dlnd_image_classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 图像分类\n",
    "\n",
    "此项目，将对[CIFAR-10 数据集](https://www.cs.toronto.edu/~kriz/cifar.html) 中的图片进行分类。\n",
    "该数据集包含飞机、猫狗和其他物体。导入数据后，需要预处理这些图片，然后用所有样本训练一个卷积神经网络。图片需要标准化（normalized），标签需要采用 one-hot 编码。\n",
    "\n",
    "\n",
    "## 获取数据\n",
    "\n",
    "运行以下单元，以下载 [CIFAR-10 数据集（Python版）](https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz)。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All files found!\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "from urllib.request import urlretrieve\n",
    "from os.path import isfile, isdir\n",
    "from tqdm import tqdm\n",
    "import problem_unittests as tests\n",
    "import tarfile\n",
    "\n",
    "cifar10_dataset_folder_path = 'cifar-10-batches-py'\n",
    "\n",
    "# Use Floyd's cifar-10 dataset if present\n",
    "floyd_cifar10_location = '/input/cifar-10/python.tar.gz'\n",
    "if isfile(floyd_cifar10_location):\n",
    "    tar_gz_path = floyd_cifar10_location\n",
    "else:\n",
    "    tar_gz_path = 'cifar-10-python.tar.gz'\n",
    "\n",
    "class DLProgress(tqdm):\n",
    "    last_block = 0\n",
    "\n",
    "    def hook(self, block_num=1, block_size=1, total_size=None):\n",
    "        self.total = total_size\n",
    "        self.update((block_num - self.last_block) * block_size)\n",
    "        self.last_block = block_num\n",
    "\n",
    "if not isfile(tar_gz_path):\n",
    "    with DLProgress(unit='B', unit_scale=True, miniters=1, desc='CIFAR-10 Dataset') as pbar:\n",
    "        urlretrieve(\n",
    "            'https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz',\n",
    "            tar_gz_path,\n",
    "            pbar.hook)\n",
    "\n",
    "if not isdir(cifar10_dataset_folder_path):\n",
    "    with tarfile.open(tar_gz_path) as tar:\n",
    "        tar.extractall()\n",
    "        tar.close()\n",
    "\n",
    "\n",
    "tests.test_folder_path(cifar10_dataset_folder_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 探索数据\n",
    "\n",
    "该数据集分成了几部分／批次（batches），以免机器在计算时内存不足。CIFAR-10 数据集包含 5 个部分，名称分别为 `data_batch_1`、`data_batch_2`、`data_batch_3`、`data_batch_4`、`data_batch_5`。每个部分都包含以下某个类别的标签和图片：\n",
    "\n",
    "* 飞机\n",
    "* 汽车\n",
    "* 鸟类\n",
    "* 猫\n",
    "* 鹿\n",
    "* 狗\n",
    "* 青蛙\n",
    "* 马\n",
    "* 船只\n",
    "* 卡车\n",
    "\n",
    "了解数据集也是对数据进行预测的必经步骤。可以通过更改 `batch_id` 和 `sample_id` 探索下面的代码单元。`batch_id` 是数据集一个部分的 ID（1 到 5）。`sample_id` 是该部分中图片和标签对（label pair）的 ID。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Stats of batch 1:\n",
      "Samples: 10000\n",
      "Label Counts: {0: 1005, 1: 974, 2: 1032, 3: 1016, 4: 999, 5: 937, 6: 1030, 7: 1001, 8: 1025, 9: 981}\n",
      "First 20 Labels: [6, 9, 9, 4, 1, 1, 2, 7, 8, 3, 4, 7, 7, 2, 9, 9, 9, 3, 2, 6]\n",
      "\n",
      "Example of Image 5:\n",
      "Image - Min Value: 0 Max Value: 252\n",
      "Image - Shape: (32, 32, 3)\n",
      "Label - Label Id: 1 Name: automobile\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfoAAAH0CAYAAADVH+85AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1b8024efa58>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 250,
       "width": 253
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import helper\n",
    "import numpy as np\n",
    "\n",
    "# Explore the dataset\n",
    "batch_id = 1\n",
    "sample_id = 5\n",
    "helper.display_stats(cifar10_dataset_folder_path, batch_id, sample_id)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 实现预处理函数\n",
    "\n",
    "### 标准化\n",
    "\n",
    "在下面的单元中，实现 `normalize` 函数，传入图片数据 `x`，并返回标准化 Numpy 数组。值应该在 0 到 1 的范围内（含 0 和 1）。返回对象应该和 `x` 的形状一样。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def normalize(x):\n",
    "    \"\"\"\n",
    "    Normalize a list of sample image data in the range of 0 to 1\n",
    "    : x: List of image data.  The image shape is (32, 32, 3)\n",
    "    : return: Numpy array of normalize data\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    xmax,xmin = x.max(),x.min()\n",
    "    return (x - xmin)/(xmax - xmin)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_normalize(normalize)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### One-hot 编码\n",
    "\n",
    "实现 `one_hot_encode` 函数， `x`，是一个标签列表。实现该函数，以返回为 one_hot 编码的 Numpy 数组的标签列表。标签的可能值为 0 到 9。每次调用 `one_hot_encode` 时，对于每个值，one_hot 编码函数应该返回相同的编码。确保将编码映射保存到该函数外面。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def one_hot_encode(x):\n",
    "    \"\"\"\n",
    "    One hot encode a list of sample labels. Return a one-hot encoded vector for each label.\n",
    "    : x: List of sample Labels\n",
    "    : return: Numpy array of one-hot encoded labels\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    return np.eye(10)[x]\n",
    "    \n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_one_hot_encode(one_hot_encode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 预处理所有数据并保存\n",
    "\n",
    "运行下方的代码单元，将预处理所有 CIFAR-10 数据，并保存到文件中。下面的代码还使用了 10% 的训练数据，用来验证。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "# Preprocess Training, Validation, and Testing Data\n",
    "helper.preprocess_and_save_data(cifar10_dataset_folder_path, normalize, one_hot_encode)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 检查点\n",
    "\n",
    "这是第一个检查点。如果需要重新启动该记事本，可以从这里开始。预处理的数据已保存到本地。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "import pickle\n",
    "import problem_unittests as tests\n",
    "import helper\n",
    "\n",
    "# Load the Preprocessed Validation data\n",
    "valid_features, valid_labels = pickle.load(open('preprocess_validation.p', mode='rb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 构建网络\n",
    "\n",
    "对于该神经网络，需要将每层都构建为一个函数。\n",
    "\n",
    "\n",
    "### 输入\n",
    "\n",
    "神经网络需要读取图片数据、one-hot 编码标签和丢弃保留概率（dropout keep probability）。需要实现的函数：\n",
    "\n",
    "* 实现 `neural_net_image_input`\n",
    " * 返回 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder)\n",
    " * 使用 `image_shape` 设置形状，部分大小设为 `None`\n",
    " * 使用 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder) 中的 TensorFlow `name` 参数对 TensorFlow 占位符 \"x\" 命名\n",
    "* 实现 `neural_net_label_input`\n",
    " * 返回 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder)\n",
    " * 使用 `n_classes` 设置形状，部分大小设为 `None`\n",
    " * 使用 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder) 中的 TensorFlow `name` 参数对 TensorFlow 占位符 \"y\" 命名\n",
