[scikit-learn] Label encoding for classifiers and soft targets
Andreas Mueller
t3kcit at gmail.com
Mon Mar 13 17:17:22 EDT 2017
On 03/12/2017 03:11 PM, Javier López Peña wrote:
> The purpose is two-fold, on the one hand use the probabilities
> generated by a very complex
> model (e.g. a massive ensemble) to train a simpler one that achieves
> comparable performance at a
> fraction of the cost. Any universal classifier will do (neural
> networks are the prime example).
You could use a regression model with a logistic sigmoid in the output
layer.
More information about the scikit-learn
mailing list