SageMaker Fridays Season 3, Episode 1 - The complete ML lifecycle with Amazon SageMaker

⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos ⭐️⭐️⭐️

This 90-minute special is the perfect starting point for SageMaker beginners and experienced users alike. After a quick introduction to SageMaker, we walk you through the 9 Sagemaker launches from AWS re:Invent 2020: what they are, what problems they solve, and a quick demo.

* SageMaker Data Wrangler: data preparation
* SageMaker Clarify: bias detection and model explainability
* SageMaker Feature store: offline and online storage for your engineered features
* SageMaker JumpStart: one-click deployment for ML solutions and pre-trained models
* SageMaker Data Parallelism: optimize large scale distributed training jobs
* SageMaker Model Parallelism: automatically split and train large models on a GPU clusters
* Profiling capability in SageMaker Debugger: collect and visualize training performance metrics with no code change
* SageMaker Pipelines: automate model deployment end-to-end with quality gates
* SageMaker Edge Manager: manage multiple ML models on edge devices

[juliensimonfr]

00:00 Welcome and quick introduction to SageMaker
09:20 Bias detection with SageMaker Clarify
16:30 Data preparation with SageMaker Data Wrangler
23:10 Storing features with SageMaker Feature Store
31:00 Deploying off the shelf solutions and models with SageMaker JumpStart
41:14 Training on large datasets with SageMaker Data Parallelism
49:10 Training very large models with SageMaker Model Parallelism
54:00 Profiling training jobs with SageMaker Debugger
1:02:05 Model explainability with SageMaker Clarify
1:03:25 Building end to end deployment workflows with SageMaker Pipelines
1:17:20 Managing models at the edge with SageMaker Edge Manager
1:21:00 Wrap up

[ElboxD]

I didn´t know I understood french! Oh là là !

[patricktchuente]

Salut, Episode 2 ou? je ne truves pas de Live quelequepart