    "* 实现 `neural_net_keep_prob_input`\n",
    " * 返回 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder)，用于丢弃保留概率\n",
    " * 使用 [TF Placeholder](https://www.tensorflow.org/api_docs/python/tf/placeholder) 中的 TensorFlow `name` 参数对 TensorFlow 占位符 \"keep_prob\" 命名\n",
    "\n",
    "这些名称将在项目结束时，用于加载保存的模型。\n",
    "\n",
    "注意：TensorFlow 中的 `None` 表示形状可以是动态大小。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image Input Tests Passed.\n",
      "Label Input Tests Passed.\n",
      "Keep Prob Tests Passed.\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "def neural_net_image_input(image_shape):\n",
    "    \"\"\"\n",
    "    Return a Tensor for a batch of image input\n",
    "    : image_shape: Shape of the images\n",
    "    : return: Tensor for image input.\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    x = tf.placeholder(tf.float32, [None, image_shape[0], image_shape[1], image_shape[2]],name='x')\n",
    "    return x\n",
    "\n",
    "\n",
    "def neural_net_label_input(n_classes):\n",
    "    \"\"\"\n",
    "    Return a Tensor for a batch of label input\n",
    "    : n_classes: Number of classes\n",
    "    : return: Tensor for label input.\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    y = tf.placeholder(tf.float32, [None, n_classes],name='y')\n",
    "    return y\n",
    "\n",
    "\n",
    "def neural_net_keep_prob_input():\n",
    "    \"\"\"\n",
    "    Return a Tensor for keep probability\n",
    "    : return: Tensor for keep probability.\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    keep_prob = tf.placeholder(tf.float32,name='keep_prob')\n",
    "    return keep_prob\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tf.reset_default_graph()\n",
    "tests.test_nn_image_inputs(neural_net_image_input)\n",
    "tests.test_nn_label_inputs(neural_net_label_input)\n",
    "tests.test_nn_keep_prob_inputs(neural_net_keep_prob_input)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 卷积和最大池化层\n",
    "\n",
    "卷积层级适合处理图片。对于此代码单元，应该实现函数 `conv2d_maxpool` 以便应用卷积然后进行最大池化：\n",
    "\n",
    "* 使用 `conv_ksize`、`conv_num_outputs` 和 `x_tensor` 的形状创建权重（weight）和偏置（bias）。\n",
    "* 使用权重和 `conv_strides` 对 `x_tensor` 应用卷积。\n",
    " * 建议使用我们建议的间距（padding），当然也可以使用任何其他间距。\n",
    "* 添加偏置\n",
    "* 向卷积中添加非线性激活（nonlinear activation）\n",
    "* 使用 `pool_ksize` 和 `pool_strides` 应用最大池化\n",
    " \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def conv2d_maxpool(x_tensor, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides):\n",
    "    \"\"\"\n",
    "    Apply convolution then max pooling to x_tensor\n",
    "    :param x_tensor: TensorFlow Tensor\n",
    "    :param conv_num_outputs: Number of outputs for the convolutional layer\n",
    "    :param conv_ksize: kernal size 2-D Tuple for the convolutional layer\n",
    "    :param conv_strides: Stride 2-D Tuple for convolution\n",
    "    :param pool_ksize: kernal size 2-D Tuple for pool\n",
    "    :param pool_strides: Stride 2-D Tuple for pool\n",
    "    : return: A tensor that represents convolution and max pooling of x_tensor\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    weight = tf.Variable(tf.random_normal([conv_ksize[0], conv_ksize[1], x_tensor.get_shape().as_list()[-1], conv_num_outputs], stddev=5e-2))\n",
    "    bias = tf.Variable(tf.zeros(conv_num_outputs))\n",
    "\n",
    "    conv_layer = tf.nn.conv2d(x_tensor, weight, strides=[1, conv_strides[0], conv_strides[1], 1], padding='SAME')\n",
    "    conv_layer = tf.nn.bias_add(conv_layer, bias)\n",
    "    conv_layer = tf.nn.relu(conv_layer)\n",
    "\n",
    "    conv_layer = tf.nn.max_pool(conv_layer, ksize=[1, pool_ksize[0], pool_ksize[1], 1], strides=[1,pool_strides[0],pool_strides[1],1], padding='SAME')\n",
    "    return conv_layer\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_con_pool(conv2d_maxpool)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 扁平化层\n",
    "\n",
    "实现 `flatten` 函数，将 `x_tensor` 的维度从四维张量（4-D tensor）变成二维张量。输出应该是形状（*部分大小（Batch Size）*，*扁平化图片大小（Flattened Image Size）*）。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def flatten(x_tensor):\n",
    "    \"\"\"\n",
    "    Flatten x_tensor to (Batch Size, Flattened Image Size)\n",
    "    : x_tensor: A tensor of size (Batch Size, ...), where ... are the image dimensions.\n",
    "    : return: A tensor of size (Batch Size, Flattened Image Size).\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    w1 = x_tensor.get_shape().as_list()[1]\n",
    "    h1 = x_tensor.get_shape().as_list()[2]\n",
    "    d1 = x_tensor.get_shape().as_list()[3]\n",
    "    x_flat = tf.reshape(x_tensor,[-1, w1 * h1 *d1])\n",
    "    return x_flat\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_flatten(flatten)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 全连接层\n",
    "\n",
    "实现 `fully_conn` 函数，以向 `x_tensor` 应用完全连接的层级，形状为（*部分大小（Batch Size）*，*num_outputs*）。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def fully_conn(x_tensor, num_outputs):\n",
    "    \"\"\"\n",
    "    Apply a fully connected layer to x_tensor using weight and bias\n",
    "    : x_tensor: A 2-D tensor where the first dimension is batch size.\n",
    "    : num_outputs: The number of output that the new tensor should be.\n",
    "    : return: A 2-D tensor where the second dimension is num_outputs.\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    return tf.contrib.layers.fully_connected(x_tensor, num_outputs)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_fully_conn(fully_conn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 输出层\n",
    "\n",
    "实现 `output` 函数，向 x_tensor 应用完全连接的层级，形状为（*部分大小（Batch Size）*，*num_outputs*）。\n",
    "\n",
    "**注意**：该层级不应应用 Activation、softmax 或交叉熵（cross entropy）。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def output(x_tensor, num_outputs):\n",
    "    \"\"\"\n",
    "    Apply a output layer to x_tensor using weight and bias\n",
    "    : x_tensor: A 2-D tensor where the first dimension is batch size.\n",
    "    : num_outputs: The number of output that the new tensor should be.\n",
    "    : return: A 2-D tensor where the second dimension is num_outputs.\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    return tf.contrib.layers.legacy_fully_connected(x_tensor, num_outputs)\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_output(output)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 创建卷积模型\n",
