dotnet/machinelearning

Add FieldAwareFactorizationMachine to AutoML

Open

#3,985 创建于 2019年7月10日

在 GitHub 查看
 (6 评论) (0 反应) (2 负责人)C# (9,125 star) (1,901 fork)batch import
AutoML.NETP2command-linegood first issue

描述

FieldAwareFactorizationMachine is good for large dataset like the Criteo 1TB dataset.

Currently FieldAwareFactorizationMachine is not swept over in AutoML.

Task:

  • Add trainer to default list of binary learners to try
  • Add sweep range
  • Add to CLI's C# CodeGen

Should be easy to just replicate an existing trainer like SDCA: https://github.com/dotnet/machinelearning/blob/d518b587b06ac3896a48646622b0f2169a230855/src/Microsoft.ML.AutoML/TrainerExtensions/BinaryTrainerExtensions.cs#L150-L169

贡献者指南

技术栈
csharp
领域
machine learning
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
2
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-3 hours
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
blocked
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
C#ML.NET AutoML
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
35
研究方向
Refer to the linked file src/Microsoft.ML.AutoML/TrainerExtensions/BinaryTrainerExtensions.cs for the pattern used by SDCA. Add a new entry for FieldAwareFactorizationMachine similar to SDCA. The sweep range and CLI CodeGen changes should follow the same pattern. Since there are assignees, discuss progress with them.
Add FieldAwareFactorizationMachine to AutoML · dotnet/machinelearning#3985 | Good First Issue