feast-dev/feast

Loading 600K features from a parquet file

Open

#4,589 建立於 2024年9月30日

在 GitHub 查看
 (5 留言) (0 反應) (1 負責人)Python (5,029 star) (896 fork)batch import
good first issuekind/bugpriority/p2wontfix

描述

Context

Given the sample files from the OpenShift AI tutorial - Fraud detection example, repo link, including 600,000 rows modelling a fraud detection dataset.

Objective

I want to model the same training dataset as a Feast offline store using a FileSource and then fetch all the historical features at once.

See all the details to replicate the issue here: https://github.com/dmartinol/feast_issue_600K

貢獻者指南

技術棧
pythonpandas
領域
datamachine learning
議題類型
bug
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。
active
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
mostly clear
前置要求
Understanding of Feast's FileSourceExperience with large parquet datasetsKnowledge of Python and pandas
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。
40
研究方向
Review the reproduction repository at https://github.com/dmartinol/feast issue 600K for exact steps and configuration. Investigate the Feast FileSource and offline store code in the feast dev/feast repository, specifically the handling of parquet files with many columns. Check any comments in the issue for additional context from the assignee.