    "\n",
    "实现函数 `conv_net`， 创建卷积神经网络模型。该函数传入一批图片 `x`，并输出对数（logits）。使用上方创建的层创建此模型：\n",
    "\n",
    "* 应用 1、2 或 3 个卷积和最大池化层（Convolution and Max Pool layers）\n",
    "* 应用一个扁平层（Flatten Layer）\n",
    "* 应用 1、2 或 3 个完全连接层（Fully Connected Layers）\n",
    "* 应用一个输出层（Output Layer）\n",
    "* 返回输出\n",
    "* 使用 `keep_prob` 向模型中的一个或多个层应用 [TensorFlow 的 Dropout](https://www.tensorflow.org/api_docs/python/tf/nn/dropout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Neural Network Built!\n"
     ]
    }
   ],
   "source": [
    "def conv_net(x, keep_prob):\n",
    "    \"\"\"\n",
    "    Create a convolutional neural network model\n",
    "    : x: Placeholder tensor that holds image data.\n",
    "    : keep_prob: Placeholder tensor that hold dropout keep probability.\n",
    "    : return: Tensor that represents logits\n",
    "    \"\"\"\n",
    "    # TODO: Apply 1, 2, or 3 Convolution and Max Pool layers\n",
    "    #    Play around with different number of outputs, kernel size and stride\n",
    "    # Function Definition from Above:\n",
    "    #    conv2d_maxpool(x_tensor, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides)\n",
    "    conv_num_outputs = 64\n",
    "    conv_ksize = (3, 3)\n",
    "    conv_strides = (1, 1)\n",
    "    pool_ksize = (2, 2)\n",
    "    pool_strides = (2, 2)\n",
    "    conv1 = conv2d_maxpool(x, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides)\n",
    "\n",
    "    # TODO: Apply a Flatten Layer\n",
    "    # Function Definition from Above:\n",
    "    #   flatten(x_tensor)\n",
    "    fc1 =  flatten(conv1)\n",
    "\n",
    "    # TODO: Apply 1, 2, or 3 Fully Connected Layers\n",
    "    #    Play around with different number of outputs\n",
    "    # Function Definition from Above:\n",
    "    #   fully_conn(x_tensor, num_outputs)\n",
    "    fc2 = fully_conn(fc1, 192)\n",
    "    fc3 = tf.nn.dropout(fc2, keep_prob)\n",
    "\n",
    "    # TODO: Apply an Output Layer\n",
    "    #    Set this to the number of classes\n",
    "    # Function Definition from Above:\n",
    "    #   output(x_tensor, num_outputs)\n",
    "    \n",
    "    out = output(fc3, 10)\n",
    "    # TODO: return output\n",
    "    return out\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "\n",
    "##############################\n",
    "## Build the Neural Network ##\n",
    "##############################\n",
    "\n",
    "# Remove previous weights, bias, inputs, etc..\n",
    "tf.reset_default_graph()\n",
    "\n",
    "# Inputs\n",
    "x = neural_net_image_input((32, 32, 3))\n",
    "y = neural_net_label_input(10)\n",
    "keep_prob = neural_net_keep_prob_input()\n",
    "\n",
    "# Model\n",
    "logits = conv_net(x, keep_prob)\n",
    "\n",
    "# Name logits Tensor, so that is can be loaded from disk after training\n",
    "logits = tf.identity(logits, name='logits')\n",
    "\n",
    "# Loss and Optimizer\n",
    "cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y))\n",
    "optimizer = tf.train.AdamOptimizer().minimize(cost)\n",
    "\n",
    "# Accuracy\n",
    "correct_pred = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1))\n",
    "accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32), name='accuracy')\n",
    "\n",
    "tests.test_conv_net(conv_net)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 训练神经网络\n",
    "\n",
    "### 单次优化\n",
    "\n",
    "实现函数 `train_neural_network` 以进行单次优化（single optimization）。该优化应该使用 `optimizer` 优化 `session`，其中 `feed_dict` 具有以下参数：\n",
    "\n",
    "* `x` 表示图片输入\n",
    "* `y` 表示标签\n",
    "* `keep_prob` 表示丢弃的保留率\n",
    "\n",
    "每个部分都会调用该函数，所以 `tf.global_variables_initializer()` 已经被调用。\n",
    "\n",
    "注意：不需要返回任何内容。该函数只是用来优化神经网络。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tests Passed\n"
     ]
    }
   ],
   "source": [
    "def train_neural_network(session, optimizer, keep_probability, feature_batch, label_batch):\n",
    "    \"\"\"\n",
    "    Optimize the session on a batch of images and labels\n",
    "    : session: Current TensorFlow session\n",
    "    : optimizer: TensorFlow optimizer function\n",
    "    : keep_probability: keep probability\n",
    "    : feature_batch: Batch of Numpy image data\n",
    "    : label_batch: Batch of Numpy label data\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    session.run(optimizer, feed_dict={\n",
    "                x: feature_batch,\n",
    "                y: label_batch,\n",
    "                keep_prob: keep_probability})\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE\n",
    "\"\"\"\n",
    "tests.test_train_nn(train_neural_network)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 显示数据\n",
    "\n",
    "实现函数 `print_stats` 以输出损失和验证准确率。使用全局变量 `valid_features` 和 `valid_labels` 计算验证准确率。使用保留率 `1.0` 计算损失和验证准确率（loss and validation accuracy）。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def print_stats(session, feature_batch, label_batch, cost, accuracy):\n",
    "    \"\"\"\n",
    "    Print information about loss and validation accuracy\n",
    "    : session: Current TensorFlow session\n",
    "    : feature_batch: Batch of Numpy image data\n",
    "    : label_batch: Batch of Numpy label data\n",
    "    : cost: TensorFlow cost function\n",
    "    : accuracy: TensorFlow accuracy function\n",
    "    \"\"\"\n",
    "    # TODO: Implement Function\n",
    "    loss = session.run(cost, feed_dict={\n",
    "                x: feature_batch,\n",
    "                y: label_batch,\n",
    "                keep_prob: 1.})\n",
    "    valid_acc = session.run(accuracy, feed_dict={\n",
    "                x: valid_features,\n",
    "                y: valid_labels,\n",
    "                keep_prob: 1.})\n",
    "\n",
    "    print('loss: {:>10.4f}, Validation Accuracy: {:.6f}'.format(loss,valid_acc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 超参数\n",
    "\n",
    "调试以下超参数：\n",
    "* 设置 `epochs` 表示神经网络停止学习或开始过拟合的迭代次数\n",
    "* 设置 `batch_size`，表示机器内存允许的部分最大体积。大部分人设为以下常见内存大小：\n",
    "\n",
    " * 64\n",
    " * 128\n",
    " * 256\n",
    " * ...\n",
    "* 设置 `keep_probability` 表示使用丢弃时保留节点的概率"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# TODO: Tune Parameters\n",
    "epochs = 50\n",
    "batch_size = 128\n",
