dotnet / dotnet/machinelearning-samples

Could not detect change points by DetectIidChangePoint

Open
#800 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
PowerShell
Stars
4.7k
Forks
2.7k
Avg merge
2d 22h
Merged PRs (30d)
1

Description

Hi, I'm quite new for ML.
I have data that looks like:
![changepoint](https://user-images.githubusercontent.com/3241829/79059719-f1702100-7caf-11ea-9590-223eaa019bb9.png)

I'm using ` mlContext.Transforms.DetectIidChangePoint` with different combination of `confidence` and `changeHistoryLength` (by nested `for loops`), but still could not get the right change point of that **unusal, fast rising** area, though it's obvious for human, am I missing sth?

Contributor guide

Open the contributing guide

Research direction

Start with the MLContext.Transforms.DetectIidChangePoint entry point and reproduce the reported behavior using the data shown in the issue. Review the API's confidence and changeHistoryLength behavior, then determine whether the missed fast-rising area is a configuration problem or an algorithm limitation.

Written by the indexing model from the issue text.

Assessment

Tech stack
csharp, machine-learning
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.