{"id":330,"date":"2022-07-31T14:34:58","date_gmt":"2022-07-31T06:34:58","guid":{"rendered":"http:\/\/www.gislxz.top\/?p=330"},"modified":"2022-07-31T22:43:13","modified_gmt":"2022-07-31T14:43:13","slug":"%e6%b7%b1%e5%ba%a6%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%ef%bc%883%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%883%ef%bc%89\/","title":{"rendered":"\u6df1\u5ea6\u5b66\u4e60\u7b14\u8bb0\uff083\uff09"},"content":{"rendered":"\n<p>mnist\u6570\u636e\u96c6\u662f\u56fe\u50cf\u5206\u7c7b\u7684\u201chello world\u201d\uff0cpaddle\u8fd9\u4e00\u7ae0\u7684\u6559\u7a0b\u4f7f\u7528\u6781\u7b80\u65b9\u5f0f\u5b8c\u6210\u8bc6\u522b\u6a21\u578b\u7684\u642d\u5efa\uff0c\u8be6\u89c1<a href=\"https:\/\/aistudio.baidu.com\/aistudio\/projectdetail\/1334045\" target=\"_blank\"  rel=\"nofollow\" >\u901a\u8fc7\u6781\u7b80\u65b9\u6848\u6784\u5efa\u624b\u5199\u6570\u5b57\u8bc6\u522b\u6a21\u578b<\/a>\uff0c\u7136\u540e\u6211\u4e5f\u4f1a\u7528pytorch\u540c\u6837\u4ee5\u6781\u7b80\u65b9\u5f0f\u5b8c\u6210\u6a21\u578b\u5730\u642d\u5efa\u4ee5\u8fdb\u884c\u5bf9\u6bd4\u5b66\u4e60\u3002\u6559\u7a0b\u7684\u4e0b\u4e00\u7ae0\u662f\u624b\u5199\u6a21\u578b\uff0c\u5305\u62ec\u6570\u636e\u8bfb\u53d6\uff0c\u7f51\u7edc\u7ed3\u6784\u548c\u8bad\u7ec3\u8fc7\u7a0b\uff0c\u5230\u65f6\u5019\u4e5f\u4f1a\u91cd\u5199pytorch\u7248\u672c\u8fdb\u884c\u5bf9\u6bd4\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>#\u52a0\u8f7d\u98de\u6868\u548c\u76f8\u5173\u7c7b\u5e93\nimport paddle\nimport paddle.fluid as fluid\nfrom paddle.fluid.dygraph.nn import Linear\nimport numpy as np\nimport os\nfrom PIL import Image<\/code><\/pre>\n\n\n\n<p>\u901a\u8fc7paddle\u672c\u8eab\u7684api\u4e0b\u8f7d\u6570\u636e\u96c6<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5982\u679c\uff5e\/.cache\/paddle\/dataset\/mnist\/\u76ee\u5f55\u4e0b\u6ca1\u6709MNIST\u6570\u636e\uff0cAPI\u4f1a\u81ea\u52a8\u5c06MINST\u6570\u636e\u4e0b\u8f7d\u5230\u8be5\u6587\u4ef6\u5939\u4e0b\n# \u8bbe\u7f6e\u6570\u636e\u8bfb\u53d6\u5668\uff0c\u8bfb\u53d6MNIST\u6570\u636e\u8bad\u7ec3\u96c6\ntrainset = paddle.dataset.mnist.train()\n# \u5305\u88c5\u6570\u636e\u8bfb\u53d6\u5668\uff0c\u6bcf\u6b21\u8bfb\u53d6\u7684\u6570\u636e\u6570\u91cf\u8bbe\u7f6e\u4e3abatch_size=8\ntrain_reader = paddle.batch(trainset, batch_size=8)\ntrain_reader()\/\/\u8fd4\u56de\u4e00\u4e2a\u6279\u6b21\u7684\u6570\u636e<\/code><\/pre>\n\n\n\n<p>\u4e4b\u540e\u7528\u8fed\u4ee3\u7684\u65b9\u6cd5\u8bfb\u53d6\u6570\u636e\uff0c\u8fed\u4ee3\u5668\uff0c\u51fd\u6570\u4e2d\u7684yield\u4f1a\u5728\u4e0b\u4e00\u7ae0\u5177\u4f53\u63a2\u8ba8\u4e00\u4e0b\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u4ee5\u8fed\u4ee3\u7684\u5f62\u5f0f\u8bfb\u53d6\u6570\u636e\nfor batch_id, data in enumerate(train_reader()):\n    # \u83b7\u5f97\u56fe\u50cf\u6570\u636e\uff0c\u5e76\u8f6c\u4e3afloat32\u7c7b\u578b\u7684\u6570\u7ec4\n    img_data = np.array(&#91;x&#91;0] for x in data]).astype('float32')\n    # \u83b7\u5f97\u56fe\u50cf\u6807\u7b7e\u6570\u636e\uff0c\u5e76\u8f6c\u4e3afloat32\u7c7b\u578b\u7684\u6570\u7ec4\n    label_data = np.array(&#91;x&#91;1] for x in