    "keep_probability = 0.5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 在单个 CIFAR-10 部分上训练\n",
    "\n",
    "先用单个部分，而不是用所有的 CIFAR-10 批次训练神经网络。这样可以节省时间，并对模型进行迭代，以提高准确率。最终验证准确率达到 50% 或以上之后，在下一部分对所有数据运行模型。\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checking the Training on a Single Batch...\n",
      "Epoch  1, CIFAR-10 Batch 1:  loss:     1.9073, Validation Accuracy: 0.395600\n",
      "Epoch  2, CIFAR-10 Batch 1:  loss:     1.7249, Validation Accuracy: 0.452000\n",
      "Epoch  3, CIFAR-10 Batch 1:  loss:     1.5303, Validation Accuracy: 0.480600\n",
      "Epoch  4, CIFAR-10 Batch 1:  loss:     1.2854, Validation Accuracy: 0.503600\n",
      "Epoch  5, CIFAR-10 Batch 1:  loss:     1.1951, Validation Accuracy: 0.515200\n",
      "Epoch  6, CIFAR-10 Batch 1:  loss:     1.0568, Validation Accuracy: 0.517800\n",
      "Epoch  7, CIFAR-10 Batch 1:  loss:     0.8888, Validation Accuracy: 0.537600\n",
      "Epoch  8, CIFAR-10 Batch 1:  loss:     0.8458, Validation Accuracy: 0.537200\n",
      "Epoch  9, CIFAR-10 Batch 1:  loss:     0.7409, Validation Accuracy: 0.533800\n",
      "Epoch 10, CIFAR-10 Batch 1:  loss:     0.7196, Validation Accuracy: 0.531600\n",
      "Epoch 11, CIFAR-10 Batch 1:  loss:     0.6203, Validation Accuracy: 0.553400\n",
      "Epoch 12, CIFAR-10 Batch 1:  loss:     0.5870, Validation Accuracy: 0.553400\n",
      "Epoch 13, CIFAR-10 Batch 1:  loss:     0.5545, Validation Accuracy: 0.561000\n",
      "Epoch 14, CIFAR-10 Batch 1:  loss:     0.5017, Validation Accuracy: 0.565200\n",
      "Epoch 15, CIFAR-10 Batch 1:  loss:     0.4217, Validation Accuracy: 0.558200\n",
      "Epoch 16, CIFAR-10 Batch 1:  loss:     0.3817, Validation Accuracy: 0.566000\n",
      "Epoch 17, CIFAR-10 Batch 1:  loss:     0.3448, Validation Accuracy: 0.568800\n",
      "Epoch 18, CIFAR-10 Batch 1:  loss:     0.3232, Validation Accuracy: 0.569600\n",
      "Epoch 19, CIFAR-10 Batch 1:  loss:     0.2981, Validation Accuracy: 0.544800\n",
      "Epoch 20, CIFAR-10 Batch 1:  loss:     0.2615, Validation Accuracy: 0.563000\n",
      "Epoch 21, CIFAR-10 Batch 1:  loss:     0.2639, Validation Accuracy: 0.557600\n",
      "Epoch 22, CIFAR-10 Batch 1:  loss:     0.2129, Validation Accuracy: 0.563800\n",
      "Epoch 23, CIFAR-10 Batch 1:  loss:     0.1886, Validation Accuracy: 0.574200\n",
      "Epoch 24, CIFAR-10 Batch 1:  loss:     0.1461, Validation Accuracy: 0.572000\n",
      "Epoch 25, CIFAR-10 Batch 1:  loss:     0.1638, Validation Accuracy: 0.570200\n",
      "Epoch 26, CIFAR-10 Batch 1:  loss:     0.1554, Validation Accuracy: 0.576000\n",
      "Epoch 27, CIFAR-10 Batch 1:  loss:     0.1254, Validation Accuracy: 0.565600\n",
      "Epoch 28, CIFAR-10 Batch 1:  loss:     0.1122, Validation Accuracy: 0.565400\n",
      "Epoch 29, CIFAR-10 Batch 1:  loss:     0.0883, Validation Accuracy: 0.568200\n",
      "Epoch 30, CIFAR-10 Batch 1:  loss:     0.0809, Validation Accuracy: 0.569600\n",
      "Epoch 31, CIFAR-10 Batch 1:  loss:     0.0832, Validation Accuracy: 0.574400\n",
      "Epoch 32, CIFAR-10 Batch 1:  loss:     0.0846, Validation Accuracy: 0.558600\n",
      "Epoch 33, CIFAR-10 Batch 1:  loss:     0.0812, Validation Accuracy: 0.562200\n",
      "Epoch 34, CIFAR-10 Batch 1:  loss:     0.0748, Validation Accuracy: 0.561800\n",
      "Epoch 35, CIFAR-10 Batch 1:  loss:     0.0644, Validation Accuracy: 0.567800\n",
      "Epoch 36, CIFAR-10 Batch 1:  loss:     0.0516, Validation Accuracy: 0.562000\n",
      "Epoch 37, CIFAR-10 Batch 1:  loss:     0.0677, Validation Accuracy: 0.562800\n",
      "Epoch 38, CIFAR-10 Batch 1:  loss:     0.0435, Validation Accuracy: 0.564000\n",
      "Epoch 39, CIFAR-10 Batch 1:  loss:     0.0538, Validation Accuracy: 0.559200\n",
      "Epoch 40, CIFAR-10 Batch 1:  loss:     0.0491, Validation Accuracy: 0.563600\n",
      "Epoch 41, CIFAR-10 Batch 1:  loss:     0.0513, Validation Accuracy: 0.564600\n",
      "Epoch 42, CIFAR-10 Batch 1:  loss:     0.0471, Validation Accuracy: 0.561600\n",
      "Epoch 43, CIFAR-10 Batch 1:  loss:     0.0617, Validation Accuracy: 0.566600\n",
      "Epoch 44, CIFAR-10 Batch 1:  loss:     0.0399, Validation Accuracy: 0.561800\n",
      "Epoch 45, CIFAR-10 Batch 1:  loss:     0.0358, Validation Accuracy: 0.564200\n",
      "Epoch 46, CIFAR-10 Batch 1:  loss:     0.0354, Validation Accuracy: 0.558400\n",
      "Epoch 47, CIFAR-10 Batch 1:  loss:     0.0347, Validation Accuracy: 0.557000\n",
      "Epoch 48, CIFAR-10 Batch 1:  loss:     0.0209, Validation Accuracy: 0.573000\n",
      "Epoch 49, CIFAR-10 Batch 1:  loss:     0.0256, Validation Accuracy: 0.566800\n",
      "Epoch 50, CIFAR-10 Batch 1:  loss:     0.0197, Validation Accuracy: 0.562600\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "print('Checking the Training on a Single Batch...')\n",
    "with tf.Session() as sess:\n",
    "    # Initializing the variables\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    \n",
    "    # Training cycle\n",
    "    for epoch in range(epochs):\n",
    "        batch_i = 1\n",
    "        for batch_features, batch_labels in helper.load_preprocess_training_batch(batch_i, batch_size):\n",
    "            train_neural_network(sess, optimizer, keep_probability, batch_features, batch_labels)\n",
    "        print('Epoch {:>2}, CIFAR-10 Batch {}:  '.format(epoch + 1, batch_i), end='')\n",
    "        print_stats(sess, batch_features, batch_labels, cost, accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 完全训练模型\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "Epoch  1, CIFAR-10 Batch 1:  loss:     2.0538, Validation Accuracy: 0.348800\n",
      "Epoch  1, CIFAR-10 Batch 2:  loss:     1.6861, Validation Accuracy: 0.426200\n",
      "Epoch  1, CIFAR-10 Batch 3:  loss:     1.4966, Validation Accuracy: 0.458600\n",
      "Epoch  1, CIFAR-10 Batch 4:  loss:     1.5010, Validation Accuracy: 0.479800\n",
      "Epoch  1, CIFAR-10 Batch 5:  loss:     1.5355, Validation Accuracy: 0.502400\n",
      "Epoch  2, CIFAR-10 Batch 1:  loss:     1.6391, Validation Accuracy: 0.495400\n",
      "Epoch  2, CIFAR-10 Batch 2:  loss:     1.2858, Validation Accuracy: 0.521200\n",
      "Epoch  2, CIFAR-10 Batch 3:  loss:     1.2167, Validation Accuracy: 0.512000\n",
      "Epoch  2, CIFAR-10 Batch 4:  loss:     1.3681, Validation Accuracy: 0.526800\n",
      "Epoch  2, CIFAR-10 Batch 5:  loss:     1.3652, Validation Accuracy: 0.530000\n",
