Learning Notebook - David Rostcheck
Public View
learning_event details
Learning Event ID
Subject
Topic
Program
Length
Institution
Presenter
Format
Recorded Date
Completed Date
Notes
Large lambda will create high bias. Too small will create high variance. Use regularization in the training algorithm, but then calculate Jtrain (again), Jtest, and Jcv w/o regularization, pick theta that gives lowest error Jcv. Try a range of lambda, plo
Personal Notes
Link
Review
Return to
main screen