data]).astype('float32')\n    # \u6253\u5370\u6570\u636e\u5f62\u72b6\n    print(\"\u56fe\u50cf\u6570\u636e\u5f62\u72b6\u548c\u5bf9\u5e94\u6570\u636e\u4e3a:\", img_data.shape, img_data&#91;0])\n    print(\"\u56fe\u50cf\u6807\u7b7e\u5f62\u72b6\u548c\u5bf9\u5e94\u6570\u636e\u4e3a:\", label_data.shape, label_data&#91;0])\n    break\n\nprint(\"\\n\u6253\u5370\u7b2c\u4e00\u4e2abatch\u7684\u7b2c\u4e00\u4e2a\u56fe\u50cf\uff0c\u5bf9\u5e94\u6807\u7b7e\u6570\u5b57\u4e3a{}\".format(label_data&#91;0]))\n# \u663e\u793a\u7b2c\u4e00batch\u7684\u7b2c\u4e00\u4e2a\u56fe\u50cf\nimport matplotlib.pyplot as plt\nimg = np.array(img_data&#91;0]+1)*127.5\nimg = np.reshape(img, &#91;28, 28]).astype(np.uint8)\n\nplt.figure(\"Image\") # \u56fe\u50cf\u7a97\u53e3\u540d\u79f0\nplt.imshow(img)\nplt.axis('on') # \u5173\u6389\u5750\u6807\u8f74\u4e3a off\nplt.title('image') # \u56fe\u50cf\u9898\u76ee\nplt.show()<\/code><\/pre>\n\n\n\n<p># \u83b7\u5f97\u56fe\u50cf\u6570\u636e\uff0c\u5e76\u8f6c\u4e3afloat32\u7c7b\u578b\u7684\u6570\u7ec4     img_data = np.array([x[0] for x in data]).astype('float32')     # \u83b7\u5f97\u56fe\u50cf\u6807\u7b7e\u6570\u636e\uff0c\u5e76\u8f6c\u4e3afloat32\u7c7b\u578b\u7684\u6570\u7ec4     label_data = np.array([x[1] for x in data]).astype('float32')<\/p>\n\n\n\n<p>\u8fd9\u4e24\u53e5\u5199\u7684\u5f88\u5e72\u7ec3\uff0c\u8fed\u4ee3\u5668\u8fd4\u56de\u7684\u662f\u516b\u5f20\u56fe\u7247\u4e00\u7ec4\u7684list\uff0clist\u91cc\u6bcf\u4e00\u4e2a\u5143\u7d20\u53c8\u662f\u4e00\u4e2a\u6807\u7b7e\u548c\u4e00\u4e2a\u5b58\u50a8\u56fe\u50cf\u7684\u4e00\u7ef4list\uff0c\u8fd9\u6837\u5c31\u5206\u522b\u5b58\u50a8\u6210\u4e86\u6807\u7b7e\u7684\u4e00\u7ef4\u6570\u7ec4\uff088\uff0c\uff09\u548c\u56fe\u50cf\u7684\u6570\u7ec4\uff088\uff0c784\uff09\u3002<\/p>\n\n\n\n<p>\u63a5\u4e0b\u6765\u5efa\u7acb\u6a21\u578b\u7c7b\uff0c\u8fd9\u91cc\u5c31\u7528\u6700\u7b80\u5355\u7684\u5355\u5c42\u7ebf\u6027\u7f51\u7edc\uff0c\u53ef\u60f3\u800c\u77e5\u6548\u679c\u4f1a\u5f88\u5dee\uff0c\u4f46\u7f51\u7edc\u7ed3\u6784\u4e0d\u662f\u8fd9\u4e00\u7ae0\u7684\u91cd\u70b9\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5b9a\u4e49mnist\u6570\u636e\u8bc6\u522b\u7f51\u7edc\u7ed3\u6784\uff0c\u540c\u623f\u4ef7\u9884\u6d4b\u7f51\u7edc\nclass MNIST(fluid.dygraph.Layer):\n    def __init__(self):\n        super(MNIST, self).__init__()\n        \n        # \u5b9a\u4e49\u4e00\u5c42\u5168\u8fde\u63a5\u5c42\uff0c\u8f93\u51fa\u7ef4\u5ea6\u662f1\uff0c\u6fc0\u6d3b\u51fd\u6570\u4e3aNone\uff0c\u5373\u4e0d\u4f7f\u7528\u6fc0\u6d3b\u51fd\u6570\n        self.fc = Linear(input_dim=784, output_dim=1, act=None)\n        \n    # \u5b9a\u4e49\u7f51\u7edc\u7ed3\u6784\u7684\u524d\u5411\u8ba1\u7b97\u8fc7\u7a0b\n    def forward(self, inputs):\n        outputs = self.fc(inputs)\n        return outputs<\/code><\/pre>\n\n\n\n<p>\u4e4b\u540e\u8bbe\u7f6e\u52a8\u6001\u56fe\u6a21\u5f0f<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5b9a\u4e49\u98de\u6868\u52a8\u6001\u56fe\u5de5\u4f5c\u73af\u5883\nwith fluid.dygraph.guard():\n    # \u58f0\u660e\u7f51\u7edc\u7ed3\u6784\n    model = MNIST()\n    # \u542f\u52a8\u8bad\u7ec3\u6a21\u5f0f\n    