      "Epoch  3, CIFAR-10 Batch 1:  loss:     1.3644, Validation Accuracy: 0.548800\n",
      "Epoch  3, CIFAR-10 Batch 2:  loss:     1.1626, Validation Accuracy: 0.533600\n",
      "Epoch  3, CIFAR-10 Batch 3:  loss:     1.0495, Validation Accuracy: 0.548600\n",
      "Epoch  3, CIFAR-10 Batch 4:  loss:     1.2273, Validation Accuracy: 0.565600\n",
      "Epoch  3, CIFAR-10 Batch 5:  loss:     1.2422, Validation Accuracy: 0.537000\n",
      "Epoch  4, CIFAR-10 Batch 1:  loss:     1.1809, Validation Accuracy: 0.560200\n",
      "Epoch  4, CIFAR-10 Batch 2:  loss:     1.0332, Validation Accuracy: 0.572800\n",
      "Epoch  4, CIFAR-10 Batch 3:  loss:     0.9133, Validation Accuracy: 0.553200\n",
      "Epoch  4, CIFAR-10 Batch 4:  loss:     1.1253, Validation Accuracy: 0.577400\n",
      "Epoch  4, CIFAR-10 Batch 5:  loss:     1.0730, Validation Accuracy: 0.579200\n",
      "Epoch  5, CIFAR-10 Batch 1:  loss:     1.1679, Validation Accuracy: 0.576400\n",
      "Epoch  5, CIFAR-10 Batch 2:  loss:     0.9422, Validation Accuracy: 0.569800\n",
      "Epoch  5, CIFAR-10 Batch 3:  loss:     0.8542, Validation Accuracy: 0.572000\n",
      "Epoch  5, CIFAR-10 Batch 4:  loss:     0.9886, Validation Accuracy: 0.589400\n",
      "Epoch  5, CIFAR-10 Batch 5:  loss:     0.9743, Validation Accuracy: 0.584000\n",
      "Epoch  6, CIFAR-10 Batch 1:  loss:     1.0241, Validation Accuracy: 0.590400\n",
      "Epoch  6, CIFAR-10 Batch 2:  loss:     0.8794, Validation Accuracy: 0.588000\n",
      "Epoch  6, CIFAR-10 Batch 3:  loss:     0.7603, Validation Accuracy: 0.594600\n",
      "Epoch  6, CIFAR-10 Batch 4:  loss:     0.9298, Validation Accuracy: 0.596600\n",
      "Epoch  6, CIFAR-10 Batch 5:  loss:     0.9312, Validation Accuracy: 0.592200\n",
      "Epoch  7, CIFAR-10 Batch 1:  loss:     0.9722, Validation Accuracy: 0.594000\n",
      "Epoch  7, CIFAR-10 Batch 2:  loss:     0.7979, Validation Accuracy: 0.600400\n",
      "Epoch  7, CIFAR-10 Batch 3:  loss:     0.7425, Validation Accuracy: 0.605600\n",
      "Epoch  7, CIFAR-10 Batch 4:  loss:     0.8449, Validation Accuracy: 0.610800\n",
      "Epoch  7, CIFAR-10 Batch 5:  loss:     0.8481, Validation Accuracy: 0.611400\n",
      "Epoch  8, CIFAR-10 Batch 1:  loss:     0.9369, Validation Accuracy: 0.612200\n",
      "Epoch  8, CIFAR-10 Batch 2:  loss:     0.7735, Validation Accuracy: 0.610400\n",
      "Epoch  8, CIFAR-10 Batch 3:  loss:     0.6896, Validation Accuracy: 0.607800\n",
      "Epoch  8, CIFAR-10 Batch 4:  loss:     0.7247, Validation Accuracy: 0.612800\n",
      "Epoch  8, CIFAR-10 Batch 5:  loss:     0.7529, Validation Accuracy: 0.620400\n",
      "Epoch  9, CIFAR-10 Batch 1:  loss:     0.8154, Validation Accuracy: 0.625600\n",
      "Epoch  9, CIFAR-10 Batch 2:  loss:     0.7363, Validation Accuracy: 0.617600\n",
      "Epoch  9, CIFAR-10 Batch 3:  loss:     0.6879, Validation Accuracy: 0.596000\n",
      "Epoch  9, CIFAR-10 Batch 4:  loss:     0.7410, Validation Accuracy: 0.602800\n",
      "Epoch  9, CIFAR-10 Batch 5:  loss:     0.7141, Validation Accuracy: 0.622200\n",
      "Epoch 10, CIFAR-10 Batch 1:  loss:     0.8292, Validation Accuracy: 0.621400\n",
      "Epoch 10, CIFAR-10 Batch 2:  loss:     0.6549, Validation Accuracy: 0.628800\n",
      "Epoch 10, CIFAR-10 Batch 3:  loss:     0.5829, Validation Accuracy: 0.620600\n",
      "Epoch 10, CIFAR-10 Batch 4:  loss:     0.7098, Validation Accuracy: 0.607000\n",
      "Epoch 10, CIFAR-10 Batch 5:  loss:     0.7267, Validation Accuracy: 0.622400\n",
      "Epoch 11, CIFAR-10 Batch 1:  loss:     0.7545, Validation Accuracy: 0.626600\n",
      "Epoch 11, CIFAR-10 Batch 2:  loss:     0.6103, Validation Accuracy: 0.633600\n",
      "Epoch 11, CIFAR-10 Batch 3:  loss:     0.5614, Validation Accuracy: 0.619800\n",
      "Epoch 11, CIFAR-10 Batch 4:  loss:     0.5897, Validation Accuracy: 0.630800\n",
      "Epoch 11, CIFAR-10 Batch 5:  loss:     0.6363, Validation Accuracy: 0.623400\n",
      "Epoch 12, CIFAR-10 Batch 1:  loss:     0.6884, Validation Accuracy: 0.632200\n",
      "Epoch 12, CIFAR-10 Batch 2:  loss:     0.6170, Validation Accuracy: 0.627200\n",
      "Epoch 12, CIFAR-10 Batch 3:  loss:     0.5364, Validation Accuracy: 0.625400\n",
      "Epoch 12, CIFAR-10 Batch 4:  loss:     0.5357, Validation Accuracy: 0.621800\n",
      "Epoch 12, CIFAR-10 Batch 5:  loss:     0.5943, Validation Accuracy: 0.618800\n",
      "Epoch 13, CIFAR-10 Batch 1:  loss:     0.6429, Validation Accuracy: 0.629800\n",
      "Epoch 13, CIFAR-10 Batch 2:  loss:     0.5363, Validation Accuracy: 0.643800\n",
      "Epoch 13, CIFAR-10 Batch 3:  loss:     0.4698, Validation Accuracy: 0.634200\n",
      "Epoch 13, CIFAR-10 Batch 4:  loss:     0.5176, Validation Accuracy: 0.629000\n",
      "Epoch 13, CIFAR-10 Batch 5:  loss:     0.5593, Validation Accuracy: 0.638200\n",
      "Epoch 14, CIFAR-10 Batch 1:  loss:     0.6246, Validation Accuracy: 0.639800\n",
      "Epoch 14, CIFAR-10 Batch 2:  loss:     0.4753, Validation Accuracy: 0.640800\n",
      "Epoch 14, CIFAR-10 Batch 3:  loss:     0.4435, Validation Accuracy: 0.634600\n",
      "Epoch 14, CIFAR-10 Batch 4:  loss:     0.4810, Validation Accuracy: 0.631200\n",
      "Epoch 14, CIFAR-10 Batch 5:  loss:     0.4918, Validation Accuracy: 0.632600\n",
      "Epoch 15, CIFAR-10 Batch 1:  loss:     0.5871, Validation Accuracy: 0.639400\n",
      "Epoch 15, CIFAR-10 Batch 2:  loss:     0.4569, Validation Accuracy: 0.645000\n",
      "Epoch 15, CIFAR-10 Batch 3:  loss:     0.3989, Validation Accuracy: 0.646400\n",
      "Epoch 15, CIFAR-10 Batch 4:  loss:     0.4709, Validation Accuracy: 0.628600\n",
      "Epoch 15, CIFAR-10 Batch 5:  loss:     0.5277, Validation Accuracy: 0.639800\n",
      "Epoch 16, CIFAR-10 Batch 1:  loss:     0.5195, Validation Accuracy: 0.644200\n",
      "Epoch 16, CIFAR-10 Batch 2:  loss:     0.4452, Validation Accuracy: 0.647400\n",
      "Epoch 16, CIFAR-10 Batch 3:  loss:     0.3931, Validation Accuracy: 0.642200\n",
      "Epoch 16, CIFAR-10 Batch 4:  loss:     0.4523, Validation Accuracy: 0.633000\n",
      "Epoch 16, CIFAR-10 Batch 5:  loss:     0.4539, Validation Accuracy: 0.644400\n",
      "Epoch 17, CIFAR-10 Batch 1:  loss:     0.5045, Validation Accuracy: 0.630000\n",
      "Epoch 17, CIFAR-10 Batch 2:  loss:     0.4361, Validation Accuracy: 0.640400\n",
