Friends, I’m extremely happy to announce that my book on Amazon SageMaker is now available on Amazon.com and on other sites as well.

Image for post
Image for post


This is a guest post by Chaim Rand, Machine Learning Algorithm Developer at Mobileye. You can also read part 1 and part 3 for more!

In a previous post, I told you the story of how my team at Mobileye (officially known as Mobileye, an Intel company), transitioned to using the Amazon SageMaker service to train our TensorFlow neural networks in the cloud. In particular, I told you about how one could use SageMaker Pipe Mode to stream training data directly from Amazon S3 storage to training instances, and how this leads to reductions in both training time and cost.

The easiest way to adopt Pipe Mode, is to use PipeModeDataset, a SageMaker implementation of the TensorFlow Dataset interface, which hides all the low level pipe management from you. Using PipeModeDataset requires reformatting your training data into one of the supported file formats (text records, TFRecord and Protobuf). We chose the TFRecord format. …


In this episode, I go through our latest announcements on Amazon Forecast, Amazon Personalize, Amazon SageMaker Ground Truth, AWS Deep Learning AMIs, and Amazon Elastic Inference.

Video episodes are available on my YouTube channel.


Folding@home is a long running project focused on disease research using distributed computing, and they recently launched a number of projects related to COVID-19.

This blog post will show you how to use Amazon EC2 GPU instances with Folding@home. This is a great way to help researchers, so please consider donating some GPU time.

Initial setup

First, I fire up an Amazon EC2 P3 instance, which hosts an NVIDIA V100 GPU. I use the NVIDIA Deep Learning AMI 19.11.3 in order to make sure that I have the latest NVIDIA drivers. …


In this episode, I focus on Amazon Kendra, an enterprise search service powered by machine learning… but you don’t need any ML skills to set it up and use it! I show you how to create an index, add data sources, and then I run queries using the AWS console and the AWS CLI.

Video episodes are available on my YouTube channel.


In this episode, I talk about XGBoost 1.0, a major milestone for this very popular algorithm. Then, I discuss the three options you have for running XGBoost on Amazon SageMaker: built-in algo, built-in framework, and bring your own container. Code included, of course!

Video episodes are available on my YouTube channel.


In this video, I show you how to train on your local machine using SageMaker APIs. I use Jupyter, and this would also work with your preferred IDE (PyCharm, etc.). This is a friendly and fast way to write and debug your code before running it at scale on managed instances. This technique also saves you money, as you’re not using notebook instances or SageMaker Studio.


In this episode, I talk about XGBoost 1.0, a major milestone for this very popular algorithm. Then, I discuss the three options you have for running XGBoost on Amazon SageMaker: built-in algo, built-in framework, and bring your own container. Code included, of course!

Video episodes are available on my YouTube channel.


In this episode, I cover new features on Amazon Personalize (recommendation & personalization), Amazon Polly (text to speech), and Apache MXNet (Deep Learning). I also point out new notebooks for Amazon SageMaker Debugger, a couple of recent videos that I recorded, and an upcoming SageMaker webinar.

Video episodes are available on my YouTube channel.


In this video, I first train an XGBoost model on my local machine (I use PyCharm), and visualize results in the mlflow UI. Then, I deploy the model locally, and predict test data. Next, I create a Docker container, push it to Amazon ECR, and use it to deploy my model on Amazon SageMaker.

Happy to answer any question you may have!

Documentation: https://www.mlflow.org/

About

Julien Simon

Hacker. Headbanger. Harley rider. Hunter. https://aws.amazon.com/evangelists/julien-simon/

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store