AWS SageMaker
ML Model Lifecycle
Demo Script
Initial Demo Script for SageMaker
1AWS Sagemaker for ML-ops
2
3MLOps Devlopment - Key Problems
4-----------------------------------------
51. Lack of standardization
62. Complexity in model deployment
73. Tools overloaded - Too many tools are needed to maintain the MLOps eco-system
8
9
10Amazon SageMaker - Build, Train, and deploy ML Models with fully managed insfrastructure and tools.
AWS SageMaker Overview
SageMaker Core Components
AWS SageMaker - Core Components
1AWS Sagemker comprises of wide-range of tools
2
31. Amazon SageMaker Studio Lab - For experiments
42. Notebook - For development
53. Amazon SageMaker ground truth - For labeling
64. Canvas - For visualization
75. Shadown Testig - for validation
86. Train - training ml models
97. Model
10
11
121. SageMaker Studio an IDE for ML with tools for data prep, model training, and deployment.
132. SageMaker Notebooks - Cloud-hosted JUpyter Notebooks
143. SageMaker Training - a managed service for scalable ML model training with various algorithms and frameworks.
154. SageMaker INference - Simplify deployment process
165. SageMaker Ground Truth - A data labeling service.
17
186. SageMaker Autopilot - Automatically builds, trains and tunes ML models
197. SageMaker Model Monitor
208. SageMaker pipelines - CICD framework models
219. SageMaker Feature Store -
2210. SageMaker canvas - visualization toll inside AWS SageMaker
SageMaker Use Case - IoT