      "Epoch 17, CIFAR-10 Batch 3:  loss:     0.3767, Validation Accuracy: 0.638200\n",
      "Epoch 17, CIFAR-10 Batch 4:  loss:     0.4023, Validation Accuracy: 0.642200\n",
      "Epoch 17, CIFAR-10 Batch 5:  loss:     0.4313, Validation Accuracy: 0.638600\n",
      "Epoch 18, CIFAR-10 Batch 1:  loss:     0.4567, Validation Accuracy: 0.638600\n",
      "Epoch 18, CIFAR-10 Batch 2:  loss:     0.4133, Validation Accuracy: 0.642400\n",
      "Epoch 18, CIFAR-10 Batch 3:  loss:     0.3414, Validation Accuracy: 0.655200\n",
      "Epoch 18, CIFAR-10 Batch 4:  loss:     0.3895, Validation Accuracy: 0.633600\n",
      "Epoch 18, CIFAR-10 Batch 5:  loss:     0.3701, Validation Accuracy: 0.645800\n",
      "Epoch 19, CIFAR-10 Batch 1:  loss:     0.4446, Validation Accuracy: 0.643000\n",
      "Epoch 19, CIFAR-10 Batch 2:  loss:     0.4075, Validation Accuracy: 0.636200\n",
      "Epoch 19, CIFAR-10 Batch 3:  loss:     0.2996, Validation Accuracy: 0.646400\n",
      "Epoch 19, CIFAR-10 Batch 4:  loss:     0.3690, Validation Accuracy: 0.647400\n",
      "Epoch 19, CIFAR-10 Batch 5:  loss:     0.3504, Validation Accuracy: 0.649000\n",
      "Epoch 20, CIFAR-10 Batch 1:  loss:     0.4024, Validation Accuracy: 0.638200\n",
      "Epoch 20, CIFAR-10 Batch 2:  loss:     0.3702, Validation Accuracy: 0.650600\n",
      "Epoch 20, CIFAR-10 Batch 3:  loss:     0.3070, Validation Accuracy: 0.645400\n",
      "Epoch 20, CIFAR-10 Batch 4:  loss:     0.3622, Validation Accuracy: 0.636800\n",
      "Epoch 20, CIFAR-10 Batch 5:  loss:     0.3484, Validation Accuracy: 0.643800\n",
      "Epoch 21, CIFAR-10 Batch 1:  loss:     0.4266, Validation Accuracy: 0.648000\n",
      "Epoch 21, CIFAR-10 Batch 2:  loss:     0.3103, Validation Accuracy: 0.655000\n",
      "Epoch 21, CIFAR-10 Batch 3:  loss:     0.3029, Validation Accuracy: 0.656200\n",
      "Epoch 21, CIFAR-10 Batch 4:  loss:     0.3529, Validation Accuracy: 0.637200\n",
      "Epoch 21, CIFAR-10 Batch 5:  loss:     0.3240, Validation Accuracy: 0.648200\n",
      "Epoch 22, CIFAR-10 Batch 1:  loss:     0.4027, Validation Accuracy: 0.655400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22, CIFAR-10 Batch 2:  loss:     0.3100, Validation Accuracy: 0.653400\n",
      "Epoch 22, CIFAR-10 Batch 3:  loss:     0.2750, Validation Accuracy: 0.649400\n",
      "Epoch 22, CIFAR-10 Batch 4:  loss:     0.3330, Validation Accuracy: 0.636600\n",
      "Epoch 22, CIFAR-10 Batch 5:  loss:     0.3161, Validation Accuracy: 0.643000\n",
      "Epoch 23, CIFAR-10 Batch 1:  loss:     0.3567, Validation Accuracy: 0.642200\n",
      "Epoch 23, CIFAR-10 Batch 2:  loss:     0.3668, Validation Accuracy: 0.645600\n",
      "Epoch 23, CIFAR-10 Batch 3:  loss:     0.2577, Validation Accuracy: 0.652400\n",
      "Epoch 23, CIFAR-10 Batch 4:  loss:     0.3284, Validation Accuracy: 0.647800\n",
      "Epoch 23, CIFAR-10 Batch 5:  loss:     0.2691, Validation Accuracy: 0.648400\n",
      "Epoch 24, CIFAR-10 Batch 1:  loss:     0.3757, Validation Accuracy: 0.657800\n",
      "Epoch 24, CIFAR-10 Batch 2:  loss:     0.2955, Validation Accuracy: 0.650200\n",
      "Epoch 24, CIFAR-10 Batch 3:  loss:     0.2373, Validation Accuracy: 0.648400\n",
      "Epoch 24, CIFAR-10 Batch 4:  loss:     0.2799, Validation Accuracy: 0.643800\n",
      "Epoch 24, CIFAR-10 Batch 5:  loss:     0.2668, Validation Accuracy: 0.655000\n",
      "Epoch 25, CIFAR-10 Batch 1:  loss:     0.3638, Validation Accuracy: 0.653600\n",
      "Epoch 25, CIFAR-10 Batch 2:  loss:     0.2652, Validation Accuracy: 0.648400\n",
      "Epoch 25, CIFAR-10 Batch 3:  loss:     0.2162, Validation Accuracy: 0.648000\n",
      "Epoch 25, CIFAR-10 Batch 4:  loss:     0.2602, Validation Accuracy: 0.650000\n",
      "Epoch 25, CIFAR-10 Batch 5:  loss:     0.2536, Validation Accuracy: 0.652600\n",
      "Epoch 26, CIFAR-10 Batch 1:  loss:     0.3340, Validation Accuracy: 0.653600\n",
      "Epoch 26, CIFAR-10 Batch 2:  loss:     0.2513, Validation Accuracy: 0.649800\n",
      "Epoch 26, CIFAR-10 Batch 3:  loss:     0.2221, Validation Accuracy: 0.654600\n",
      "Epoch 26, CIFAR-10 Batch 4:  loss:     0.2657, Validation Accuracy: 0.642000\n",
      "Epoch 26, CIFAR-10 Batch 5:  loss:     0.2226, Validation Accuracy: 0.658000\n",
      "Epoch 27, CIFAR-10 Batch 1:  loss:     0.3254, Validation Accuracy: 0.654600\n",
      "Epoch 27, CIFAR-10 Batch 2:  loss:     0.2204, Validation Accuracy: 0.660000\n",
      "Epoch 27, CIFAR-10 Batch 3:  loss:     0.2284, Validation Accuracy: 0.643400\n",
      "Epoch 27, CIFAR-10 Batch 4:  loss:     0.2435, Validation Accuracy: 0.648800\n",
      "Epoch 27, CIFAR-10 Batch 5:  loss:     0.2227, Validation Accuracy: 0.659200\n",
      "Epoch 28, CIFAR-10 Batch 1:  loss:     0.3045, Validation Accuracy: 0.657400\n",
      "Epoch 28, CIFAR-10 Batch 2:  loss:     0.2368, Validation Accuracy: 0.655400\n",
      "Epoch 28, CIFAR-10 Batch 3:  loss:     0.2261, Validation Accuracy: 0.644800\n",
      "Epoch 28, CIFAR-10 Batch 4:  loss:     0.2469, Validation Accuracy: 0.651800\n",
      "Epoch 28, CIFAR-10 Batch 5:  loss:     0.2214, Validation Accuracy: 0.653200\n",
      "Epoch 29, CIFAR-10 Batch 1:  loss:     0.3133, Validation Accuracy: 0.652800\n",
      "Epoch 29, CIFAR-10 Batch 2:  loss:     0.1920, Validation Accuracy: 0.652800\n",
      "Epoch 29, CIFAR-10 Batch 3:  loss:     0.1957, Validation Accuracy: 0.640600\n",
      "Epoch 29, CIFAR-10 Batch 4:  loss:     0.2235, Validation Accuracy: 0.643400\n",
      "Epoch 29, CIFAR-10 Batch 5:  loss:     0.1920, Validation Accuracy: 0.654200\n",
      "Epoch 30, CIFAR-10 Batch 1:  loss:     0.3490, Validation Accuracy: 0.655000\n",
      "Epoch 30, CIFAR-10 Batch 2:  loss:     0.1973, Validation Accuracy: 0.657800\n",
      "Epoch 30, CIFAR-10 Batch 3:  loss:     0.1569, Validation Accuracy: 0.644200\n",
      "Epoch 30, CIFAR-10 Batch 4:  loss:     0.2024, Validation Accuracy: 0.651000\n",
      "Epoch 30, CIFAR-10 Batch 5:  loss:     0.2382, Validation Accuracy: 0.647800\n",
      "Epoch 31, CIFAR-10 Batch 1:  loss:     0.2940, Validation Accuracy: 0.652600\n",
      "Epoch 31, CIFAR-10 Batch 2:  loss:     0.2074, Validation Accuracy: 0.654400\n",
      "Epoch 31, CIFAR-10 Batch 3:  loss:     0.1862, Validation Accuracy: 0.628600\n",
      "Epoch 31, CIFAR-10 Batch 4:  loss:     0.1981, Validation Accuracy: 0.646600\n",
