scalanlp/breeze

Windowed Time Series Data in Matrices

Open

#586 建立於 2016年9月22日

在 GitHub 查看
 (0 留言) (0 反應) (0 負責人)Scala (3,453 star) (690 fork)batch import
help wantedproject

描述

As new data becomes available you might need to drop the oldest column or row and insert a new column or row with the new data. Is it possible to provide this functionality with a circular indexing along the particular dimension?

貢獻者指南

技術棧
scala
領域
data
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
1-2 days
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
stale
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
basic Scalafamiliarity with Breeze
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
30
研究方向
This is a feature request for windowed time series data handling in Breeze matrices. The core idea is to enable circular indexing along a dimension to efficiently drop old columns/rows and insert new ones. To implement, explore the current matrix slicing and indexing capabilities in Breeze (e.g., DenseMatrix, CSCMatrix). Look at existing views or slice operations, then design a wrapper or additional method that uses modular arithmetic for index mapping. Since the issue is old and has no comments, it would be wise to open a discussion on the Breeze issue tracker to clarify design requirements and seek maintainer input before implementation.