model.train()\n    # \u5b9a\u4e49\u6570\u636e\u8bfb\u53d6\u51fd\u6570\uff0c\u6570\u636e\u8bfb\u53d6batch_size\u8bbe\u7f6e\u4e3a16\n    train_loader = paddle.batch(paddle.dataset.mnist.train(), batch_size=16)\n    # \u5b9a\u4e49\u4f18\u5316\u5668\uff0c\u4f7f\u7528\u968f\u673a\u68af\u5ea6\u4e0b\u964dSGD\u4f18\u5316\u5668\uff0c\u5b66\u4e60\u7387\u8bbe\u7f6e\u4e3a0.001\n    optimizer = fluid.optimizer.SGDOptimizer(learning_rate=0.001, parameter_list=model.parameters())<\/code><\/pre>\n\n\n\n<p>\u63a5\u7740\u5b9a\u4e49\u8bad\u7ec3\u8fc7\u7a0b<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u901a\u8fc7with\u8bed\u53e5\u521b\u5efa\u4e00\u4e2adygraph\u8fd0\u884c\u7684context\n# \u52a8\u6001\u56fe\u4e0b\u7684\u4e00\u4e9b\u64cd\u4f5c\u9700\u8981\u5728guard\u4e0b\u8fdb\u884c\nwith fluid.dygraph.guard():\n    model = MNIST()\n    model.train()\n    train_loader = paddle.batch(paddle.dataset.mnist.train(), batch_size=16)\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\u683c\u5f0f\u9700\u8981\u8f6c\u6362\u6210\u7b26\u5408\u6846\u67b6\u8981\u6c42\n            image_data = np.array(&#91;x&#91;0] for x in data]).astype('float32')\n            label_data = np.array(&#91;x&#91;1] for x in data]).astype('float32').reshape(-1, 1)\n            # \u5c06\u6570\u636e\u8f6c\u4e3a\u98de\u6868\u52a8\u6001\u56fe\u683c\u5f0f\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\u4e861000\u6279\u6b21\u7684\u6570\u636e\uff0c\u6253\u5370\u4e0b\u5f53\u524dLoss\u7684\u60c5\u51b5\n            if batch_id !=0 and batch_id  % 1000 == 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\n    fluid.save_dygraph(model.state_dict(), 'mnist')<\/code><\/pre>\n\n\n\n<p>\u6a21\u578b\u6d4b\u8bd5<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5bfc\u5165\u56fe\u50cf\u8bfb\u53d6\u7b2c\u4e09\u65b9\u5e93\nimport matplotlib.image as mpimg\nimport matplotlib.pyplot as plt\nimport cv2\nimport numpy as np\n# \u8bfb\u53d6\u56fe\u50cf\nimg1 = cv2.imread('.\/work\/example_0.png')\nexample = mpimg.imread('.\/work\/example_0.png')\n# \u663e\u793a\u56fe\u50cf\nplt.imshow(example)\nplt.show()\nim = Image.open('.\/work\/example_0.png').convert('L')\nprint(np.array(im).shape)\nim = im.resize((28, 28), Image.ANTIALIAS)\nplt.imshow(im)\nplt.show()\nprint(np.array(im).shape)\n# \u8bfb\u53d6\u4e00\u5f20\u672c\u5730\u7684\u6837\u4f8b\u56fe\u7247\uff0c\u8f6c\u53d8\u6210\u6a21\u578b\u8f93\u5165\u7684\u683c\u5f0f\ndef load_image(img_path):\n    # \u4eceimg_path\u4e2d\u8bfb\u53d6\u56fe\u50cf\uff0c\u5e76\u8f6c\u4e3a\u7070\u5ea6\u56fe\n    im = Image.open(img_path).convert('L')\n    print(np.array(im))\n    im = im.resize((28, 28), Image.ANTIALIAS)\n    im = np.array(im).reshape(1, -1).astype(np.float32)\n    # \u56fe\u50cf\u5f52\u4e00\u5316\uff0c\u4fdd\u6301\u548c\u6570\u636e\u96c6\u7684\u6570\u636e\u8303\u56f4\u4e00\u81f4\n    im = 1 - im \/ 127.5\n    return im\n\n# \u5b9a\u4e49\u9884\u6d4b\u8fc7\u7a0b\nwith fluid.dygraph.guard():\n    model = MNIST()\n    params_file_path = 'mnist'\n    img_path = '.