search

AICoE / prometheus-anomaly-detector

A newer more updated version of the prometheus anomaly detector (https://github.com/AICoE/prometheus-anomaly-detector-legacy)
GNU General Public License v3.0
597 stars 151 forks source link

How to get multiple timestep predictions into the future #130

Closed tarakdatanow closed 3 years ago

tarakdatanow commented 4 years ago

Hi,

Which values/parameters are used to get multi-step output into the future? As per my understanding, currently I can only see 1 output value on the Tornado web app with four different vaues for each reading - yhat, yhat-lower, yhat-upper and anomaly.

How can I get multiple timesteps into future?

Example: Using last 100 days data, how can I predict for the next 10 days?

4n4nd commented 4 years ago

currently I can only see 1 output value on the Tornado web app

that's because these outputs are for Prometheus to scrape and the anomaly value can only be calculated using the current metric value

tarakdatanow commented 4 years ago

Okay, how should I get the forecasting for multiple input timeseries? Also, how can I plot it on a dashboard just as you've shown in your examples in the Readme?

4n4nd commented 4 years ago

You can use Grafana to plot metrics once the predicted metrics are scraped by the Prometheus instance

sesheta commented 3 years ago

Issues go stale after 90d of inactivity. Mark the issue as fresh with /remove-lifecycle stale. Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

/lifecycle stale

chiaminchuang commented 3 years ago

Hello @tarakdatanow, I have the same issue. Do you solve this problem? Thanks in advance.

You can use Grafana to plot metrics once the predicted metrics are scraped by the Prometheus instance

Hello @4n4nd , In the source code, the Prometheus instance scrapes one prediction whose timestamp is closest to the current timestamp. We cannot plot the metrics on the Grafana dashboard. How do we make the Prometheus instance scrape the future predictions not only the most recent one?

Thank you for your great repo.

sesheta commented 3 years ago

Stale issues rot after 30d of inactivity. Mark the issue as fresh with /remove-lifecycle rotten. Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

/lifecycle rotten

sesheta commented 3 years ago

Rotten issues close after 30d of inactivity. Reopen the issue with /reopen. Mark the issue as fresh with /remove-lifecycle rotten.

/close

sesheta commented 3 years ago

@sesheta: Closing this issue.

In response to [this](https://github.com/AICoE/prometheus-anomaly-detector/issues/130#issuecomment-968179091): >Rotten issues close after 30d of inactivity. >Reopen the issue with `/reopen`. >Mark the issue as fresh with `/remove-lifecycle rotten`. > >/close Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.