Install Multiple Package using Sagemaker Life Cycle configuration file

Question

I wanted to know how I can pre-install Python packages in Sagemaker before spinning it up?

For example, I want to install Tensorfliw, LightFM, and Scikit-optimize

How can i code a Lifecycle Configuration file which will tell sagemaker to install these packages before it spins up and have it ready when I am ready to code.

I know the following resources:

1. https://aws.amazon.com/blogs/machine-learning/customize-your-amazon-sagemaker-notebook-instances-with-lifecycle-configurations-and-the-option-to-disable-internet-access/
2. https://docs.aws.amazon.com/sagemaker/latest/dg/nbi-add-external.html

Answer 1

For Tensorflow, there is an existing Conda environment(tensorflow_p36) with TensorFlow preinstalled that you can use. For other packages that aren't present by default, you can use this Lifecycle Configuration script sample to install them into the tensorflow_p36 environment.