在使用torch.nn.functional模块时，需要导入包：
from torch.nn import functional以下是常见激活函数的介绍以及对应的代码示例：
tanh (双曲正切)
import torch
x = torch.tensor([-2.0, -1.0, 0.0, 1.0, 2.0])
y = torch.nn.funcationl.tanh(x)
print(y) # 输出：tensor([-0.9640, -0.7616, 0.0000, 0.7616, 0.9640])sigmoid (S形函数)
y = torch.nn.funcational.sigmoid(x)
print(y) # 输出：tensor([0.1192, 0.2689, 0.5000, 0.7311, 0.8808])SiLU (Sigmoid Linear Unit，也称Swish)
y = torch.nn.funcationl.silu(x)
print(y) # 输出：tensor([-0.2384, -0.2689, 0.0000, 0.7311, 1.7616])GELU (Gaussian Error Linear Unit)
y = torch.nn.functional.gelu(x)
print(y) # 输出：tensor([-0.0454, -0.1588, 0.0000, 0.8413, 1.9546])ReLU (Rectified Linear Unit)
y = torch.nn.funcationl.relu(x)
print(y) # 输出：tensor([0., 0., 0., 1., 2.])ReLU_ (In-place ReLU)
x.relu_() # 注意：会改变x本身
print(x) # x的值被修改为：tensor([0., 0., 0., 1., 2.])Leaky ReLU
x = torch.tensor([-2.0, -1.0, 0.0, 1.0, 2.0])
y = torch.nn.functional.leaky_relu(x, negative_slope=0.01)
print(y) # 输出：tensor([-0.0200, -0.0100, 0.0000, 1.0000, 2.0000])Leaky ReLU_ (In-place Leaky ReLU)
x.leaky_relu_(negative_slope=0.01)
print(x) # x的值被修改Softmax
x = torch.tensor([1.0, 2.0, 3.0])
y = torch.nn.functional.softmax(x, dim=0)
print(y) # 输出：tensor([0.0900, 0.2447, 0.6652])Threshold
x = torch.tensor([-1.0, 0.0, 1.0, 2.0])
y = torch.nn.functional.threshold(x, threshold=0.5, value=0.0)
print(y) # 输出：tensor([0., 0., 0., 2.])Normalize
x = torch.tensor([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
y = torch.nn.functional.normalize(x, p=2, dim=1)
print(y) # 输出：标准化到单位向量
