Day085_HW.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Work\n",
    "1. 試改變 monitor \"Validation Accuracy\" 並比較結果\n",
    "2. 調整 earlystop 的等待次數至 10, 25 並比較結果"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:528: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:529: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:530: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\dtypes.py:535: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import keras\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "\n",
    "# Disable GPU\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = keras.datasets.cifar10.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "## 資料前處理\n",
    "def preproc_x(x, flatten=True):\n",
    "    x = x / 255.\n",
    "    if flatten:\n",
    "        x = x.reshape((len(x), -1))\n",
    "    return x\n",
    "\n",
    "def preproc_y(y, num_classes=10):\n",
    "    if y.shape[-1] == 1:\n",
    "        y = keras.utils.to_categorical(y, num_classes)\n",
    "    return y    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, y_train = train\n",
    "x_test, y_test = test\n",
    "\n",
    "# Preproc the inputs\n",
    "x_train = preproc_x(x_train)\n",
    "x_test = preproc_x(x_test)\n",
    "\n",
    "# Preprc the outputs\n",
    "y_train = preproc_y(y_train)\n",
    "y_test = preproc_y(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import BatchNormalization\n",
    "\n",
    "\"\"\"\n",
    "建立神經網路，並加入 BN layer\n",
    "\"\"\"\n",
    "def build_mlp(input_shape, output_units=10, num_neurons=[128, 128, 128, 128]):\n",
    "    input_layer = keras.layers.Input(input_shape)\n",
    "    \n",
    "    for i, n_units in enumerate(num_neurons):\n",
    "        if i == 0:\n",
    "            x = keras.layers.Dense(units=n_units, \n",
    "                                   activation=\"relu\", \n",
    "                                   name=\"hidden_layer\"+str(i+1))(input_layer)\n",
    "            x = BatchNormalization()(x)\n",
    "        else:\n",
    "            x = keras.layers.Dense(units=n_units, \n",
    "                                   activation=\"relu\", \n",
    "                                   name=\"hidden_layer\"+str(i+1))(x)\n",
    "            x = BatchNormalization()(x)\n",
    "    \n",
    "    out = keras.layers.Dense(units=output_units, activation=\"softmax\", name=\"output\")(x)\n",
    "    \n",
    "    model = keras.models.Model(inputs=[input_layer], outputs=[out])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "## 超參數設定\n",
    "LEARNING_RATE = 3e-3\n",
    "EPOCHS = 50\n",
    "BATCH_SIZE = 1024\n",
    "MOMENTUM = 0.95"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\users\\henry\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\tensorflow\\python\\framework\\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 3072)              0         \n",
      "_________________________________________________________________\n",
      "hidden_layer1 (Dense)        (None, 128)               393344    \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 128)               512       \n",
      "_________________________________________________________________\n",
      "hidden_layer2 (Dense)        (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 128)               512       \n",
      "_________________________________________________________________\n",
      "hidden_layer3 (Dense)        (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 128)               512       \n",
      "_________________________________________________________________\n",
      "hidden_layer4 (Dense)        (None, 128)               16512     \n",
      "_________________________________________________________________\n",
      "batch_normalization_4 (Batch (None, 128)               512       \n",
      "_________________________________________________________________\n",
      "output (Dense)               (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 446,218\n",
      "Trainable params: 445,194\n",
      "Non-trainable params: 1,024\n",
      "_________________________________________________________________\n",
      "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 50000 samples, validate on 10000 samples\n",
      "Epoch 1/50\n",
      "50000/50000 [==============================] - 4s 85us/step - loss: 2.1123 - acc: 0.2881 - val_loss: 2.2282 - val_acc: 0.2968\n",
      "Epoch 2/50\n",
      "50000/50000 [==============================] - 3s 63us/step - loss: 1.6911 - acc: 0.4034 - val_loss: 1.7728 - val_acc: 0.3828\n",
