Skip to content
This repository has been archived by the owner on Nov 13, 2021. It is now read-only.

Anomaly on small data #89

Open
doriwal opened this issue Aug 17, 2017 · 1 comment
Open

Anomaly on small data #89

doriwal opened this issue Aug 17, 2017 · 1 comment

Comments

@doriwal
Copy link

doriwal commented Aug 17, 2017

I have a basic question

I have small amount of data , its data of 1 day with 24 hours records
I want to find anomaly , which is in the example below the record with [2017-09-15 01:00:00,5]

my data is
[2017-09-15 00:00:00,140]
[2017-09-15 01:00:00,5]
[2017-09-15 02:00:00,141]
[2017-09-15 03:00:00,144]
...
[2017-09-15 23:00:00,135]

I used AnomalyDetectionVec instead of AnomalyDetectionTs
res = AnomalyDetectionVec(df[,2], max_anoms=0.02,period= 24 , direction='both', plot=TRUE )

Its not working , I tried to change period value to 1 or 2 and still not working

Does this library works on small amount of data ? when I tested it with large set of data It did worked

@doriwal
Copy link
Author

doriwal commented Aug 17, 2017

I found the solution
I changed

res = AnomalyDetectionVec(df[,2], max_anoms=0.2,period= 11 , direction='both', plot=TRUE , only_last=TRUE )

I added only_last

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant