In this course, you will:
- The NannyML library and its basic capabilities are introduced. You will first learn how to prepare raw data for reference and analysis sets ready for production monitoring.
- You will study forecasting the tip amount for taxi rides in New York as a practical example. You will also learn how to estimate the performance of the tip prediction model using NannyML at the end of the chapter.
- When ground truth becomes available, realized performance calculators are provided.
- Learn about more advanced ways for dealing with findings, including as filtering, graphing, converting to data frames, chunking, and setting custom thresholds.
- Use this understanding to compute the business value of a model trained on the hotel booking dataset.
- Discover how to find the root cause. You will learn about multivariate and univariate drift detection methods in this chapter.
- Learn how to spot data quality concerns and how to solve the underlying issues you find.
- Data preparation and performance estimation
- Monitoring Performance and Business Value
- Root Cause Analysis and Issue Resolution