训练数据集：{\"
测试数据集：{\"
我们用下面代码查询jupyter的工作目录。" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2020-09-04T06:15:58.114167Z", "start_time": "2020-09-04T06:15:58.105497Z" } }, "outputs": [ { "data": { "text/plain": [ "'C:\\\\Users\\\\Administrator'" ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import os\n", "\n", "os.getcwd()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "训练数据集放在----`Jupyter工作目录+\\MNIST_Data\\train\\`，此时train文件夹内应该包含两个文件，`train-images-idx3-ubyte`和`train-labels-idx1-ubyte`
测试数据集放在----`Jupyter工作目录+\\MNIST_Data\\test\\`，此时test文件夹内应该包含两个文件，`t10k-images-idx3-ubyte`和`t10k-labels-idx1-ubyte`" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### 方法二：\n", "直接执行下面代码，会自动进行训练集的下载与解压，但是整个过程根据网络好坏情况会需要花费几分钟时间。" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "ExecuteTime": { "end_time": "2020-09-04T06:15:58.130999Z", "start_time": "2020-09-04T06:15:58.115177Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "******Downloading the MNIST dataset******\n" ] } ], "source": [ "import urllib.request \n", "from urllib.parse import urlparse\n", "import gzip \n", "import os\n", "\n", "def unzip_file(gzip_path):\n", " \"\"\"\n", " unzip dataset file\n", " \n", " Args:\n", " gzip_path (str): dataset file path\n", " \"\"\"\n", " open_file = open(gzip_path.replace('.gz',''), 'wb')\n", " gz_file = gzip.GzipFile(gzip_path)\n", " open_file.write(gz_file.read())\n", " gz_file.close()\n", " \n", "def download_dataset():\n", " \"\"\"Download the dataset from http://yann.lecun.com/exdb/mnist/.\"\"\"\n", " print(\"******Downloading the MNIST dataset******\")\n", " train_path = \"./MNIST_Data/train/\" \n", " test_path = \"./MNIST_Data/test/\"\n", " train_path_check = os.path.exists(train_path)\n", " test_path_check = os.path.exists(test_path)\n", " if train_path_check == False and test_path_check == False:\n", " os.makedirs(train_path)\n", " os.makedirs(test_path)\n", " train_url = {\"http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\", \"http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\"}\n", " test_url = {\"http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\", \"http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\"}\n", " \n", " for url in train_url:\n", " url_parse = urlparse(url)\n", " # split the file name from url\n", " file_name = os.path.join(train_path,url_parse.path.split('/')[-1])\n", " if not os.path.exists(file_name.replace('.gz', '')):\n", " file = urllib.request.urlretrieve(url, file_name)\n", " unzip_file(file_name)\n", " os.remove(file_name)\n", " \n", " for url in test_url:\n", " url_parse = urlparse(url)\n", " # split the file name from url\n", " file_name = os.path.join(test_path,url_parse.path.split('/')[-1])\n", " if not os.path.exists(file_name.replace('.gz', '')):\n", " file = urllib.request.urlretrieve(url, file_name)\n", " unzip_file(file_name)\n", " os.remove(file_name)\n", "\n", "download_dataset()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "这样就完成了数据集的下载解压缩工作。" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 处理MNIST数据集" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "由于我们后面会采用LeNet这样的卷积神经网络对数据集进行训练，而采用LeNet在训练数据时，对数据格式是有所要求的，所以接下来的工作需要我们先查看数据集内的数据是什么样的，这样才能构造一个针对性的数据转换函数，将数据集数据转换成符合训练要求的数据形式。" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "更多的LeNet网络的介绍不在此赘述，希望详细了解LeNet网络，可以查询
其中分为：
1、输入层；
2、卷积层C1；
3、池化层S2；
4、卷积层C3；
5、池化层S4；
6、全连接F6；
7、全连接；
8、全连接OUTPUT。
结构示意如下图:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### LeNet5结构图" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
![\"LeNet5\"](\"https://www.mindspore.cn/tutorial/zh-CN/master/_images/LeNet_5.jpg\")
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"在构建LeNet5前，我们需要对全连接层以及卷积层进行初始化。\n",
"\n",
"`Normal`：参数初始化方法，MindSpore支持`TruncatedNormal`、`Normal`、`Uniform`等多种参数初始化方法，具体可以参考MindSpore API的`mindspore.common.initializer`模块说明。\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"使用MindSpore定义神经网络需要继承`mindspore.nn.cell.Cell`，`Cell`是所有神经网络（`Conv2d`等）的基类。\n",
"\n",
"神经网络的各层需要预先在`__init__`方法中定义，然后通过定义`construct`方法来完成神经网络的前向构造，按照LeNet5的网络结构，定义网络各层如下："
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-04T06:16:00.336989Z",
"start_time": "2020-09-04T06:16:00.328353Z"
}
},
"outputs": [],
"source": [
"import mindspore.nn as nn\n",
"from mindspore.common.initializer import Normal\n",
"\n",
"class LeNet5(nn.Cell):\n",
" \"\"\"Lenet network structure.\"\"\"\n",
" # define the operator required\n",
" def __init__(self, num_class=10, num_channel=1):\n",
" super(LeNet5, self).__init__()\n",
" self.conv1 = nn.Conv2d(num_channel, 6, 5, pad_mode='valid')\n",
" self.conv2 = nn.Conv2d(6, 16, 5, pad_mode='valid')\n",
" self.fc1 = nn.Dense(16 * 5 * 5, 120, weight_init=Normal(0.02))\n",
" self.fc2 = nn.Dense(120, 84, weight_init=Normal(0.02))\n",
" self.fc3 = nn.Dense(84, num_class, weight_init=Normal(0.02))\n",
" self.relu = nn.ReLU()\n",
" self.max_pool2d = nn.MaxPool2d(kernel_size=2, stride=2)\n",
" self.flatten = nn.Flatten()\n",
"\n",
" # use the preceding operators to construct networks\n",
" def construct(self, x):\n",
" x = self.max_pool2d(self.relu(self.conv1(x)))\n",
" x = self.max_pool2d(self.relu(self.conv2(x)))\n",
" x = self.flatten(x)\n",
" x = self.relu(self.fc1(x))\n",
" x = self.relu(self.fc2(x))\n",
" x = self.fc3(x) \n",
" return x"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"构建完成后，可以使用`print(LeNet5())`将神经网络中的各层参数全部打印出来，也可以使用`LeNet().{layer名称}`打印相应的参数信息。下面例子选择打印第一个卷积层和第一个全连接层的相应参数，使用如下："
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"ExecuteTime": {
"end_time": "2020-09-04T06:16:00.652544Z",
"start_time": "2020-09-04T06:16:00.338003Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"layer conv1: Conv2d