{"id":373,"date":"2022-07-31T23:24:06","date_gmt":"2022-07-31T15:24:06","guid":{"rendered":"http:\/\/www.gislxz.top\/?p=373"},"modified":"2022-07-31T23:52:53","modified_gmt":"2022-07-31T15:52:53","slug":"%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%ef%bc%884%ef%bc%89","status":"publish","type":"post","link":"https:\/\/www.gislxz.com\/index.php\/2022\/07\/31\/%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%ef%bc%884%ef%bc%89\/","title":{"rendered":"\u6df1\u5ea6\u5b66\u4e60\u7b14\u8bb0\uff084\uff09"},"content":{"rendered":"\n<p>\u8fd9\u4e00\u7ae0\u8ddf\u7740<a href=\"https:\/\/aistudio.baidu.com\/aistudio\/projectdetail\/1334010\" target=\"_blank\"  rel=\"nofollow\" >paddle\u7684\u5b98\u65b9\u6559\u7a0b<\/a>\u6765\u624b\u5199\u6570\u636e\u5904\u7406\u90e8\u5206\u3002<\/p>\n\n\n\n<p>\u9996\u5148\u5bfc\u5165\u76f8\u5173\u5e93<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u52a0\u8f7d\u98de\u6868\u548c\u76f8\u5173\u6570\u636e\u5904\u7406\u7684\u5e93\nimport paddle\nimport paddle.fluid as fluid\nfrom paddle.fluid.dygraph.nn import Linear\nimport numpy as np\nimport os\nimport gzip\nimport json\nimport random<\/code><\/pre>\n\n\n\n<p>\u9996\u5148\u67e5\u770bmnist\u6570\u636e\u7684\u5b58\u50a8\u5f62\u5f0f\uff0c\u8fd9\u91cc\u7ed9\u7684\u6570\u636e\u662fjson\u683c\u5f0f\uff0c\u76f4\u63a5\u7528json\u5e93\u6253\u5f00\u5f97\u5230\u4e00\u4e2alist\uff0c\u7ed3\u6784\u5982\u56fe<\/p>\n\n\n\n<figure class=\"wp-block-image is-resized\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/ai-studio-static-online.cdn.bcebos.com\/7d278024d7ac4d6689fdbe0aa1729181699444730e3941d386a55a1ff8ab4276\" alt=\"\" width=\"670\" height=\"335\"\/><\/figure>\n\n\n\n<pre class=\"wp-block-code\"><code># \u58f0\u660e\u6570\u636e\u96c6\u6587\u4ef6\u4f4d\u7f6e\ndatafile = '.\/work\/mnist.json.gz'\nprint('loading mnist dataset from {} ......'.format(datafile))\n# \u52a0\u8f7djson\u6570\u636e\u6587\u4ef6\ndata = json.load(gzip.open(datafile))\nprint('mnist dataset load done')\n# \u8bfb\u53d6\u5230\u7684\u6570\u636e\u533a\u5206\u8bad\u7ec3\u96c6\uff0c\u9a8c\u8bc1\u96c6\uff0c\u6d4b\u8bd5\u96c6\ntrain_set, val_set, eval_set = data\n\n# \u6570\u636e\u96c6\u76f8\u5173\u53c2\u6570\uff0c\u56fe\u7247\u9ad8\u5ea6IMG_ROWS, \u56fe\u7247\u5bbd\u5ea6IMG_COLS\nIMG_ROWS = 28\nIMG_COLS = 28\n\n# \u6253\u5370\u6570\u636e\u4fe1\u606f\nimgs, labels = train_set&#91;0], train_set&#91;1]\nprint(\"\u8bad\u7ec3\u6570\u636e\u96c6\u6570\u91cf: \", len(imgs))\n\n# \u89c2\u5bdf\u9a8c\u8bc1\u96c6\u6570\u91cf\nimgs, labels = val_set&#91;0], val_set&#91;1]\nprint(\"\u9a8c\u8bc1\u6570\u636e\u96c6\u6570\u91cf: \", len(imgs))\n\n# \u89c2\u5bdf\u6d4b\u8bd5\u96c6\u6570\u91cf\nimgs, labels = val= eval_set&#91;0], eval_set&#91;1]\nprint(\"\u6d4b\u8bd5\u6570\u636e\u96c6\u6570\u91cf: \", len(imgs))<\/code><\/pre>\n\n\n\n<p>\u4e0b\u9762\u6765\u5b9e\u73b0\u8fd4\u56de\u4e00\u6279\u6b21\u7684\u6570\u636e\u7684\u6570\u636e\u751f\u6210\u5668\uff0c\u8fd9\u91cc\u7528\u5230\u4e86\u8fed\u4ee3\u5668\uff0c\u4e5f\u5c31\u662fyield\u3002\n\u8fd9\u4e2a\u535a\u4e3b\u8bb2\u7684\u6bd4\u8f83\u7b80\u5355\u6613\u61c2<a href=\"https:\/\/blog.csdn.net\/mieleizhi0522\/article\/details\/82142856\/\" target=\"_blank\"  rel=\"nofollow\" >https:\/\/blog.csdn.net\/mieleizhi0522\/article\/details\/82142856\/<\/a>\n\u90a3\u4e48\u5c31\u6765\u5199\u6211\u4eec\u81ea\u5df1\u7684\u6570\u636e\u8fed\u4ee3\u5668<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>imgs, labels = train_set&#91;0], train_set&#91;1]\nprint(\"\u8bad\u7ec3\u6570\u636e\u96c6\u6570\u91cf: \", len(imgs))\n# \u83b7\u5f97\u6570\u636e\u96c6\u957f\u5ea6\nimgs_length = len(imgs)\n# \u5b9a\u4e49\u6570\u636e\u96c6\u6bcf\u4e2a\u6570\u636e\u7684\u5e8f\u53f7\uff0c\u6839\u636e\u5e8f\u53f7\u8bfb\u53d6\u6570\u636e\nindex_list = list(range(imgs_length))\n# \u8bfb\u5165\u6570\u636e\u65f6\u7528\u5230\u7684\u6279\u6b21\u5927\u5c0f\nBATCHSIZE = 100\n\n# \u968f\u673a\u6253\u4e71\u8bad\u7ec3\u6570\u636e\u7684\u7d22\u5f15\u5e8f\u53f7\nrandom.shuffle(index_list)\n\n# \u5b9a\u4e49\u6570\u636e\u751f\u6210\u5668\uff0c\u8fd4\u56de\u6279\u6b21\u6570\u636e\ndef data_generator():\n\n    imgs_list = &#91;]\n    labels_list = &#91;]\n    for i in index_list:\n        # \u5c06\u6570\u636e\u5904\u7406\u6210\u671f\u671b\u7684\u683c\u5f0f\uff0c\u6bd4\u5982\u7c7b\u578b\u4e3afloat32\uff0cshape\u4e3a&#91;1, 28, 28]\n        img = np.reshape(imgs&#91;i], &#91;1, IMG_ROWS, IMG_COLS]).astype('float32')\n        label = np.reshape(labels&#91;i], &#91;1]).astype('float32')\n        imgs_list.append(img) \n        labels_list.append(label)\n        if len(imgs_list) == BATCHSIZE:\n            # \u83b7\u5f97\u4e00\u4e2abatchsize\u7684\u6570\u636e\uff0c\u5e76\u8fd4\u56de\n            yield np.array(imgs_list), np.array(labels_list)\n            # \u6e05\u7a7a\u6570\u636e\u8bfb\u53d6\u5217\u8868\n            imgs_list = &#91;]\n            labels_list = &#91;]\n\n    # \u5982\u679c\u5269\u4f59\u6570\u636e\u7684\u6570\u76ee\u5c0f\u4e8eBATCHSIZE\uff0c\n    # \u5219\u5269\u4f59\u6570\u636e\u4e00\u8d77\u6784\u6210\u4e00\u4e2a\u5927\u5c0f\u4e3alen(imgs_list)\u7684mini-batch\n    if len(imgs_list) &gt; 0:\n        yield np.array(imgs_list), np.array(labels_list)\n    return data_generator\n\n# \u58f0\u660e\u6570\u636e\u8bfb\u53d6\u51fd\u6570\uff0c\u4ece\u8bad\u7ec3\u96c6\u4e2d\u8bfb\u53d6\u6570\u636e\ntrain_loader = data_generator\n# \u4ee5\u8fed\u4ee3\u7684\u5f62\u5f0f\u8bfb\u53d6\u6570\u636e\nfor batch_id, data in enumerate(train_loader()):\n    image_data, label_data = data\n# \u58f0\u660e\u6570\u636e\u8bfb\u53d6\u51fd\u6570\uff0c\u4ece\u8bad\u7ec3\u96c6\u4e2d\u8bfb\u53d6\u6570\u636e\ntrain_loader = data_generator\n# \u4ee5\u8fed\u4ee3\u7684\u5f62\u5f0f\u8bfb\u53d6\u6570\u636e\nfor batch_id, data in enumerate(train_loader()):\n    image_data, label_data = data\n    if batch_id == 0:\n        # \u6253\u5370\u6570\u636eshape\u548c\u7c7b\u578b\n        print(\"\u6253\u5370\u7b2c\u4e00\u4e2abatch\u6570\u636e\u7684\u7ef4\u5ea6\uff0c\u4ee5\u53ca\u6570\u636e\u7684\u7c7b\u578b:\")\n        print(\"\u56fe\u50cf\u7ef4\u5ea6: {}, \u6807\u7b7e\u7ef4\u5ea6: {}, \u56fe\u50cf\u6570\u636e\u7c7b\u578b: {}, \u6807\u7b7e\u6570\u636e\u7c7b\u578b: {}\".format(image_data.shape, label_data.shape, type(image_data), type(label_data)))\n    break<\/code><\/pre>\n\n\n\n<p>\u6570\u636e\u91cc\u96be\u514d\u4f1a\u51fa\u73b0\u70b9\u5947\u602a\u7684\u9519\u8bef\uff0c\u6bd4\u5982\u4e4b\u524d\u6211\u5c31\u505a\u8fc7\u4e00\u4e2a\u9879\u76ee\uff0c\u6570\u636e\u96c6\u91cc\u6709\u4e00\u5f20\u56fe\u7247\u5c45\u7136\u662f\u56db\u901a\u9053\u5c31\u4f1a\u83ab\u540d\u5176\u5999\u62a5\u9519\uff0c\u4e8e\u662f\u9700\u8981\u5728\u52a0\u8f7d\u6570\u636e\u65f6\u5c31\u5199\u4e00\u4e9b\u68c0\u9a8c\u51fd\u6570\uff0c\u4e00\u822c\u5c31\u662f\u68c0\u9a8c\u4e00\u4e0b\u56fe\u50cf\u7684shape\u548c\u901a\u9053\u6570\u5bf9\u4e0d\u5bf9\uff0c\u548c\u4fbf\u7b7e\u6570\u91cf\u5339\u4e0d\u5339\u914d\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code> imgs_length = len(imgs)\n\n    assert len(imgs) == len(labels), \\\n          \"length of train_imgs({}) should be the same as train_labels({})\".format(len(imgs), len(label))<\/code><\/pre>\n\n\n\n<p>\u63a5\u7740\u628a\u5b83\u5c01\u88c5\u6210\u51fd\u6570<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>def load_data(mode='train'):\n    datafile = '.\/work\/mnist.json.gz'\n    print('loading mnist dataset from {} ......'.format(datafile))\n    # \u52a0\u8f7djson\u6570\u636e\u6587\u4ef6\n    data = json.load(gzip.open(datafile))\n    print('mnist dataset load done')\n   \n    # \u8bfb\u53d6\u5230\u7684\u6570\u636e\u533a\u5206\u8bad\u7ec3\u96c6\uff0c\u9a8c\u8bc1\u96c6\uff0c\u6d4b\u8bd5\u96c6\n    train_set, val_set, eval_set = data\n    if mode=='train':\n        # \u83b7\u5f97\u8bad\u7ec3\u6570\u636e\u96c6\n        imgs, labels = train_set&#91;0], train_set&#91;1]\n    elif mode=='valid':\n        # \u83b7\u5f97\u9a8c\u8bc1\u6570\u636e\u96c6\n        imgs, labels = val_set&#91;0], val_set&#91;1]\n    elif mode=='eval':\n        # \u83b7\u5f97\u6d4b\u8bd5\u6570\u636e\u96c6\n        imgs, labels = eval_set&#91;0], eval_set&#91;1]\n    else:\n        raise Exception(\"mode can only be one of &#91;'train', 'valid', 'eval']\")\n    print(\"\u8bad\u7ec3\u6570\u636e\u96c6\u6570\u91cf: \", len(imgs))\n    \n    # \u6821\u9a8c\u6570\u636e\n    imgs_length = len(imgs)\n\n    assert len(imgs) == len(labels), \\\n          \"length of train_imgs({}) should be the same as train_labels({})\".format(len(imgs), len(labels))\n    \n    # \u83b7\u5f97\u6570\u636e\u96c6\u957f\u5ea6\n    imgs_length = len(imgs)\n    \n    # \u5b9a\u4e49\u6570\u636e\u96c6\u6bcf\u4e2a\u6570\u636e\u7684\u5e8f\u53f7\uff0c\u6839\u636e\u5e8f\u53f7\u8bfb\u53d6\u6570\u636e\n    