The k-NN assignment (part of assignment 1) asks this: "The training error of a 1-NN will always be lower than that of 5-NN." However, the notes at http://cs231n.github.io/classification/ (which should be sufficient to solve the assignment IMHO) do not define that term.
The k-NN assignment (part of assignment 1) asks this: "The training error of a 1-NN will always be lower than that of 5-NN." However, the notes at http://cs231n.github.io/classification/ (which should be sufficient to solve the assignment IMHO) do not define that term.