soft dtw用代码写出

以下是Python实现的soft dtw代码：

import numpy as np
from scipy.spatial.distance import euclidean

def soft_dtw(x, y, gamma):
    # 计算距离矩阵
    D = np.zeros((len(x), len(y)))
    for i in range(len(x)):
        for j in range(len(y)):
            D[i, j] = euclidean(x[i], y[j])

    # 初始化cost矩阵和accumulator矩阵
    M = np.zeros((len(x)+1, len(y)+1))
    R = np.ones((len(x)+1, len(y)+1, 2)) * np.inf
    R[0, 0] = 0

    # 计算cost矩阵
    for i in range(1, len(x)+1):
        for j in range(1, len(y)+1):
            diagonal = R[i-1, j-1]
            vertical = R[i-1, j]
            horizontal = R[i, j-1]
            r_min = np.argmin([diagonal, vertical, horizontal])
            if r_min == 0:
                r = diagonal
            elif r_min == 1:
                r = vertical
            else:
                r = horizontal
            R[i, j, 0] = i-1+r_min
            R[i, j, 1] = j-1+r_min
            M[i, j] = D[i-1, j-1] + r

            # 软对角惩罚
            if abs(i-j) > gamma:
                M[i, j] += np.inf
            
            # 更新accumulator矩阵
            s = (M[i-1, j] - M[i, j], 
                 M[i, j-1] - M[i, j])
            R[i, j, :] += np.array(s)

    return M[-1, -1]

其中x和y为两个时间序列，每个时间点上的数值是一个向量。gamma是软对角惩罚的超参数。函数返回x和y之间的soft DTW距离。