Preventing high cost with Aws Personalize

I was testing AWS Personalize recommenders and I realize that I was using all which charge for real-time recommendations. Is there a way to use them not for charge on real-time?

Tópicos

Análise Machine Learning e IA

Tags

Análise Machine Learning e IA Amazon Personalize

Idioma

English

AWS-User-3037600

feita há 2 anos736 visualizações

1 Resposta

Mais recentes
Mais votos
Mais comentários

Resposta aceita

Yes, take a look at asynchronous batch for Amazon Personalize here:

https://docs.aws.amazon.com/personalize/latest/dg/recommendations-batch.html

ESPECIALISTA

AshleyMiller

respondido há 2 anos

ESPECIALISTA

Fabrizio@AWS

avaliado há 2 anos

AWS-User-3037600
há 2 anos
Hi @amiller. Thanks for your answer. After tried this approach, I'm having this error on output file: "output":{"recommendedItems":[]},"error":"User error: Required fields for the provided recipe are userId"}. My input file its a json with:

{"userId":"244922300825"} {"userId":"244922300827"} ...
AshleyMiller ESPECIALISTA
há 2 anos
Can you post your the line of code that invokes the Personlize API?

Conteúdo relevante

Por que recebi o erro de IAM “AWS was not able to validate the provided access credentials” (A AWS não pôde validar as credenciais de acesso fornecidas) em algumas regiões da AWS?
AWS OFICIALAtualizada há 2 anos
Why can't I detach or delete an elastic network interface that Lambda created?
AWS OFICIALAtualizada há 2 meses
Como resolvo o erro "[AWS service] was unable to place a task because no container instance met all of its requirements" no Amazon ECS?
AWS OFICIALAtualizada há um ano
How do I connect to my WorkSpace with RDP?
AWS OFICIALAtualizada há 2 anos