      "Epoch 31, CIFAR-10 Batch 5:  loss:     0.1549, Validation Accuracy: 0.647000\n",
      "Epoch 32, CIFAR-10 Batch 1:  loss:     0.3190, Validation Accuracy: 0.649400\n",
      "Epoch 32, CIFAR-10 Batch 2:  loss:     0.1654, Validation Accuracy: 0.654800\n",
      "Epoch 32, CIFAR-10 Batch 3:  loss:     0.1777, Validation Accuracy: 0.639800\n",
      "Epoch 32, CIFAR-10 Batch 4:  loss:     0.2285, Validation Accuracy: 0.641200\n",
      "Epoch 32, CIFAR-10 Batch 5:  loss:     0.1726, Validation Accuracy: 0.647600\n",
      "Epoch 33, CIFAR-10 Batch 1:  loss:     0.2762, Validation Accuracy: 0.658200\n",
      "Epoch 33, CIFAR-10 Batch 2:  loss:     0.1859, Validation Accuracy: 0.658600\n",
      "Epoch 33, CIFAR-10 Batch 3:  loss:     0.1581, Validation Accuracy: 0.643000\n",
      "Epoch 33, CIFAR-10 Batch 4:  loss:     0.2299, Validation Accuracy: 0.637200\n",
      "Epoch 33, CIFAR-10 Batch 5:  loss:     0.1181, Validation Accuracy: 0.647400\n",
      "Epoch 34, CIFAR-10 Batch 1:  loss:     0.2656, Validation Accuracy: 0.649000\n",
      "Epoch 34, CIFAR-10 Batch 2:  loss:     0.1783, Validation Accuracy: 0.648000\n",
      "Epoch 34, CIFAR-10 Batch 3:  loss:     0.1218, Validation Accuracy: 0.646400\n",
      "Epoch 34, CIFAR-10 Batch 4:  loss:     0.2021, Validation Accuracy: 0.642600\n",
      "Epoch 34, CIFAR-10 Batch 5:  loss:     0.1237, Validation Accuracy: 0.655200\n",
      "Epoch 35, CIFAR-10 Batch 1:  loss:     0.2564, Validation Accuracy: 0.631400\n",
      "Epoch 35, CIFAR-10 Batch 2:  loss:     0.1864, Validation Accuracy: 0.641000\n",
      "Epoch 35, CIFAR-10 Batch 3:  loss:     0.1302, Validation Accuracy: 0.639600\n",
      "Epoch 35, CIFAR-10 Batch 4:  loss:     0.1915, Validation Accuracy: 0.645400\n",
      "Epoch 35, CIFAR-10 Batch 5:  loss:     0.1305, Validation Accuracy: 0.659600\n",
      "Epoch 36, CIFAR-10 Batch 1:  loss:     0.2479, Validation Accuracy: 0.644400\n",
      "Epoch 36, CIFAR-10 Batch 2:  loss:     0.1535, Validation Accuracy: 0.644200\n",
      "Epoch 36, CIFAR-10 Batch 3:  loss:     0.1647, Validation Accuracy: 0.635000\n",
      "Epoch 36, CIFAR-10 Batch 4:  loss:     0.1424, Validation Accuracy: 0.655400\n",
      "Epoch 36, CIFAR-10 Batch 5:  loss:     0.1275, Validation Accuracy: 0.651800\n",
      "Epoch 37, CIFAR-10 Batch 1:  loss:     0.2618, Validation Accuracy: 0.648400\n",
      "Epoch 37, CIFAR-10 Batch 2:  loss:     0.1470, Validation Accuracy: 0.643200\n",
      "Epoch 37, CIFAR-10 Batch 3:  loss:     0.1360, Validation Accuracy: 0.634200\n",
      "Epoch 37, CIFAR-10 Batch 4:  loss:     0.1690, Validation Accuracy: 0.645200\n",
      "Epoch 37, CIFAR-10 Batch 5:  loss:     0.1092, Validation Accuracy: 0.663000\n",
      "Epoch 38, CIFAR-10 Batch 1:  loss:     0.2091, Validation Accuracy: 0.643600\n",
      "Epoch 38, CIFAR-10 Batch 2:  loss:     0.1472, Validation Accuracy: 0.640800\n",
      "Epoch 38, CIFAR-10 Batch 3:  loss:     0.1273, Validation Accuracy: 0.639800\n",
      "Epoch 38, CIFAR-10 Batch 4:  loss:     0.1662, Validation Accuracy: 0.650000\n",
      "Epoch 38, CIFAR-10 Batch 5:  loss:     0.1098, Validation Accuracy: 0.656800\n",
      "Epoch 39, CIFAR-10 Batch 1:  loss:     0.2324, Validation Accuracy: 0.634400\n",
      "Epoch 39, CIFAR-10 Batch 2:  loss:     0.1161, Validation Accuracy: 0.638600\n",
      "Epoch 39, CIFAR-10 Batch 3:  loss:     0.1416, Validation Accuracy: 0.643800\n",
      "Epoch 39, CIFAR-10 Batch 4:  loss:     0.1462, Validation Accuracy: 0.643800\n",
      "Epoch 39, CIFAR-10 Batch 5:  loss:     0.0884, Validation Accuracy: 0.656000\n",
      "Epoch 40, CIFAR-10 Batch 1:  loss:     0.2356, Validation Accuracy: 0.645000\n",
      "Epoch 40, CIFAR-10 Batch 2:  loss:     0.1161, Validation Accuracy: 0.647400\n",
      "Epoch 40, CIFAR-10 Batch 3:  loss:     0.1235, Validation Accuracy: 0.642000\n",
      "Epoch 40, CIFAR-10 Batch 4:  loss:     0.1317, Validation Accuracy: 0.647800\n",
      "Epoch 40, CIFAR-10 Batch 5:  loss:     0.1023, Validation Accuracy: 0.655600\n",
      "Epoch 41, CIFAR-10 Batch 1:  loss:     0.2263, Validation Accuracy: 0.644400\n",
      "Epoch 41, CIFAR-10 Batch 2:  loss:     0.1044, Validation Accuracy: 0.646200\n",
      "Epoch 41, CIFAR-10 Batch 3:  loss:     0.1273, Validation Accuracy: 0.625400\n",
      "Epoch 41, CIFAR-10 Batch 4:  loss:     0.1396, Validation Accuracy: 0.636600\n",
      "Epoch 41, CIFAR-10 Batch 5:  loss:     0.0919, Validation Accuracy: 0.647000\n",
      "Epoch 42, CIFAR-10 Batch 1:  loss:     0.2265, Validation Accuracy: 0.632800\n",
      "Epoch 42, CIFAR-10 Batch 2:  loss:     0.0945, Validation Accuracy: 0.650600\n",
      "Epoch 42, CIFAR-10 Batch 3:  loss:     0.0997, Validation Accuracy: 0.645200\n",
      "Epoch 42, CIFAR-10 Batch 4:  loss:     0.1384, Validation Accuracy: 0.634000\n",
      "Epoch 42, CIFAR-10 Batch 5:  loss:     0.1015, Validation Accuracy: 0.650600\n",
      "Epoch 43, CIFAR-10 Batch 1:  loss:     0.2246, Validation Accuracy: 0.636600\n",
      "Epoch 43, CIFAR-10 Batch 2:  loss:     0.0972, Validation Accuracy: 0.651800\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43, CIFAR-10 Batch 3:  loss:     0.1080, Validation Accuracy: 0.639800\n",
      "Epoch 43, CIFAR-10 Batch 4:  loss:     0.1144, Validation Accuracy: 0.642800\n",
      "Epoch 43, CIFAR-10 Batch 5:  loss:     0.1120, Validation Accuracy: 0.652400\n",
      "Epoch 44, CIFAR-10 Batch 1:  loss:     0.2559, Validation Accuracy: 0.630400\n",
      "Epoch 44, CIFAR-10 Batch 2:  loss:     0.1223, Validation Accuracy: 0.649600\n",
      "Epoch 44, CIFAR-10 Batch 3:  loss:     0.0994, Validation Accuracy: 0.640800\n",
      "Epoch 44, CIFAR-10 Batch 4:  loss:     0.1019, Validation Accuracy: 0.645800\n",
      "Epoch 44, CIFAR-10 Batch 5:  loss:     0.1399, Validation Accuracy: 0.646800\n",
      "Epoch 45, CIFAR-10 Batch 1:  loss:     0.2053, Validation Accuracy: 0.632000\n",
      "Epoch 45, CIFAR-10 Batch 2:  loss:     0.1091, Validation Accuracy: 0.645000\n",
      "Epoch 45, CIFAR-10 Batch 3:  loss:     0.0824, Validation Accuracy: 0.641000\n",
      "Epoch 45, CIFAR-10 Batch 4:  loss:     0.1206, Validation Accuracy: 0.640800\n",
      "Epoch 45, CIFAR-10 Batch 5:  loss:     0.1369, Validation Accuracy: 0.648400\n",
      "Epoch 46, CIFAR-10 Batch 1:  loss:     0.2002, Validation Accuracy: 0.634800\n",
