GPU for REST API on AWS?

Question

I have an EasyOCR model that I have trained with personnal data and I need to deploy it and make it available with an API REST. My project is on Github.

Problem: I have saw that we can't use GPU on AWS LAMBDA so how can we deploy a REST API that can use GPU on AWS ?

(EasyOCR is really slow when we don't use GPU with CUDA)

Answer 1

You can have a look at AWS EC2 Instances with GPU Support. There are 3 or 4 classes available (p3, g3, p4 and g4). They should be sufficient for your usecase. Make sure to use the AWS Deep Learning AMI, to get out of the Box NVIDIA Drivers and CUDA Support.

Edit: Ref- https://docs.aws.amazon.com/dlami/latest/devguide/gpu.html