2. # 这里用np.random创建一个随机数组作为输入数据
    3. x = np.random.randn(*[3,1,28,28])
    4. x = x.astype('float32')
    5. with fluid.dygraph.guard():
    6.     # 创建LeNet类的实例，指定模型名称和分类的类别数目
    7.     m = LeNet('LeNet', num_classes=10)
    8.     # 通过调用LeNet从基类继承的sublayers()函数，
    9.     # 查看LeNet中所包含的子层
    10.     print(m.sublayers())
    11.     x = fluid.dygraph.to_variable(x)
    12.     for item in m.sublayers():
    13.         # item是LeNet类中的一个子层
    14.         # 查看经过子层之后的输出数据形状
    15.         try:
    16.             x = item(x)
    17.         except:
    18.             x = fluid.layers.reshape(x, [x.shape[0], -1])
    19.             x = item(x)
    20.         if len(item.parameters())==2:
    21.             # 查看卷积和全连接层的数据和参数的形状，
    22.             # 其中item.parameters()[0]是权重参数w，item.parameters()[1]是偏置参数b
    23.             print(item.full_name(), x.shape, item.parameters()[0].shape, item.parameters()[1].shape)
    24.         else:
    25.             # 池化层没有参数
    26.             print(item.full_name(), x.shape)
    1. # -*- coding: utf-8 -*-
    3. # LeNet 识别手写数字
    5. import os
    6. import random
    8. import paddle.fluid as fluid
    9. import numpy as np
    11. # 定义训练过程
    12.     print('start training ... ')
    13.     model.train()
    14.     epoch_num = 5
    15.     opt = fluid.optimizer.Momentum(learning_rate=0.001, momentum=0.9, parameter_list=model.parameters())
    16.     # 使用Paddle自带的数据读取器
    17.     train_loader = paddle.batch(paddle.dataset.mnist.train(), batch_size=10)
    18.     valid_loader = paddle.batch(paddle.dataset.mnist.test(), batch_size=10)
    19.     for epoch in range(epoch_num):
    20.         for batch_id, data in enumerate(train_loader()):
    21.             # 调整输入数据形状和类型
    22.             x_data = np.array([item[0] for item in data], dtype='float32').reshape(-1, 1, 28, 28)
    23.             y_data = np.array([item[1] for item in data], dtype='int64').reshape(-1, 1)
    24.             # 将numpy.ndarray转化成Tensor
    25.             img = fluid.dygraph.to_variable(x_data)
    26.             label = fluid.dygraph.to_variable(y_data)
    27.             # 计算模型输出
    28.             logits = model(img)
    29.             # 计算损失函数
    30.             loss = fluid.layers.softmax_with_cross_entropy(logits, label)
    31.             avg_loss = fluid.layers.mean(loss)
    32.             if batch_id % 1000 == 0:
    33.                 print("epoch: {}, batch_id: {}, loss is: {}".format(epoch, batch_id, avg_loss.numpy()))
    34.             avg_loss.backward()
    35.             opt.minimize(avg_loss)
    36.             model.clear_gradients()
    39.         accuracies = []
    40.         losses = []
    41.         for batch_id, data in enumerate(valid_loader()):
    42.             # 调整输入数据形状和类型
    43.             x_data = np.array([item[0] for item in data], dtype='float32').reshape(-1, 1, 28, 28)
    44.             y_data = np.array([item[1] for item in data], dtype='int64').reshape(-1, 1)
    45.             # 将numpy.ndarray转化成Tensor
    46.             img = fluid.dygraph.to_variable(x_data)
    47.             label = fluid.dygraph.to_variable(y_data)
    48.             # 计算模型输出
    49.             pred = fluid.layers.softmax(logits)
    50.             # 计算损失函数
    51.             loss = fluid.layers.softmax_with_cross_entropy(logits, label)
    52.             acc = fluid.layers.accuracy(pred, label)
    53.             accuracies.append(acc.numpy())
    54.             losses.append(loss.numpy())
    55.         print("[validation] accuracy/loss: {}/{}".format(np.mean(accuracies), np.mean(losses)))
    56.         model.train()
    58.     # 保存模型参数
    59.     fluid.save_dygraph(model.state_dict(), 'mnist')
    62. if __name__ == '__main__':
    63.     # 创建模型
    64.     with fluid.dygraph.guard():
    65.         model = LeNet("LeNet", num_classes=10)
    66.         #启动训练过程
    67.         train(model)
    1. 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 
    2. Begin to download
    3.  
    4. Download finished
    5. 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 
    6. Begin to download
    7. ........
    8. Download finished
    9. 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 
    10. Begin to download
    11.  
    12. Download finished
    13. 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 
    14. Begin to download
    15. ..
    16. Download finished