classifier.py
733 Bytes
import torch
import torch.nn as nn
class FinalClassifier(nn.Module):
def __init__(self, input_dim, num_classes, hidden_dim=512, dropout_rate=0.3):
super(FinalClassifier, self).__init__()
# 增加一个隐藏层
self.fc1 = nn.Linear(input_dim, hidden_dim) # 第一层全连接层
self.fc2 = nn.Linear(hidden_dim, num_classes) # 第二层全连接层
self.dropout = nn.Dropout(dropout_rate) # Dropout 防止过拟合
self.relu = nn.ReLU() # 激活函数
def forward(self, x):
x = self.relu(self.fc1(x)) # 第一层全连接 + ReLU 激活
x = self.dropout(x) # Dropout
out = self.fc2(x) # 最终输出层(未应用 softmax)
return out