Day70-Keras_Mnist_MLP_HW.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 作業目標:¶\n",
    "利用手寫辨識資料說明MLP 中, 神經網路層的多寡跟預測結果的差異"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 作業重點:\n",
    "[練習]請嘗試新增一層或兩層看看最後預測結果是否有差異\n",
    "model.add(Dense(units=256, input_dim=784, kernel_initializer='normal', activation='relu'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "zfI5zCjRPlnc"
   },
   "source": [
    "# 資料預處理"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 20601,
     "status": "ok",
     "timestamp": 1533441753568,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "GyIQ7tpGPlnf",
    "outputId": "f3642b99-3ef2-43fe-ee42-e17b1ef50cbb"
   },
   "outputs": [],
   "source": [
    "from keras.utils import np_utils\n",
    "import numpy as np\n",
    "np.random.seed(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5937,
     "status": "ok",
     "timestamp": 1533441759521,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "OLzoKlJRPlno",
    "outputId": "c75f10a4-951d-4b3b-ad5f-c8fc4dff0e96"
   },
   "outputs": [],
   "source": [
    "#載入手寫辨識的資料集\n",
    "from keras.datasets import mnist\n",
    "(x_train_image,y_train_label),\\\n",
    "(x_test_image,y_test_label)= mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "QEuQwa4nPlnu"
   },
   "outputs": [],
   "source": [
    "#指定測試集與訓練資料集\n",
    "x_Train =x_train_image.reshape(60000, 784).astype('float32')\n",
    "x_Test = x_test_image.reshape(10000, 784).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "YisRq-SxPln0"
   },
   "outputs": [],
   "source": [
    "# normalize inputs from 0-255 to 0-1\n",
    "x_Train_normalize = x_Train / 255\n",
    "x_Test_normalize = x_Test / 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "AHtfmzpfPln4"
   },
   "outputs": [],
   "source": [
    "#把LABEL轉成NUMERICAL Categorical \n",
    "y_Train_OneHot = np_utils.to_categorical(y_train_label)\n",
    "y_Test_OneHot = np_utils.to_categorical(y_test_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0s8PjvqJPln7"
   },
   "source": [
    "# 建立模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "N10v1eX2Pln8"
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "jmaxrsXwPln-"
   },
   "outputs": [],
   "source": [
    "#建立模型\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Dense(units=128, \n",
    "                input_dim=784, \n",
    "                kernel_initializer='normal', \n",
    "                activation='relu'))\n",
    "model.add(Dense(64,activation=\"relu\", kernel_initializer=\"normal\"))\n",
    "\n",
    "model.add(Dense(units=10, kernel_initializer='normal', activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jdoAZgv1PloK"
   },
   "source": [
    "# 訓練模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "5d5wrEL3PloK"
   },
   "outputs": [],
   "source": [
    "model.compile(loss='categorical_crossentropy', \n",
    "              optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 374
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 15426,
     "status": "ok",
     "timestamp": 1533441786624,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "VlpjoqQ9PloM",
    "outputId": "b91454c9-af9a-40fe-9736-b4a1c97462cd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Train on 48000 samples, validate on 12000 samples\n",
      "Epoch 1/10\n",
      "48000/48000 [==============================] - 3s 69us/step - loss: 0.3289 - acc: 0.9055 - val_loss: 0.1624 - val_acc: 0.9523\n",
      "Epoch 2/10\n",
      "48000/48000 [==============================] - 3s 64us/step - loss: 0.1384 - acc: 0.9585 - val_loss: 0.1224 - val_acc: 0.9646\n",
      "Epoch 3/10\n",
      "48000/48000 [==============================] - 3s 58us/step - loss: 0.0951 - acc: 0.9715 - val_loss: 0.1101 - val_acc: 0.9676\n",
      "Epoch 4/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0713 - acc: 0.9782 - val_loss: 0.1102 - val_acc: 0.9667\n",
      "Epoch 5/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0550 - acc: 0.9830 - val_loss: 0.1306 - val_acc: 0.9624\n",
      "Epoch 6/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0443 - acc: 0.9865 - val_loss: 0.0940 - val_acc: 0.9732\n",
      "Epoch 7/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0352 - acc: 0.9890 - val_loss: 0.1020 - val_acc: 0.9720\n",
      "Epoch 8/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0282 - acc: 0.9907 - val_loss: 0.0981 - val_acc: 0.9758\n",
      "Epoch 9/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0269 - acc: 0.9916 - val_loss: 0.1070 - val_acc: 0.9736\n",
      "Epoch 10/10\n",
      "48000/48000 [==============================] - 3s 57us/step - loss: 0.0212 - acc: 0.9932 - val_loss: 0.1206 - val_acc: 0.9729\n"
     ]
    }
   ],
   "source": [
    "train_history =model.fit(x=x_Train_normalize,\n",
    "                         y=y_Train_OneHot,validation_split=0.2, \n",
    "                         epochs=10, batch_size=32,verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pV1Ftdb0PloP"
   },
   "source": [
    "# 以圖形顯示訓練過程"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     }
    },
    "colab_type": "code",
    "id": "EEmuoa1dPloQ"
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "def show_train_history(train_history,train,validation):\n",
    "    plt.plot(train_history.history[train])\n",
    "    plt.plot(train_history.history[validation])\n",
    "    plt.title('Train History')\n",
    "    plt.ylabel(train)\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend(['train', 'validation'], loc='upper left')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1451,
     "status": "ok",
     "timestamp": 1533441789030,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "fvE2WsejPloT",
    "outputId": "bb56e170-8850-409e-bbb2-eb6b5b89e528"
   },
   "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": [
    "show_train_history(train_history,'acc','val_acc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1937,
     "status": "ok",
     "timestamp": 1533441791049,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "zIYyrnFWPloW",
    "outputId": "4bdbd1fa-6f25-49af-88d0-8d35f6e2ecaa"
   },
   "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": [
    "show_train_history(train_history,'loss','val_loss')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "h1z2y0g-PloZ"
   },
   "source": [
    "# 評估模型準確率"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "autoexec": {
      "startup": false,
      "wait_interval": 0
     },
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1425,
     "status": "ok",
     "timestamp": 1533441792584,
     "user": {
      "displayName": "Edward Cheng",
      "photoUrl": "//lh4.googleusercontent.com/-ANnwjGu3IBk/AAAAAAAAAAI/AAAAAAAAAAc/qXXg6Jek9xw/s50-c-k-no/photo.jpg",
      "userId": "112554017642991017343"
     },
     "user_tz": -480
    },
    "id": "vMawurJqPloZ",
    "outputId": "31465f8d-cfad-49b8-dc9e-90d643e6b4a8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 0s 24us/step\n",
      "\n",
      "accuracy= 0.9757\n"
     ]
    }
   ],
   "source": [
    "scores = model.evaluate(x_Test_normalize, y_Test_OneHot)\n",
    "print()\n",
    "print('accuracy=',scores[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "default_view": {},
   "name": "Keras_Mnist_MLP_h256.ipynb",
   "provenance": [],
   "version": "0.3.2",
   "views": {}
  },
  "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": 1
}