Could I save the predicted value after using ML model?

[rickyhsu7]

As title, I am using Splunk Machine Learning Toolkit now.
I'm confused about whether I could save the result of predicting.
In my opinion, I might save the result into my raw data. That's exactly what I want to do.
Besides, could I make predictions while uploading raw data?
Thank you.

[skoelpin]

Are you wanting to save the metrics? Why not just apply the model each time you want to use it?

[rickyhsu7]

Well... I want to monitor a folder, and I want to predict some value by an existing model when the data in the folder is uploaded to Splunk.

[skoelpin]

What? This makes zero sense.

What exactly are you trying to predict? Are you trying to forecast or predict? What type of algorithm are you using? Once again, why not just apply the model rather than saving the results?

[rickyhsu7]

In my opinion, I might find the solution of the issue.
Since I want to alert user that there are something wrong after predicting, the immediate predicting is necessary. Thank you for your advice.

[rickyhsu7]

Immediately

[jkat54]

Would the “collect” command suffice? Aka summary indexing?

http://docs.splunk.com/Documentation/Splunk/7.2.0/SearchReference/Collect