index_list = list(range(imgs_length))\n    # \u8bfb\u5165\u6570\u636e\u65f6\u7528\u5230\u7684\u6279\u6b21\u5927\u5c0f\n    BATCHSIZE = 100\n    \n    # \u5b9a\u4e49\u6570\u636e\u751f\u6210\u5668\n    def data_generator():\n        if mode == 'train':\n            # \u8bad\u7ec3\u6a21\u5f0f\u4e0b\u6253\u4e71\u6570\u636e\n            random.shuffle(index_list)\n        imgs_list = &#91;]\n        labels_list = &#91;]\n        for i in index_list:\n            # \u5c06\u6570\u636e\u5904\u7406\u6210\u5e0c\u671b\u7684\u683c\u5f0f\uff0c\u6bd4\u5982\u7c7b\u578b\u4e3afloat32\uff0cshape\u4e3a&#91;1, 28, 28]\n            img = np.reshape(imgs&#91;i], &#91;1, IMG_ROWS, IMG_COLS]).astype('float32')\n            label = np.reshape(labels&#91;i], &#91;1]).astype('float32')\n            imgs_list.append(img) \n            labels_list.append(label)\n            if len(imgs_list) == BATCHSIZE:\n                # \u83b7\u5f97\u4e00\u4e2abatchsize\u7684\u6570\u636e\uff0c\u5e76\u8fd4\u56de\n                yield np.array(imgs_list), np.array(labels_list)\n                # \u6e05\u7a7a\u6570\u636e\u8bfb\u53d6\u5217\u8868\n                imgs_list = &#91;]\n                labels_list = &#91;]\n    \n        # \u5982\u679c\u5269\u4f59\u6570\u636e\u7684\u6570\u76ee\u5c0f\u4e8eBATCHSIZE\uff0c\n        # \u5219\u5269\u4f59\u6570\u636e\u4e00\u8d77\u6784\u6210\u4e00\u4e2a\u5927\u5c0f\u4e3alen(imgs_list)\u7684mini-batch\n        if len(imgs_list) &gt; 0:\n            yield np.array(imgs_list), np.array(labels_list)\n    return data_generator<\/code><\/pre>\n\n\n\n<p>\u4e0b\u9762\u5b9a\u4e49\u4e00\u5c42\u795e\u7ecf\u7f51\u7edc\uff0c\u5229\u7528\u5b9a\u4e49\u597d\u7684\u6570\u636e\u5904\u7406\u51fd\u6570\uff0c\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u7684\u8bad\u7ec3\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u6570\u636e\u5904\u7406\u90e8\u5206\u4e4b\u540e\u7684\u4ee3\u7801\uff0c\u6570\u636e\u8bfb\u53d6\u7684\u90e8\u5206\u8c03\u7528load_data\u51fd\u6570\n# \u5b9a\u4e49\u7f51\u7edc\u7ed3\u6784\uff0c\u540c\u4e0a\u4e00\u8282\u6240\u4f7f\u7528\u7684\u7f51\u7edc\u7ed3\u6784\nclass MNIST(fluid.dygraph.Layer):\n    def __init__(self):\n        super(MNIST, self).__init__()\n        self.fc = Linear(input_dim=784, output_dim=1, act=None)\n\n    def forward(self, inputs):\n        inputs = fluid.layers.reshape(inputs, (-1, 784))\n        outputs = self.fc(inputs)\n        return outputs\n\n# \u8bad\u7ec3\u914d\u7f6e\uff0c\u5e76\u542f\u52a8\u8bad\u7ec3\u8fc7\u7a0b\nwith fluid.dygraph.guard():\n    model = MNIST()\n    model.train()\n    #\u8c03\u7528\u52a0\u8f7d\u6570\u636e\u7684\u51fd\u6570\n    train_loader = load_data('train')\n    optimizer = fluid.optimizer.SGDOptimizer(learning_rate=0.001, parameter_list=model.parameters())\n    EPOCH_NUM = 10\n    for epoch_id in range(EPOCH_NUM):\n        for batch_id, data in