日韩性视频-久久久蜜桃-www中文字幕-在线中文字幕av-亚洲欧美一区二区三区四区-撸久久-香蕉视频一区-久久无码精品丰满人妻-国产高潮av-激情福利社-日韩av网址大全-国产精品久久999-日本五十路在线-性欧美在线-久久99精品波多结衣一区-男女午夜免费视频-黑人极品ⅴideos精品欧美棵-人人妻人人澡人人爽精品欧美一区-日韩一区在线看-欧美a级在线免费观看

歡迎訪問 生活随笔!

生活随笔

當(dāng)前位置: 首頁 > 编程语言 > python >内容正文

python

python增量更新数据,Python中的增量最近邻算法

發(fā)布時間:2023/12/19 python 23 豆豆
生活随笔 收集整理的這篇文章主要介紹了 python增量更新数据,Python中的增量最近邻算法 小編覺得挺不錯的,現(xiàn)在分享給大家,幫大家做個參考.

Is anyone aware of a nearest neighbor algorithm implemented in Python that can be updated incrementally? All the ones I've found, such as this one, appear to be batch processes. Is it possible to implement an incremental NN algorithm?

解決方案

I think the problem with incremental construction of a KD-tree or KNN-tree is, as you've alluded to in a comment, that the tree will eventually become unbalanced and you can't do simple tree rotation to fix balance problems and keep consistency. At the minimum, the re-balancing task is not trivial and one would definitely not want to do it at each insertion. Often, one will choose to build a tree with a batch method, insert a bunch of new points and allow the tree to become unbalanced up to a point, and then re-balance it.

A very similar thing to do is to build the data structure in batch for M points, use it for M' points, and then re-build the data structure in batch with M+M' points. Since re-balancing is not normal, fast algorithm we are familiar with for trees, rebuilding is not necessarily slow in comparison and in some cases can be faster (depending on how the sequence of the points entering your incremental algorithm).

That being said, the amount of code you write, debugging difficulty, and the ease of others' understanding of your code can be significantly smaller if you take the rebuild approach. If you do so, you can use a batch method and keep an external list of points not yet inserted into the tree. A brute force approach can be used to ensure none of these is closer than the ones in the tree.

Some links to Python implementations/discussions are below, but I haven't found any that explicitly claim to be incremental. Good luck.

Note: My comments here apply to high-dimensional spaces. If you're working in 2D or 3D, what I've said may not be appropriate. (If you working in very high dimensional spaces, use brute force or approximate nearest neighbor.)

總結(jié)

以上是生活随笔為你收集整理的python增量更新数据,Python中的增量最近邻算法的全部內(nèi)容,希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯,歡迎將生活随笔推薦給好友。