linkedin / linkedin/greykite

ChangePoint detection for not daily data

Open
#134 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.9k
Forks
107
PR merge metrics
No merged PRs in 30d

Description

My data consists of records taken every 20 seconds, as shown below. However, when I performed change point detection with the following parameters, I was able to detect change points, but as shown in the attached figure, the trend line ended up being straight. I've read through the documentation but haven't been able to resolve this issue. I would appreciate it if you could help me.

data
beh_time | count
2023-01-01 10:01:20 | 50
2023-01-01 10:01:40 | 47
2023-01-01 10:02:00 | 46
2023-01-01 10:02:20 | 48
2023-01-01 10:02:40 | 52
....

parameter
res = model.find_trend_changepoints(
df=df,
time_col="beh_time", # time column name
value_col="count",
actual_changepoint_min_distance=timedelta(seconds=300),
resample_freq=timedelta(seconds=20),
regularization_strength=0.1,)

result
greykite_event

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the reported model.find_trend_changepoints call and its time_col, value_col, resample_freq, and regularization_strength parameters. Review the relevant documentation and reproduce the supplied 20-second data example; done means explaining or correcting why the resulting trend line is straight.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.