Classifying Performance Test Results Using Machine Learning
summary
Covers basic machine learning algorithms and their application areas, then shows a practical example of building a classification system for Unity performance test results produced by Unity’s Performance Test Framework to detect performance regressions or flaky tests. Also examines criteria for evaluating ML algorithm performance and methods for debugging those algorithms.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:55What is machine learning
- 6:59ML algorithm types
- 7:20Performance test classification
- 9:03Classification as task
- 10:10Neural networks explained
- 15:38Neural network demo
- 18:35Training results and evaluation
- 21:03Algorithm performance
- 22:29Train-test split
- 23:54Underfitting and overfitting
- 26:25Model complexity
- 27:48Data quality
- 28:31Confusion matrix
- 30:43Feature selection and normalization
- 32:11Summary
- 33:35Learning resources
- 35:49Q&A
related talks
39mMethodology of ML in Performance Testing (How to Find Bottlenecks Automatically)
40mThe Applicability of Machine Learning and Neural Networks in Test Automation
53mUnleashing the Black Box: Testing Machine Learning Systems
50mPractical Applications of ML in Software Testing
01h 04mWhat Testers Need from Machine Learning and Artificial Intelligence
31mTesting Machine Learning Models
51mHow to Test Artificial Intelligence and Machine Learning Solutions?
10mMachine Learning & Unity ML-Agents update