ML crash course - Numerical data

Machine learning crash courseNumerical data 챕터

developers.google.com/machine-learning/crash-course/numerical-data

Introduction

How a model ingests data using feature vectors

First steps

Visualize your data

Statistically evaluate your data

Find outliers

Programming exercises

https://developers.google.com/machine-learning/crash-course/numerical-data/programming-exercises

Normalization

Linear scaling

Z-score scaling

Log scaling

Clipping

Summary of normalization techniques

Binning

Binning example: number of shoppers versus temperature

Quantile Bucketing

Scrubbing

Qualities of good numerical features

Clearly named

Checked or tested before training

Sensible

Polynomial transforms

Conclusion

What’s next

2024 © ak