      "Epoch 3/50\n",
      "50000/50000 [==============================] - 3s 62us/step - loss: 1.5756 - acc: 0.4452 - val_loss: 1.6563 - val_acc: 0.4132\n",
      "Epoch 4/50\n",
      "50000/50000 [==============================] - 3s 58us/step - loss: 1.5048 - acc: 0.4695 - val_loss: 1.6250 - val_acc: 0.4321\n",
      "Epoch 5/50\n",
      "50000/50000 [==============================] - 3s 57us/step - loss: 1.4534 - acc: 0.4873 - val_loss: 1.5792 - val_acc: 0.4458\n",
      "Epoch 6/50\n",
      "50000/50000 [==============================] - 3s 56us/step - loss: 1.4083 - acc: 0.5030 - val_loss: 1.5266 - val_acc: 0.4581\n",
      "Epoch 7/50\n",
      "50000/50000 [==============================] - 3s 56us/step - loss: 1.3738 - acc: 0.5143 - val_loss: 1.5155 - val_acc: 0.4614\n",
      "Epoch 8/50\n",
      "50000/50000 [==============================] - 3s 57us/step - loss: 1.3391 - acc: 0.5254 - val_loss: 1.4799 - val_acc: 0.4734\n",
      "Epoch 9/50\n",
      "50000/50000 [==============================] - 3s 57us/step - loss: 1.3084 - acc: 0.5388 - val_loss: 1.4956 - val_acc: 0.4746\n",
      "Epoch 10/50\n",
      "50000/50000 [==============================] - 4s 87us/step - loss: 1.2802 - acc: 0.5493 - val_loss: 1.4983 - val_acc: 0.4656\n",
      "Epoch 11/50\n",
      "50000/50000 [==============================] - 4s 80us/step - loss: 1.2521 - acc: 0.5576 - val_loss: 1.4705 - val_acc: 0.4796\n",
      "Epoch 12/50\n",
      "50000/50000 [==============================] - 4s 88us/step - loss: 1.2264 - acc: 0.5682 - val_loss: 1.4742 - val_acc: 0.4828\n",
      "Epoch 13/50\n",
      "50000/50000 [==============================] - 4s 82us/step - loss: 1.2014 - acc: 0.5754 - val_loss: 1.4445 - val_acc: 0.4893\n",
      "Epoch 14/50\n",
      "50000/50000 [==============================] - 4s 83us/step - loss: 1.1762 - acc: 0.5853 - val_loss: 1.4838 - val_acc: 0.4757\n",
      "Epoch 15/50\n",
      "50000/50000 [==============================] - 4s 86us/step - loss: 1.1512 - acc: 0.5963 - val_loss: 1.4804 - val_acc: 0.4749\n",
      "Epoch 16/50\n",
      "50000/50000 [==============================] - 4s 84us/step - loss: 1.1285 - acc: 0.6043 - val_loss: 1.4859 - val_acc: 0.4813\n",
      "Epoch 17/50\n",
      "50000/50000 [==============================] - 4s 85us/step - loss: 1.1072 - acc: 0.6119 - val_loss: 1.4884 - val_acc: 0.4800\n",
      "Epoch 18/50\n",
      "50000/50000 [==============================] - 4s 83us/step - loss: 1.0859 - acc: 0.6198 - val_loss: 1.4674 - val_acc: 0.4863\n",
      "Epoch 00018: early stopping\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "# 載入 Callbacks, 並將 monitor 設定為監控 validation loss\n",
    "\"\"\"\n",
    "from keras.callbacks import EarlyStopping\n",
    "\n",
    "earlystop = EarlyStopping(monitor=\"val_loss\", \n",
    "                          patience=5, \n",
    "                          verbose=1\n",
    "                          )\n",
    "model = build_mlp(input_shape=x_train.shape[1:])\n",
    "model.summary()\n",
    "optimizer = keras.optimizers.SGD(lr=LEARNING_RATE, nesterov=True, momentum=MOMENTUM)\n",
    "model.compile(loss=\"categorical_crossentropy\", metrics=[\"accuracy\"], optimizer=optimizer)\n",
    "\n",
    "model.fit(x_train, y_train, \n",
    "          epochs=EPOCHS, \n",
    "          batch_size=BATCH_SIZE, \n",
    "          validation_data=(x_test, y_test), \n",
    "          shuffle=True,\n",
    "          callbacks=[earlystop]\n",
    "         )\n",
    "\n",
    "# Collect results\n",
    "train_loss = model.history.history[\"loss\"]\n",
    "valid_loss = model.history.history[\"val_loss\"]\n",
    "train_acc = model.history.history[\"acc\"]\n",
    "valid_acc = model.history.history[\"val_acc\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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"
    },
    {
     "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",
    "%matplotlib inline\n",
    "\n",
    "plt.plot(range(len(train_loss)), train_loss, label=\"train loss\")\n",
    "plt.plot(range(len(valid_loss)), valid_loss, label=\"valid loss\")\n",
    "plt.legend()\n",
    "plt.title(\"Loss\")\n",
    "plt.show()\n",
    "\n",
    "plt.plot(range(len(train_acc)), train_acc, label=\"train accuracy\")\n",
    "plt.plot(range(len(valid_acc)), valid_acc, label=\"valid accuracy\")\n",
    "plt.legend()\n",
    "plt.title(\"Accuracy\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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
}