\/work\/example_0.png'\n# \u52a0\u8f7d\u6a21\u578b\u53c2\u6570\n    model_dict, _ = fluid.load_dygraph(\"mnist\")\n    model.load_dict(model_dict)\n# \u704c\u5165\u6570\u636e\n    model.eval()\n    tensor_img = load_image(img_path)\n    result = model(fluid.dygraph.to_variable(tensor_img))\n#  \u9884\u6d4b\u8f93\u51fa\u53d6\u6574\uff0c\u5373\u4e3a\u9884\u6d4b\u7684\u6570\u5b57\uff0c\u6253\u5370\u7ed3\u679c\n    print(\"\u672c\u6b21\u9884\u6d4b\u7684\u6570\u5b57\u662f\", result.numpy().astype('int32'))<\/code><\/pre>\n\n\n\n<p>\u63a5\u4e0b\u6765\u770b\u770bpytorch\u7684\u7248\u672c\u600e\u4e48\u5199<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>#\u5bfc\u5165\u76f8\u5173\u5e93\nimport torch\nfrom torch import nn\nfrom torch import optim\nimport torch.nn.functional as F\nimport torchvision\n\n#\u5b9a\u4e49\u6570\u636e\u8bfb\u53d6\u51fd\u6570\ndef get_dataloader(train=True):\n    assert isinstance(train,bool),\"train \u5fc5\u987b\u662fbool\u7c7b\u578b\"\n\n    #\u51c6\u5907\u6570\u636e\u96c6\uff0c\u5176\u4e2d0.1307\uff0c0.3081\u4e3aMNIST\u6570\u636e\u7684\u5747\u503c\u548c\u6807\u51c6\u5dee\uff0c\u8fd9\u6837\u64cd\u4f5c\u80fd\u591f\u5bf9\u5176\u8fdb\u884c\u6807\u51c6\u5316\n    #\u56e0\u4e3aMNIST\u53ea\u6709\u4e00\u4e2a\u901a\u9053\uff08\u9ed1\u767d\u56fe\u7247\uff09,\u6240\u4ee5\u5143\u7ec4\u4e2d\u53ea\u6709\u4e00\u4e2a\u503c\n    dataset = torchvision.datasets.MNIST(r\"E:\\NLPDATA\", train=train, download=True,\n                                         transform=torchvision.transforms.Compose(&#91;\n                                         torchvision.transforms.ToTensor(),\n                                         torchvision.transforms.Normalize((0.1307,), (0.3081,)),]))\n    #\u51c6\u5907\u6570\u636e\u8fed\u4ee3\u5668\n    batch_size = train_batch_size if train else test_batch_size\n    dataloader = torch.utils.data.DataLoader(dataset,batch_size=batch_size,shuffle=True)\n    return dataloader\n\n#\u5b9a\u4e49\u7f51\u7edc\u7ed3\u6784\nclass MnistNet(nn.Module):\r\n    def __init__(self):\r\n        super(MnistNet,self).__init__()\r\n        self.fc1 = nn.Linear(28*28*1,1)\r\n\r\n    def forward(self,x):\r\n        x = x.view(-1,28*28*1)\r\n        x = self.fc1(x)\r\n        return x\n\ntrain_batch_size = 8\ntest_batch_size = 1000\nimg_size = 28\nmnist_net = MnistNet()\noptimizer = optim.Adam(mnist_net.parameters(),lr= 0.001)\ncriterion = nn.MSELoss()\ntrain_loss_list = &#91;]\ntrain_count_list = &#91;]\ndef train(epoch):\n    mode = True\n    mnist_net.train(mode=mode)\n    train_dataloader = get_dataloader(train=mode)\n    print(len(train_dataloader.dataset))\n    print(len(train_dataloader))\n    for idx,(data,target) in enumerate(train_dataloader):\n        optimizer.zero_grad()\n        output = mnist_net(data)\n        output=output.to(torch.float32)\n        target=target.to(torch.float32)\n        loss = criterion(output,target)\n        loss.backward()\n        optimizer.step()\n        if idx % 1000 == 0:\n            print('Train Epoch: {} &#91;{}\/{} ({:.0f}%)]\\tLoss: {:.6f}'.format(\n                epoch, idx * len(data), len(train_dataloader.dataset),\n                       