Machine LearningLesson 14Free preview
ML Data
This lesson is part of A.I. · All levels
Summary
Data is the main ingredient of machine learning: without quality data, models are useless. Data is generally divided into training data for learning and data testing to gauge the model's capabilities in new examples. Data quality includes completeness, cleanliness, representativeness, and free from harmful bias.
Key points
- The data is divided into training for learning and testing for evaluation.
- General division: most for training, the rest for testing.
- Dirty data (duplicate, missing value, mislabeling) ruins the model results.
- A biased dataset makes models unfair to certain groups.
Full lesson
Code examples and quizzes
Full lessons, code examples, and quizzes are available to subscribers. 5 questions.