O'Reilly logo

Hands-On Serverless Deep Learning with TensorFlow and AWS Lambda by Rustem Feyzkhanov

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

General issues with deploying Python frameworks on AWS Lambda

In this section, we will learn about AWS Lambda main limit, which is also known as the size of the package. The current limit for the Lambda deployment package is 50 MB. It is supposed to include libraries and code. There are two main libraries that we need to fit:

  • TensorFlow
  • NumPy

These libraries are used for matrix calculations. As you may know, the libraries by themselves are pretty big and they just wouldn't work on AWS Lambda. As you have already seen in the previous section on deployment that when we deploy them through S3, we don't have this limitation, and we only have 250 MB limitation for the unzipped package. In this case to make it work, we need to reduce the size ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required