      "Epoch 46, CIFAR-10 Batch 2:  loss:     0.0976, Validation Accuracy: 0.650200\n",
      "Epoch 46, CIFAR-10 Batch 3:  loss:     0.0748, Validation Accuracy: 0.639200\n",
      "Epoch 46, CIFAR-10 Batch 4:  loss:     0.1093, Validation Accuracy: 0.643400\n",
      "Epoch 46, CIFAR-10 Batch 5:  loss:     0.1021, Validation Accuracy: 0.652000\n",
      "Epoch 47, CIFAR-10 Batch 1:  loss:     0.1948, Validation Accuracy: 0.640200\n",
      "Epoch 47, CIFAR-10 Batch 2:  loss:     0.0990, Validation Accuracy: 0.646000\n",
      "Epoch 47, CIFAR-10 Batch 3:  loss:     0.0773, Validation Accuracy: 0.640800\n",
      "Epoch 47, CIFAR-10 Batch 4:  loss:     0.0982, Validation Accuracy: 0.643200\n",
      "Epoch 47, CIFAR-10 Batch 5:  loss:     0.0914, Validation Accuracy: 0.652800\n",
      "Epoch 48, CIFAR-10 Batch 1:  loss:     0.1635, Validation Accuracy: 0.627600\n",
      "Epoch 48, CIFAR-10 Batch 2:  loss:     0.0925, Validation Accuracy: 0.651200\n",
      "Epoch 48, CIFAR-10 Batch 3:  loss:     0.0907, Validation Accuracy: 0.633200\n",
      "Epoch 48, CIFAR-10 Batch 4:  loss:     0.1047, Validation Accuracy: 0.650200\n",
      "Epoch 48, CIFAR-10 Batch 5:  loss:     0.1035, Validation Accuracy: 0.651200\n",
      "Epoch 49, CIFAR-10 Batch 1:  loss:     0.1756, Validation Accuracy: 0.637200\n",
      "Epoch 49, CIFAR-10 Batch 2:  loss:     0.0886, Validation Accuracy: 0.644600\n",
      "Epoch 49, CIFAR-10 Batch 3:  loss:     0.0805, Validation Accuracy: 0.641800\n",
      "Epoch 49, CIFAR-10 Batch 4:  loss:     0.0845, Validation Accuracy: 0.639000\n",
      "Epoch 49, CIFAR-10 Batch 5:  loss:     0.0674, Validation Accuracy: 0.653200\n",
      "Epoch 50, CIFAR-10 Batch 1:  loss:     0.1625, Validation Accuracy: 0.638000\n",
      "Epoch 50, CIFAR-10 Batch 2:  loss:     0.0830, Validation Accuracy: 0.648800\n",
      "Epoch 50, CIFAR-10 Batch 3:  loss:     0.0731, Validation Accuracy: 0.636400\n",
      "Epoch 50, CIFAR-10 Batch 4:  loss:     0.0788, Validation Accuracy: 0.642600\n",
      "Epoch 50, CIFAR-10 Batch 5:  loss:     0.0728, Validation Accuracy: 0.651000\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "save_model_path = './image_classification'\n",
    "\n",
    "print('Training...')\n",
    "with tf.Session() as sess:\n",
    "    # Initializing the variables\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    \n",
    "    # Training cycle\n",
    "    for epoch in range(epochs):\n",
    "        # Loop over all batches\n",
    "        n_batches = 5\n",
    "        for batch_i in range(1, n_batches + 1):\n",
    "            for batch_features, batch_labels in helper.load_preprocess_training_batch(batch_i, batch_size):\n",
    "                train_neural_network(sess, optimizer, keep_probability, batch_features, batch_labels)\n",
    "            print('Epoch {:>2}, CIFAR-10 Batch {}:  '.format(epoch + 1, batch_i), end='')\n",
    "            print_stats(sess, batch_features, batch_labels, cost, accuracy)\n",
    "            \n",
    "    # Save Model\n",
    "    saver = tf.train.Saver()\n",
    "    save_path = saver.save(sess, save_model_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## 测试模型\n",
    "\n",
    "利用测试数据集测试模型。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from ./image_classification\n",
      "Testing Accuracy: 0.6437895569620253\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAscAAAJ/CAYAAACUb342AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1b802461c50>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 319,
       "width": 355
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "DON'T MODIFY ANYTHING IN THIS CELL\n",
    "\"\"\"\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import tensorflow as tf\n",
    "import pickle\n",
    "import helper\n",
    "import random\n",
    "\n",
    "# Set batch size if not already set\n",
    "try:\n",
    "    if batch_size:\n",
    "        pass\n",
    "except NameError:\n",
    "    batch_size = 64\n",
    "\n",
    "save_model_path = './image_classification'\n",
    "n_samples = 4\n",
    "top_n_predictions = 3\n",
    "\n",
    "def test_model():\n",
    "    \"\"\"\n",
    "    Test the saved model against the test dataset\n",
    "    \"\"\"\n",
    "\n",
    "    test_features, test_labels = pickle.load(open('preprocess_test.p', mode='rb'))\n",
    "    loaded_graph = tf.Graph()\n",
    "\n",
    "    with tf.Session(graph=loaded_graph) as sess:\n",
    "        # Load model\n",
    "        loader = tf.train.import_meta_graph(save_model_path + '.meta')\n",
    "        loader.restore(sess, save_model_path)\n",
    "\n",
    "        # Get Tensors from loaded model\n",
    "        loaded_x = loaded_graph.get_tensor_by_name('x:0')\n",
    "        loaded_y = loaded_graph.get_tensor_by_name('y:0')\n",
    "        loaded_keep_prob = loaded_graph.get_tensor_by_name('keep_prob:0')\n",
    "        loaded_logits = loaded_graph.get_tensor_by_name('logits:0')\n",
    "        loaded_acc = loaded_graph.get_tensor_by_name('accuracy:0')\n",
    "        \n",
    "        # Get accuracy in batches for memory limitations\n",
    "        test_batch_acc_total = 0\n",
    "        test_batch_count = 0\n",
    "        \n",
    "        for test_feature_batch, test_label_batch in helper.batch_features_labels(test_features, test_labels, batch_size):\n",
    "            test_batch_acc_total += sess.run(\n",
    "                loaded_acc,\n",
    "                feed_dict={loaded_x: test_feature_batch, loaded_y: test_label_batch, loaded_keep_prob: 1.0})\n",
    "            test_batch_count += 1\n",
    "\n",
    "        print('Testing Accuracy: {}\\n'.format(test_batch_acc_total/test_batch_count))\n",
    "\n",
    "        # Print Random Samples\n",
    "        random_test_features, random_test_labels = tuple(zip(*random.sample(list(zip(test_features, test_labels)), n_samples)))\n",
    "        random_test_predictions = sess.run(\n",
    "            tf.nn.top_k(tf.nn.softmax(loaded_logits), top_n_predictions),\n",
    "            feed_dict={loaded_x: random_test_features, loaded_y: random_test_labels, loaded_keep_prob: 1.0})\n",
    "        helper.display_image_predictions(random_test_features, random_test_labels, random_test_predictions)\n",
    "\n",
    "\n",
    "test_model()"
   ]
  }
 ],
 "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}