如何解决Pytorch LSTM 和交叉熵
我正在从事情感分析,我想将输出分为 4 类。对于损失,我使用了交叉熵。
问题是 PyTorch 交叉熵需要 (batch_size,output) 的输入,这有问题。
我的批量大小为 12,序列大小为 32
public class main {
public static double computeIncome(double salesAmount) {
double num = salesAmount;
double baseSalary = 5000;
double numTotal = baseSalary;
if(num < 5000.01) {
numTotal += num * 0.08;
}
else {
if(num < 10000.01) {
numTotal += ((num - 5000) * 0.10);
numTotal += baseSalary * 0.08;
}
else {
numTotal += ((num - 10000) * 0.12);
numTotal += baseSalary * 0.10;
numTotal += baseSalary * 0.08;
}
}
return numTotal;
}
static void cd(int dataCount,double[] dataBase) {
int currentDataCount = dataCount;
double[] db = dataBase;
Scanner reader = new Scanner(System.in);
double valueFromInput = reader.nextDouble();
if (valueFromInput == -1) {
for (int i = 0; i < db.length; i++) {
System.out.println(db[i] + " ");
}
}
double x = computeIncome(valueFromInput);
db[currentDataCount] = x;
currentDataCount += 1;
cd(currentDataCount,db);
reader.close();
}
public static void main(String[] args) {
int dataCount = 0;
double[] db = new double[20];
System.out.println("Input the sales: (type '-1' to finish )");
cd(dataCount,db);
}
}
版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。