Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 2-7: (Chapter 2: Automating Machine Learning (Criteria for AutoML…
Chapter 2-7:
Chapter 2: Automating Machine Learning
process of automating machine learning, very time/knowledge intensive process
exploratory data analysis
feature engineering
algorithm selection and hyperparameter tuning
model diagnostics
autoML is not automatic ML
still need people to analyze
platforms used
context-specific tools
general platforms
Criteria for AutoML Excellence
accuracy
productivity
ease of use
understanding and learning
resource availability
process transparency
generalizable across contexts
recommend actions
Chapter 3. Specify Business Problem
perfect accuracy is of no value if wrong problem is addressed
must be described and shared with stakeholders
needs to be audience focused
Chapter 7: Identify Success Criteria
consider these factors
who will use the model?
is management on board with project?
can the model drivers be visualized?
how much value can the model produce?
Risks
model risks
too predictive or not predictive enough
ethical risks
privacy
cultural risks
missions and factors leading to success
environmental risks
external environment that provided data for your analysis changes drastically
Chapter 4: Acquire Subject Matter Expertise
need some sort of expertise to be capable
Chapter 5: Decide on Unit of Analysis
involves intricate understanding of a problem's real world context
what prediction target is usually will tell unit of analysis
everything will be different
Chapter 6: Define Prediction Target
behavior of the "thing" we need to know about the future
without a target, there is no way for humans or machines to learn what associations drive an outcome