隐马尔可夫模型(Hidden Markov Model,HMM)的裸题目,在HMM中,状态序列不可知,可知的是观测序列,状态的概率转移矩阵以及观测值在不同状态下的概率分布是可知的,HMM中常见的问题是估计概率和参数,主有有三个:
1.已知模型参数,计算某一特定输出序列的概率.通常使用forward算法解决.
2.已知模型参数,寻找最可能的能产生某一特定输出序列的隐含状态的序列.通常使用Viterbi算法解决.
3.已知输出序列,寻找最可能的状态转移以及输出概率.通常使用Baum-Welch算法以及Reversed Viterbi算法解决.
问题1,2可用动态规划解决,问题3??
本题是问题2的裸题
#include <algorithm>
#include <cstdio>
#include <cstring>
#include <iostream>
#include <map>
#include <set>
#include <string>
#include <cmath>
#include <vector>
#include <queue>
using namespace std;
#define N 111
#define LL long long
#define mod 1000000007ll
#define inf 0XFFFFFFF
#define pii pair<int, int>
#define mp make_pair
#define eps 1e-3
#define euC 0.57721566490153286060651209;
double a[3][3] = {
{0.5, 0.375, 0.125},
{0.25, 0.125, 0.625},
{0.25, 0.375, 0.375}
};
double b[3][4] = {
{0.6, 0.2, 0.15, 0.05},
{0.25, 0.3, 0.2, 0.25},
{0.05, 0.1, 0.35, 0.5}
};
int o[N];
int n;
double pai[3] = {0.63, 0.17, 0.2};
double dp[N][N];
int path[N][N];
int ans[N];
int Main() {
//freopen("cf.txt","r", stdin);
int T; cin >> T;
for (int tt = 0; tt < T; tt++) {
cin >> n;
for (int i = 0; i < n; i++) {
string s; cin >> s;
if (s == "Dry") o[i] = 0;
else if (s == "Dryish") o[i] = 1;
else if (s == "Damp") o[i] = 2;
else if (s == "Soggy") o[i] = 3;
}
for (int i = 0; i < 3; i++) {
dp[0][i] = log(pai[i]) + log(b[i][o[0]]);
}
for (int i = 1; i < n; i++) {
for (int j = 0; j < 3; j++) {
dp[i][j] = -1e200;
path[i][j] = -1;
for (int k = 0; k < 3; k++) {
if (dp[i][j] < dp[i-1][k] + log(a[k][j]) + log(b[j][o[i]])) {
dp[i][j] = dp[i-1][k] + log(a[k][j]) + log(b[j][o[i]]);
path[i][j] = k;
}
}
}
}
cout << "Case #" << tt + 1 << ":" << endl;
ans[n-1] = 0;
for (int i = 0; i < 3; i++)
if (dp[n-1][i] > dp[n-1][ans[n-1]])
ans[n-1] = i;
for (int i = n-2; i >= 0; i--) {
ans[i] = path[i+1][ans[i+1]];
}
for (int i = 0; i < n; i++) {
if (ans[i] == 0) cout << "Sunny" << endl;
else if (ans[i] == 1) cout << "Cloudy" << endl;
else cout << "Rainy" << endl;
}
}
return 0;
}
int main() {
return Main();
}