tasks:
1. unsupervised knn
https://scikit-learn.org/stable/modules/neighbors.html#unsupervised-neighbors
2. isomap
3. largevis
4. line
5.umap
1. how to construct a KNN graph?
常见的方法一般有三类: i. space-partitioning trees; ii. locality sensitive hashing; iii. neighbour exploring techniques.
Reference:
2. 聚类算法实践(二)——谱聚类、Chameleon聚类 waiting to read.
3. K-dimension tree, k-d tree. Use the k-d tree to implement knn algorithm.