Skip to content

Platt scaling

Carolin Hainke edited this page Feb 25, 2018 · 3 revisions
  • transforms output from a classification model to a probability distribution over classes
  • fits a logistic regression model to a classifier's score
  • estimates probability P(y=1|x), even though the classifier does not provide this probability:
    • P(y=1|x)= 1 / 1 + exp (A f(x) + B)
    • A and B

https://en.wikipedia.org/wiki/Platt_scaling

Clone this wiki locally