Machine Learning on AWS

This post are some notes of the course done related to ML on AWS (8 hours). Table of contents Module 1: Introduction to Machine LearningModule 2: Artificial Intelligence Services on AWSModule 3: Machine Learning ProcessModule 4: Data Collection, Integration, Preparation, Visualization, and AnalysisModule 5: Deep Learning Amazon Machine ImagesModule 6: Amazon SageMaker ConceptsModule 7: Amazon … Read more

Backstage of Amazon investments

This week Amazon announced the new “data region” Amazon Web Services (AWS) is now opened. Where the hell is Aragon? Here: The AWS Europe (Spain) Region has three Availability Zones located in: Villanueva de Gállego, El Burgo de Ebro, and the Plhus logistics platform in Huesca, where your applications are reliably distributed cloud. What does it mean in … Read more

Who is the dietitian of your cloud?

The cloud bill is an issue in many organizations. This is not something new, it comes from more than 5 -7 years ago when cloud adoption started to establish as the standard. I use to do the joke when asking the infrastructure teams about “who is the dietitian of your cloud?”, “who defines the menu … Read more

Amazon SageMaker + Spark

Some screenshots and notes for my poor memory ML Pipeline with PCA on Spark, and K-Means on Amazon SageMaker Apache Spark is an open-source unified analytics engine for large-scale data processing.  PCA = principal components analysis. Collaborative Filtering Deep Structure Semantic Module (DSSM) A matrix factorization solution in its core is multiplication of 2 matrices. Neural Networks are … Read more

AWS Honeycode

Honeycode is the low-code service in beta version being implemented by Amazon. Things I have noticed: The features are very basic, the templates available basic too. The customers using it are not top tier customers. It’s beta. The number of vacancies opened in is low and last role available is from April 2022. The … Read more

AWS Sagemaker

These are some notes about the basis of Sagemaker Sagemaker services SageMaker Neo optimizes the trained model and compiles it into an executable. Taking the target hardware where the model will be run as input; the compiler uses a ML model to apply performance optimizations on your model. Ground truth makes easy to label data. … Read more