import paddle import paddle.nn.functional as F from paddle.vision.transforms import ToTensor import numpy as np import matplotlib.pyplot as plt
transform = ToTensor()
cifar10_train = paddle.vision.datasets.Cifar10(mode='train',
transform=transform)
cifar10_test = paddle.vision.datasets.Cifar10(mode='test',
transform=transform)
class MyNet(paddle.nn.Layer):
def __init__(self, num_classes=1):
super(MyNet, self).__init__()
self.conv1 = paddle.nn.Conv2D(in_channels=3, out_channels=32, kernel_size=(2, 2))
self.pool1 = paddle.nn.MaxPool2D(kernel_size=2, stride=2)
#池化：降采样；此处采用最大值池化
self.conv2 = paddle.nn.Conv2D(in_channels=32, out_channels=64, kernel_size=(2, 2))
self.pool2 = paddle.nn.MaxPool2D(kernel_size=2, stride=2)
self.conv3 = paddle.nn.Conv2D(in_channels=64, out_channels=64, kernel_size=(2, 2))
self.pool3 = paddle.nn.MaxPool2D(kernel_size=2, stride=2)
self.conv4 = paddle.nn.Conv2D(in_channels=64, out_channels=128, kernel_size=(2, 2))
self.flatten = paddle.nn.Flatten()
self.linear1 = paddle.nn.Linear(in_features=512, out_features=64)
self.linear2 = paddle.nn.Linear(in_features=64, out_features=num_classes)
def forward(self, x):
x = self.conv1(x)
x = F.relu(x)
x = self.pool1(x)
x = self.conv2(x)
x = F.relu(x)
x = self.pool2(x)
x = self.conv3(x)
x = F.relu(x)
x = self.pool3(x)
x = self.conv4(x)
x = F.relu(x)
x = self.flatten(x)
x = self.linear1(x)
x = F.relu(x)
x = self.linear2(x)
return x
模型预测
epoch_num = 10
batch_size = 32
learning_rate = 0.001
val_acc_history = []
val_loss_history = []
def train(model):
print('start training ... ')
# turn into training mode
model.train()
opt = paddle.optimizer.Adam(learning_rate=learning_rate,
parameters=model.parameters())
train_loader = paddle.io.DataLoader(cifar10_train,
shuffle=True,
batch_size=batch_size)
valid_loader = paddle.io.DataLoader(cifar10_test, batch_size=batch_size)
for epoch in range(epoch_num):
for batch_id, data in enumerate(train_loader()):
x_data = data[0]
y_data = paddle.to_tensor(data[1])
y_data = paddle.unsqueeze(y_data, 1)
logits = model(x_data)
loss = F.cross_entropy(logits, y_data)
if batch_id % 1000 == 0:
print("epoch: {}, batch_id: {}, loss is: {}".format(epoch, batch_id, loss.numpy()))
loss.backward()
opt.step()
opt.clear_grad()
# evaluate model after one epoch
model.eval()
accuracies = []
losses = []
for batch_id, data in enumerate(valid_loader()):
x_data = data[0]
y_data = paddle.to_tensor(data[1])
y_data = paddle.unsqueeze(y_data, 1)
logits = model(x_data)
loss = F.cross_entropy(logits, y_data)
acc = paddle.metric.accuracy(logits, y_data)
accuracies.append(acc.numpy())
losses.append(loss.numpy())
avg_acc, avg_loss = np.mean(accuracies), np.mean(losses)
print("[validation] accuracy/loss: {}/{}".format(avg_acc, avg_loss))
val_acc_history.append(avg_acc)
val_loss_history.append(avg_loss)
model.train()
model = MyNet(num_classes=10)
train(model)
start training ... epoch: 0, batch_id: 0, loss is: [2.2956471] epoch: 0, batch_id: 1000, loss is: [1.1769111] [validation] accuracy/loss: 0.5558106899261475/1.2440916299819946 epoch: 1, batch_id: 0, loss is: [1.3489157] epoch: 1, batch_id: 1000, loss is: [1.2474117] [validation] accuracy/loss: 0.5992411971092224/1.1333197355270386 epoch: 2, batch_id: 0, loss is: [0.9533154] epoch: 2, batch_id: 1000, loss is: [1.0819473] [validation] accuracy/loss: 0.6656349897384644/0.9669198989868164 epoch: 3, batch_id: 0, loss is: [0.5330911] epoch: 3, batch_id: 1000, loss is: [0.64690065] [validation] accuracy/loss: 0.681010365486145/0.9109472632408142 epoch: 4, batch_id: 0, loss is: [0.67969406] epoch: 4, batch_id: 1000, loss is: [0.9760423] [validation] accuracy/loss: 0.6943889856338501/0.8789347410202026 epoch: 5, batch_id: 0, loss is: [0.47081977] epoch: 5, batch_id: 1000, loss is: [0.6279324] [validation] accuracy/loss: 0.7052715420722961/0.8517018556594849 epoch: 6, batch_id: 0, loss is: [0.51920074] epoch: 6, batch_id: 1000, loss is: [0.8535361] [validation] accuracy/loss: 0.7143570184707642/0.835128128528595 epoch: 7, batch_id: 0, loss is: [0.3145673] epoch: 7, batch_id: 1000, loss is: [0.49101165] [validation] accuracy/loss: 0.7194488644599915/0.8388176560401917 epoch: 8, batch_id: 0, loss is: [0.55036885] epoch: 8, batch_id: 1000, loss is: [0.9111887] [validation] accuracy/loss: 0.7224441170692444/0.8234149813652039 epoch: 9, batch_id: 0, loss is: [0.458919] epoch: 9, batch_id: 1000, loss is: [0.65196913] [validation] accuracy/loss: 0.7136581540107727/0.8812852501869202
绘制图像
plt.plot(val_acc_history, label = 'validation accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.ylim([0.5, 0.8])
plt.legend(loc='lower right')
