示例画廊

指标

最长公共子序列

Longest Common Subsequence

LB_Keogh

LB_Keogh

规范时间扭曲

Canonical Time Warping

sDTW 多路径匹配

sDTW multi path matching

使用自定义距离度量的最长公共子序列

Longest Commom Subsequence with a custom distance metric

动态时间规整

Dynamic Time Warping

软动态时间规整

Soft Dynamic Time Warping

使用自定义距离度量进行DTW计算

DTW computation with a custom distance metric

最近邻

k-NN 搜索

k-NN search

最近邻

Nearest neighbors

使用KNeighborsTimeSeriesClassifier进行管道的超参数调优

Hyper-parameter tuning of a Pipeline with KNeighborsTimeSeriesClassifier

使用SAX + MINDIST的1-NN

1-NN with SAX + MINDIST

聚类和重心

KShape

KShape

Kernel k-means

Kernel k-means

重心

Barycenters

Soft-DTW加权重心

Soft-DTW weighted barycenters

k-means

k-means

分类

SVM和GAK

SVM and GAK

学习形状特征

Learning Shapelets

早期分类

Early Classification

将发现的shapelets与时间序列对齐

Aligning discovered shapelets with timeseries

学习形状：二维距离空间中的决策边界

Learning Shapelets: decision boundaries in 2D distance space

自动微分

用于PyTorch神经网络的Soft-DTW损失

Soft-DTW loss for PyTorch neural network

杂项

模型持久化

Model Persistence

PAA和SAX特征

PAA and SAX features

矩阵轮廓

Matrix Profile

距离和矩阵轮廓

Distance and Matrix Profiles