COVID-19 icon indicating copy to clipboard operation
COVID-19 copied to clipboard

Create AWS Lambda Layer with PyTorch

Open nietzscheson opened this issue 4 years ago • 1 comments

AWS Lambda have restrictions. PyTorch is too longer. When trying to deploy the project, AWS Lambda reports the out of memory error. PyTorch weighs: 758MB.

There is PyTorch layers with other versions. It could be a problem. We need to control the versions of the dependencies that we are going to use.

An alternative is:

  • Create and publish our own layer. With the specifications that requirements the project.
  • Take PyTorch from requirements.txt and use it only in development or on Docker.

Related: #18, #20, #29, #30

nietzscheson avatar Mar 22 '20 17:03 nietzscheson

We can start prototyping with this layer that its publicly available:

arn:aws:lambda:AWS_REGION:934676248949:layer:pytorchv1-py36:2

caveat: its version 1.1 of pytorch while we are using 1.4

I already checked the layer and its working!

and following this tutorial:

https://course.fast.ai/deployment_aws_lambda.html

jisazaTappsi avatar Mar 23 '20 02:03 jisazaTappsi