This message was deleted.

Hey @Noam Mansovsky I am glad you find sagemaker deployment useful. For batch inferencing, what's your current solution? And what about sagemaker batchtransform attracts you?

Hi @Bo, thanks for replying. I am currently building a batch inferencing solution for simulating models. I can create a solution by myself, but I think a combination of sagemaker and bentoml will be greatest because: 1. With SageMaker BatchTransform, AWS already develops and maintains the scheduling and distributed, scalable calculations, all I have to provide is the model and the dataset file. 2. BentoML takes my model function and automatically wraps it with an interface following the BatchTransform APIs (slightly different than the Model Endpoint API) And then I have almost no work 🏆