Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 2: Automating Machine Learning (8 Criteria for ML excellence:…
Chapter 2: Automating Machine Learning
5 STEPS:
Define project objectives
acquire and explore data
model data
interpret and communicate
implement, document, and maintain
not a linear process, involves going back and repeating steps, etc
Google is finding that Auto ML outperforms their smartest data scientists
very time andknowlege intensive
Air BnB example...
Auto ML was better for:
exploratory data analysis
feature engineering
algorithm selection and hyperparameter tuning
model diagnostics
What Auto ML is NOT:
still several decisions to be made by data analyst
expert myst decide whta problems are even worth solving
8 Criteria for ML excellence:
Accuracy
Productivity
Ease of Use
Understanding and learning
resource availibility
process transparency
generalizable across contexts
recommend action