From the course: Data-Centric AI: Best Practices, Responsible AI, and More
Join today to access over 24,700 courses taught by industry experts.
Data validation, train-test validation, and model validation
From the course: Data-Centric AI: Best Practices, Responsible AI, and More
Data validation, train-test validation, and model validation
- So far, we have covered the conceptual foundations of data-centric AI, as well as how to operationalize it within MLOps system. Now it's time to put these concepts into action with concrete skills and examples. We'll start by revisiting core techniques like data validation, pre-processing, and exploratory analysis. I'll provide you additional details on methods for handling missing data, detecting outliers, and cleaning your data. This discussion will set the stage for hands-on demonstration, using TensorFlow data validation to compute statistics, identify anomalies, profile your data, and validate the models that you're building. By actually applying data-centric AI tooling to real datasets, you'll gain the experience with practical methods for ensuring data quality, robustness, and ethics. This section ties together the previous conceptual foundations with tangible skills so you can apply in your own projects. We…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.