4231 - val_accuracy: 0.8754 - val_binary_crossentropy: 0.4231\n",
      "Epoch 14/20\n",
      "25000/25000 - 3s - loss: 0.1345 - accuracy: 0.9643 - binary_crossentropy: 0.1345 - val_loss: 0.4272 - val_accuracy: 0.8758 - val_binary_crossentropy: 0.4272\n",
      "Epoch 15/20\n",
      "25000/25000 - 3s - loss: 0.1307 - accuracy: 0.9640 - binary_crossentropy: 0.1307 - val_loss: 0.4550 - val_accuracy: 0.8746 - val_binary_crossentropy: 0.4550\n",
      "Epoch 16/20\n",
      "25000/25000 - 3s - loss: 0.1217 - accuracy: 0.9668 - binary_crossentropy: 0.1217 - val_loss: 0.5008 - val_accuracy: 0.8740 - val_binary_crossentropy: 0.5008\n",
      "Epoch 17/20\n",
      "25000/25000 - 3s - loss: 0.1161 - accuracy: 0.9686 - binary_crossentropy: 0.1161 - val_loss: 0.5079 - val_accuracy: 0.8738 - val_binary_crossentropy: 0.5079\n",
      "Epoch 18/20\n",
      "25000/25000 - 3s - loss: 0.1125 - accuracy: 0.9698 - binary_crossentropy: 0.1125 - val_loss: 0.5297 - val_accuracy: 0.8718 - val_binary_crossentropy: 0.5297\n",
      "Epoch 19/20\n",
      "25000/25000 - 3s - loss: 0.1069 - accuracy: 0.9705 - binary_crossentropy: 0.1069 - val_loss: 0.5379 - val_accuracy: 0.8740 - val_binary_crossentropy: 0.5379\n",
      "Epoch 20/20\n",
      "25000/25000 - 3s - loss: 0.1068 - accuracy: 0.9720 - binary_crossentropy: 0.1068 - val_loss: 0.5721 - val_accuracy: 0.8732 - val_binary_crossentropy: 0.5721\n"
     ]
    }
   ],
   "source": [
    "dpt_model = keras.Sequential(\n",
    "[\n",
    "    layers.Dense(16, activation='relu', input_shape=(NUM_WORDS,)),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(16, activation='relu'),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "]\n",
    ")\n",
    "dpt_model.compile(optimizer='adam',\n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['accuracy', 'binary_crossentropy'])\n",
    "dpt_model.summary()\n",
    "dpt_history = dpt_model.fit(train_data, train_labels,\n",
    "                                     epochs=20, batch_size=512,\n",
    "                                     validation_data=(test_data, test_labels),\n",
    "                                     verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_history([('baseline', baseline_history),\n",
    "              ('dropout', dpt_history)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "防止神经网络中过度拟合的最常用方法：\n",
    "\n",
    "- 获取更多训练数据。\n",
    "- 减少网络容量。\n",
    "- 添加权重正规化。\n",
    "- 添加dropout。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}