使用LeNet在MNIST数据集实现图像分类

作者: PaddlePaddle
日期: 2021.05
摘要: 本示例教程演示如何在MNIST数据集上用LeNet进行图像分类。

一、环境配置

本教程基于Paddle 2.1 编写，如果您的环境不是本版本，请先参考官网安装 Paddle 2.1 。

  1. import paddle
  2. print(paddle.__version__)
  1. 2.1.0

二、数据加载

手写数字的MNIST数据集，包含60,000个用于训练的示例和10,000个用于测试的示例。这些数字已经过尺寸标准化并位于图像中心，图像是固定大小(28x28像素)，其值为0到1。该数据集的官方地址为：http://yann.lecun.com/exdb/mnist 。

我们使用飞桨框架自带的 paddle.vision.datasets.MNIST 完成mnist数据集的加载。

  1. from paddle.vision.transforms import Compose, Normalize
  3. transform = Compose([Normalize(mean=[127.5],
  4.                                std=[127.5],
  5.                                data_format='CHW')])
  6. # 使用transform对数据集做归一化
  7. print('download training data and load training data')
  8. train_dataset = paddle.vision.datasets.MNIST(mode='train', transform=transform)
  9. test_dataset = paddle.vision.datasets.MNIST(mode='test', transform=transform)
  10. print('load finished')
  1. download training data and load training data
  4. Cache file /home/aistudio/.cache/paddle/dataset/mnist/train-images-idx3-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/train-images-idx3-ubyte.gz 
  5. Begin to download
  7. Download finished
  8. Cache file /home/aistudio/.cache/paddle/dataset/mnist/train-labels-idx1-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/train-labels-idx1-ubyte.gz 
  9. Begin to download
  10. ........
  11. Download finished
  12. Cache file /home/aistudio/.cache/paddle/dataset/mnist/t10k-images-idx3-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/t10k-images-idx3-ubyte.gz 
  13. Begin to download
  15. Download finished
  16. Cache file /home/aistudio/.cache/paddle/dataset/mnist/t10k-labels-idx1-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/t10k-labels-idx1-ubyte.gz 
  17. Begin to download
  18. ..
  19. Download finished
  22. load finished

取训练集中的一条数据看一下。

  1. import numpy as np
  2. import matplotlib.pyplot as plt
  3. train_data0, train_label_0 = train_dataset[0][0],train_dataset[0][1]
  4. train_data0 = train_data0.reshape([28,28])
  5. plt.figure(figsize=(2,2))
  6. plt.imshow(train_data0, cmap=plt.cm.binary)
  7. print('train_data0 label is: ' + str(train_label_0))
  1. train_data0 label is: [5]

png

三、组网

用paddle.nn下的API，如Conv2D、MaxPool2D、Linear完成LeNet的构建。

  1. import paddle
  2. import paddle.nn.functional as F
  3. class LeNet(paddle.nn.Layer):
  4.     def __init__(self):
  5.         super(LeNet, self).__init__()
  6.         self.conv1 = paddle.nn.Conv2D(in_channels=1, out_channels=6, kernel_size=5, stride=1, padding=2)
  7.         self.max_pool1 = paddle.nn.MaxPool2D(kernel_size=2,  stride=2)
  8.         self.conv2 = paddle.nn.Conv2D(in_channels=6, out_channels=16, kernel_size=5, stride=1)
  9.         self.max_pool2 = paddle.nn.MaxPool2D(kernel_size=2, stride=2)
  10.         self.linear1 = paddle.nn.Linear(in_features=16*5*5, out_features=120)
  11.         self.linear2 = paddle.nn.Linear(in_features=120, out_features=84)
  12.         self.linear3 = paddle.nn.Linear(in_features=84, out_features=10)
  14.     def forward(self, x):
  15.         x = self.conv1(x)
  16.         x = F.relu(x)
  17.         x = self.max_pool1(x)
  18.         x = F.relu(x)
  19.         x = self.conv2(x)
  20.         x = self.max_pool2(x)
  21.         x = paddle.flatten(x, start_axis=1,stop_axis=-1)
  22.         x = self.linear1(x)
  23.         x = F.relu(x)
  24.         x = self.linear2(x)
  25.         x = F.relu(x)
  26.         x = self.linear3(x)
  27.         return x

