Predictive Analytics: Using SMOTE, Model Explanations, & Hyperparameter Tuning
Predictive Analytics | Intermediate
- 11 videos | 1h 15m 22s
- Includes Assessment
- Earns a Badge
Machine learning (ML) models can struggle with training themselves to identify failures if the dataset's number of machine failures is too low. This is a common problem that occurs when predicting very rare occurrences. Thankfully, oversampling techniques exist to mitigate such issues. In this course, learn how to use SMOTE, a widely used technique to make datasets more balanced. Next, explore model explanations, a feature of Azure Machine Learning. Finally, practice performing hyperparameter tuning by trying different model configurations to see which yields the best performance. Upon completion, you'll be able to improve the performance of a failure detection model, generate records of minority classes, and perform hyperparameter tuning.
WHAT YOU WILL LEARN
Discover the key concepts covered in this courseUse smote to improve the performance of a machine failure prediction modelView important machine failure prediction attributesObserve model explanations and performance metrics for machine failure prediction modelsStandardize data using a componentView the effects of standardizing machine failure data
Create and configure a decision forest model for machine failure predictionPredict machine failure using decision forests and compare the performance to the logistic regression modelCompare the performance of the logistic regression and decision forest modelsPerform hyperparameter tuning on a machine failure prediction model and view the resultsSummarize the key concepts covered in this course
IN THIS COURSE
EARN A DIGITAL BADGE WHEN YOU COMPLETE THIS COURSE
Skillsoft is providing you the opportunity to earn a digital badge upon successful completion on some of our courses, which can be shared on any social network or business platform.Digital badges are yours to keep, forever.
YOU MIGHT ALSO LIKE
Rating 4.4 of 136 users (136)
Rating 4.5 of 935 users (935)
Rating 4.4 of 53 users (53)