This flashcard is just one of a free flashcard set. See all flashcards!
70
Training and testing procedure for k-NN?
Training
Basically non-existing. Store the training samples and maybe perform some preprocessing to speed up queries (feature extraction, dimensionality reduction)
Testing
For classification look at the k nearest neighbors and pick the class with the most votes.
For regression average the values of the k nearest neighbors.
Basically non-existing. Store the training samples and maybe perform some preprocessing to speed up queries (feature extraction, dimensionality reduction)
Testing
For classification look at the k nearest neighbors and pick the class with the most votes.
For regression average the values of the k nearest neighbors.
Tags:
Source: CI Teil 1 Lecture 6
Source: CI Teil 1 Lecture 6
Flashcard info:
Author: Sepp Samuel
Main topic: Telematik
Topic: Computational Intelligence
School / Univ.: TU Graz
Published: 02.07.2014