四、方式1：基于高层API，完成模型的训练与预测

通过paddle提供的Model 构建实例，使用封装好的训练与测试接口，快速完成模型训练与测试。

4.1 使用 Model.fit来训练模型

  1. from paddle.metric import Accuracy
  2. model = paddle.Model(LeNet())   # 用Model封装模型
  3. optim = paddle.optimizer.Adam(learning_rate=0.001, parameters=model.parameters())
  5. # 配置模型
  6. model.prepare(
  7.     optim,
  8.     paddle.nn.CrossEntropyLoss(),
  9.     Accuracy()
  10.     )
  1. # 训练模型
  2. model.fit(train_dataset,
  3.         epochs=2,
  4.         batch_size=64,
  5.         verbose=1
  6.         )
  1. The loss value printed in the log is the current step, and the metric is the average value of previous steps.
  2. Epoch 1/2
  3. step 938/938 [==============================] - loss: 0.0025 - acc: 0.9564 - 9ms/step        
  4. Epoch 2/2
  5. step 938/938 [==============================] - loss: 0.0127 - acc: 0.9844 - 9ms/step

4.2 使用 Model.evaluate 来预测模型

  1. model.evaluate(test_dataset, batch_size=64, verbose=1)
  1. Eval begin...
  2. step 157/157 [==============================] - loss: 1.2412e-04 - acc: 0.9872 - 8ms/step     
  3. Eval samples: 10000
  5. {'loss': [0.0001241174], 'acc': 0.9872}

方式一结束

以上就是方式一，可以快速、高效的完成网络模型训练与预测。

五、方式2：基于基础API，完成模型的训练与预测

5.1 模型训练

组网后，开始对模型进行训练，先构建train_loader，加载训练数据，然后定义train函数，设置好损失函数后，按batch加载数据，完成模型的训练。

  1. import paddle.nn.functional as F
  2. train_loader = paddle.io.DataLoader(train_dataset, batch_size=64, shuffle=True)
  3. # 加载训练集 batch_size 设为 64
  4. def train(model):
  5.     model.train()
  6.     epochs = 2
  7.     optim = paddle.optimizer.Adam(learning_rate=0.001, parameters=model.parameters())
  8.     # 用Adam作为优化函数
  9.     for epoch in range(epochs):
  10.         for batch_id, data in enumerate(train_loader()):
  11.             x_data = data[0]
  12.             y_data = data[1]
  13.             predicts = model(x_data)
  14.             loss = F.cross_entropy(predicts, y_data)
  15.             # 计算损失
  16.             acc = paddle.metric.accuracy(predicts, y_data)
  17.             loss.backward()
  18.             if batch_id % 300 == 0:
  19.                 print("epoch: {}, batch_id: {}, loss is: {}, acc is: {}".format(epoch, batch_id, loss.numpy(), acc.numpy()))
  20.             optim.step()
  21.             optim.clear_grad()
  22. model = LeNet()
  23. train(model)
  1. epoch: 0, batch_id: 0, loss is: [3.0527446], acc is: [0.09375]
  2. epoch: 0, batch_id: 300, loss is: [0.05049332], acc is: [1.]
  3. epoch: 0, batch_id: 600, loss is: [0.109704], acc is: [0.953125]
  4. ...

5.2 模型验证

训练完成后，需要验证模型的效果，此时，加载测试数据集，然后用训练好的模对测试集进行预测，计算损失与精度。

  1. test_loader = paddle.io.DataLoader(test_dataset, places=paddle.CPUPlace(), batch_size=64)
  2. # 加载测试数据集
  3. def test(model):
  4.     model.eval()
  5.     batch_size = 64
  6.     for batch_id, data in enumerate(test_loader()):
  7.         x_data = data[0]
  8.         y_data = data[1]
  9.         predicts = model(x_data)
  10.         # 获取预测结果
  11.         loss = F.cross_entropy(predicts, y_data)
  12.         acc = paddle.metric.accuracy(predicts, y_data)
  13.         if batch_id % 20 == 0:
  14.             print("batch_id: {}, loss is: {}, acc is: {}".format(batch_id, loss.numpy(), acc.numpy()))
  15. test(model)
  1. batch_id: 0, loss is: [0.01972857], acc is: [0.984375]
  2. batch_id: 20, loss is: [0.19958115], acc is: [0.9375]
  3. batch_id: 40, loss is: [0.23575728], acc is: [0.953125]
  4. batch_id: 60, loss is: [0.07018849], acc is: [0.984375]
  5. batch_id: 80, loss is: [0.02309197], acc is: [0.984375]
  6. batch_id: 100, loss is: [0.00239462], acc is: [1.]
  7. batch_id: 120, loss is: [0.01583934], acc is: [1.]
  8. batch_id: 140, loss is: [0.00399609], acc is: [1.]

方式二结束

以上就是方式二，通过底层API，可以清楚的看到训练和测试中的每一步过程。但是，这种方式比较复杂。因此，我们提供了训练方式一，使用高层API来完成模型的训练与预测。对比底层API，高层API能够更加快速、高效的完成模型的训练与测试。

六、总结

以上就是用LeNet对手写数字数据及MNIST进行分类。本示例提供了两种训练模型的方式，一种可以快速完成模型的组建与预测，非常适合新手用户上手。另一种则需要多个步骤来完成模型的训练，适合进阶用户使用。