unit8co/darts

Write an `NBEATSExplainer` class for the interpretable version

Open

#1,270 建立於 2022年10月9日

在 GitHub 查看
 (1 留言) (0 反應) (0 負責人)Python (6,832 star) (762 fork)batch import
good first issueimprovement

描述

NBEATS has an "interpretable" mode where it decomposes the forecast in trend and seasonality. We should have a class NBEATSExplainer in darts.explainability which exposes these.

貢獻者指南

技術棧
python
領域
machine learningdata
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。
1-3 hours
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。
fresh
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 knowledge of NBEATS modelFamiliarity with darts library
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。
50
研究方向
Examine the existing NBEATS implementation in darts (likely in darts.models.forecasting.nbeats) to understand how the interpretable mode decomposes forecasts into trend and seasonality components. Then, create a new class NBEATSExplainer in darts.explainability that exposes these components, following the patterns used by other explainer classes in the module. Consider reviewing any related discussions in issue comments for additional context.
Write an `NBEATSExplainer` class for the interpretable version · unit8co/darts#1270 | Good First Issue