How to utilize external recommendation algortithm return in python inside Adobe Target

[RajaShankar]

Hi All,

I have a requirement from customer to use external recommendation algorithm for a estore - product recommendation written in custom python script inside adobe target recommendation. I had checked the criteria supported by Adobe target i am not able to find any criteria to leverage external algorithm. How can we leverage external custom algorithm inside Adobe target. Please suggest.

P:S : As per my understanding Custom Algorithm (Bring your Own) is supported in Adobe Target as mentioned in below guide

https://experienceleague.adobe.com/docs/target/using/recommendations/introduction-to-recommendations...

But there is no details on how to integrate custom logic using custom criteria. Its just allowing to upload a csv file with key and recommendation as columns. Some examples will be helpfull.

@kautuk_sahni @arunpatidar @Jörg_Hoh @BrianKasingli

Any pointers please.

Regards,

Rajashankar.R

[RajaShankar]

Any pointers please

[mikem321 · Accepted solution]

Hey there, 

Although the docs make it sound like you can use your own algorithm, you can't.

By "custom algorithm" they mean custom criteria. They just mean that you can customize all of the OOTB options in the criteria menu to create your own criteria configuration.

The closest thing that you can get to using your own algorithm is running your python algo on your own and uploading the output in the custom criteria configuration as a .csv.

Trust me on this, I had the exact same question and Adobe consulting confirmed it for me.

It's unfortunate because the language used in the docs is so misleading that us developers have to be the bearer of bad news that the product doesn't do something that it pretty clearly claims to do.