automl/auto-sklearn

FastICA: `n_components is too large: it will be set to _`

Open

#240 建立於 2017年2月24日

在 GitHub 查看
 (4 留言) (0 反應) (0 負責人)Python (7,270 star) (1,265 fork)batch import
Good first issueenhancement

描述

FastICA currently looks for n_components between 10 and 2000, with a default of 100. However, assuming there are more rows N than columns M, a data set can only support up to M independent components. The suggested range is therefore between 2 and M, with a default of round(M / 2).

Looking at the implementation, it appears information about the data set is not available in get_hyperparameter_search_space. As a workaround, the n_components property could be replaced by a n_components_rel property that indicates the amount of independent components relative to the total number of features M. When calling sklearn.decomposition.FastICA, you could then simply pass n_components=X.shape[1] * self.n_components_rel.

貢獻者指南

技術棧
pythonscikit learn
領域
machine learning
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。
half day
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
Pythonscikit learnauto sklearnFastICA concepts
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。
70
研究方向
Examine the FastICA component implementation in auto sklearn, focusing on the `get hyperparameter search space` method. The issue suggests replacing `n components` with a relative property `n components rel` that derives from the number of features. Look for the existing hyperparameter definitions and modify them to accept a relative value, then adjust the call to `sklearn.decomposition.FastICA` accordingly. The discussion in the issue might provide further insights.