enumerate(train_loader()):\n            #\u51c6\u5907\u6570\u636e\uff0c\u53d8\u5f97\u66f4\u52a0\u7b80\u6d01\n            image_data, label_data = data\n            image = fluid.dygraph.to_variable(image_data)\n            label = fluid.dygraph.to_variable(label_data)\n            \n            #\u524d\u5411\u8ba1\u7b97\u7684\u8fc7\u7a0b\n            predict = model(image)\n            \n            #\u8ba1\u7b97\u635f\u5931\uff0c\u53d6\u4e00\u4e2a\u6279\u6b21\u6837\u672c\u635f\u5931\u7684\u5e73\u5747\u503c\n            loss = fluid.layers.square_error_cost(predict, label)\n            avg_loss = fluid.layers.mean(loss)\n            \n            #\u6bcf\u8bad\u7ec3\u4e86200\u6279\u6b21\u7684\u6570\u636e\uff0c\u6253\u5370\u4e0b\u5f53\u524dLoss\u7684\u60c5\u51b5\n            if batch_id % 200 == 0:\n                print(\"epoch: {}, batch: {}, loss is: {}\".format(epoch_id, batch_id, avg_loss.numpy()))\n            \n            #\u540e\u5411\u4f20\u64ad\uff0c\u66f4\u65b0\u53c2\u6570\u7684\u8fc7\u7a0b\n            avg_loss.backward()\n            optimizer.minimize(avg_loss)\n            model.clear_gradients()\n\n    #\u4fdd\u5b58\u6a21\u578b\u53c2\u6570\n    fluid.save_dygraph(model.state_dict(), 'mnist')<\/code><\/pre>\n\n\n\n<p>\u628a\u5b83\u6539\u5199\u6210pytorch\u7248\u672c\u4e5f\u5f88\u7b80\u5355\uff0c\u6bd5\u7adf\u6570\u636e\u8bfb\u53d6\u7684\u51fd\u6570\u4e0e\u7528\u4ec0\u4e48\u6df1\u5ea6\u5b66\u4e60\u6846\u67b6\u6ca1\u6709\u4ec0\u4e48\u5173\u7cfb<\/p>\n\n\n\n<p>\u8fd8\u662f\u8981\u6ce8\u610fpytorch\u7684\u597d\u51e0\u4e2a\u51fd\u6570\u540d\u90fd\u4e0d\u4e00\u6837<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u6570\u636e\u5904\u7406\u90e8\u5206\u4e4b\u540e\u7684\u4ee3\u7801\uff0c\u6570\u636e\u8bfb\u53d6\u7684\u90e8\u5206\u8c03\u7528load_data\u51fd\u6570\n# \u5b9a\u4e49\u7f51\u7edc\u7ed3\u6784\uff0c\u540c\u4e0a\u4e00\u8282\u6240\u4f7f\u7528\u7684\u7f51\u7edc\u7ed3\u6784\nclass Mnist(nn.Module):\n    def __init__(self):\n        super(Mnist,self).__init__()\n        self.fc1 = nn.Linear(28*28*1,1)\n \n    def forward(self,x):\n        x = x.view(-1,28*28*1)\n        x = self.fc1(x)\n        return x\n\n# \u8bad\u7ec3\u914d\u7f6e\uff0c\u5e76\u542f\u52a8\u8bad\u7ec3\u8fc7\u7a0b\nmodel = Mnist()\nmodel.train(mode=True)\n#\u8c03\u7528\u52a0\u8f7d\u6570\u636e\u7684\u51fd\u6570\ntrain_loader = load_data('train')\noptimizer = optim.SGD(model.parameters(),lr= 0.001)\ncriterion = nn.MSELoss()\n\nEPOCH_NUM = 10\nfor epoch_id in range(EPOCH_NUM):\n    for batch_id, data in enumerate(train_loader()):\n        #\u51c6\u5907\u6570\u636e\uff0c\u53d8\u5f97\u66f4\u52a0\u7b80\u6d01\n        image_data, label_data = data\n        image = torch.tensor(image_data)\n        label = torch.tensor(label_data)\n            \n        #\u524d\u5411\u8ba1\u7b97\u7684\u8fc7\u7a0b\n        predict = model(image)\n            \n        #\u8ba1\u7b97\u635f\u5931\uff0c\u53d6\u4e00\u4e2a\u6279\u6b21\u6837\u672c\u635f\u5931\u7684\u5e73\u5747\u503c\n        loss = criterion(predict, label)\n        avg_loss = torch.mean(loss)\n            \n        #\u6bcf\u8bad\u7ec3\u4e86200\u6279\u6b21\u7684\u6570\u636e\uff0c\u6253\u5370\u4e0b\u5f53\u524dLoss\u7684\u60c5\u51b5\n        if batch_id % 200 == 0:\n            print(\"epoch: {}, batch: {}, loss is: {}\".format(epoch_id, batch_id, avg_loss.detach().numpy()))\n            \n        #\u540e\u5411\u4f20\u64ad\uff0c\u66f4\u65b0\u53c2\u6570\u7684\u8fc7\u7a0b\n        avg_loss.backward()\n        optimizer.step()\n        model.zero_grad()<\/code><\/pre>\n\n\n\n<p>\u4e4b\u540e\u6559\u7a0b\u8fd8\u8bf4\u4e86paddle\u600e\u4e48\u5f02\u6b65\u8bfb\u53d6\uff0c\u8fd9\u4e2a\u4e4b\u540e\u518d\u8bf4\u5427<\/p>\n\n\n\n<p>\u8bdd\u8bf4\u4eca\u5929lycoris\u66f4\u65b0\u4e86\uff0c\u771f\u7684\u597d\u751c\u554a<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"576\" src=\"http:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-1024x576.png\" alt=\"\" class=\"wp-image-376\" srcset=\"https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-1024x576.png 1024w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-300x169.png 300w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-768x432.png 768w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-1536x864.png 1536w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172934-2048x1152.png 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"576\" src=\"http:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-1024x576.png\" alt=\"\" class=\"wp-image-377\" srcset=\"https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-1024x576.png 1024w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-300x169.png 300w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-768x432.png 768w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-1536x864.png 1536w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/IMG_20220731_172946-2048x1152.png 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>mnist\u6570\u636e\u96c6\u624b\u5199\u6570\u636e\u5904\u7406<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[21],"tags":[],"class_list":["post-373","post","type-post","status-publish","format-standard","hentry","category-21"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts\/373","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/comments?post=373"}],"version-history":[{"count":3,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts\/373\/revisions"}],"predecessor-version":[{"id":380,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts\/373\/revisions\/380"}],"wp:attachment":[{"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/media?parent=373"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/categories?post=373"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/tags?post=373"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}