We're in development. Going live soon.

Data Pipeline Design - ML System Design | OneNoughtOne

Loading content...

Data Quality & Analytics Engineering0/34 completed

Data Normalization Techniques: Z-Score and Min-Max Scaling
Synchronized Data Permutation
Dataset Batch Generator for Training Pipelines
Dataset quality checks reject values on valid limits
Include the configured data-quality endpoints

1.Requirements Gathering 2.Data Pipeline Design 3.Model Architecture 4.Serving Infrastructure 5.Monitoring and Maintenance

One bad dataset row consumes several quality allowances
Count bad dataset rows once
Published datasets differ from the snapshot that passed review
Publish the rows captured by the quality run
Design a Machine Learning Feature Store

Course overview & roadmap

101OneNoughtOne

0/34

Data Pipeline Design

Sign in with a free account to read every lesson and save your progress.

Read the first lesson sample