100. * idx \/ len(train_dataloader), loss.item()))\n\n            train_loss_list.append(loss.item())\n            train_count_list.append(idx*train_batch_size+(epoch-1)*len(train_dataloader))\n\ndef test():\n    test_loss = 0\n    correct = 0\n    mnist_net.eval()\n    test_dataloader = get_dataloader(train=False)\n    with torch.no_grad():\n        for data, target in test_dataloader:\n            output = mnist_net(data)\n            output=output.to(torch.float32)\n            target=target.to(torch.float32)\n            test_loss += criterion(output, target)\n            pred = output.round()\n            correct += pred.eq(target.data.view_as(pred)).sum()\n    test_loss \/= len(test_dataloader.dataset)\n    print('\\nTest set: Avg. loss: {:.4f}, Accuracy: {}\/{} ({:.2f}%)\\n'.format(\n        test_loss, correct, len(test_dataloader.dataset),\n        100. * correct \/ len(test_dataloader.dataset)))<\/code><\/pre>\n\n\n\n<p>\u540c\u6837\u7684pytorch\u4e5f\u6709api\u5305\u542b\u4e86\u8fd9\u4e9b\u5e38\u7528\u6570\u636e\u96c6\u7684dataloader\uff0c\u4f46\u8fd9\u7ae0\u53ea\u662f\u63cf\u8ff0\u4e86\u4e00\u4e0b\u6781\u7b80\u7684\u6a21\u578b\u642d\u5efa\u65b9\u6cd5\uff0c\u91cd\u70b9\u8fd8\u662f\u5728\u540e\u9762\u7684\u624b\u52a8\u7f16\u5199\u6570\u636e\u8bfb\u53d6\u5904\u7406\u8fc7\u7a0b\uff0c\u624b\u5de5\u642d\u5efa\u6a21\u578b\u7f51\u7edc\uff0c\u624b\u5de5\u7f16\u5199\u8bad\u7ec3\u6d4b\u8bd5\u8fc7\u7a0b\uff0c\u4ee5\u8fbe\u5230\u5b8c\u5168\u6e05\u695a\u53ef\u63a7\u3002\u6bd5\u7adf\u5b9e\u9645\u505a\u9879\u76ee\u65f6\u6570\u636e\u96c6\u90fd\u662f\u81ea\u5df1\u7684\uff0c\u4e5f\u8981\u4ece\u5934\u5f00\u59cb\u5199\u5b8c\u6574\u8fc7\u7a0b\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"310\" height=\"310\" src=\"http:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/113-1.png\" alt=\"\" class=\"wp-image-369\" srcset=\"https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/113-1.png 310w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/113-1-300x300.png 300w, https:\/\/www.gislxz.com\/wp-content\/uploads\/2022\/07\/113-1-150x150.png 150w\" sizes=\"auto, (max-width: 310px) 100vw, 310px\" \/><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>\u901a\u8fc7\u6781\u7b80\u65b9\u6848\u5feb\u901f\u6784\u5efa\u624b\u5199\u6570\u5b57\u8bc6\u522b\u6a21\u578b-paddle\uff0cpytorch\u5b9e\u73b0\u5bf9\u6bd4<\/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-330","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\/330","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=330"}],"version-history":[{"count":3,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts\/330\/revisions"}],"predecessor-version":[{"id":375,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/posts\/330\/revisions\/375"}],"wp:attachment":[{"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/media?parent=330"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/categories?post=330"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.gislxz.com\/index.php\/wp-json\/wp\/v2\/tags?post=330"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}