Test Time Complexity

From GM-RKB
(Redirected from test time)
Jump to navigation Jump to search

A Test Time Complexity is a Computational Complexity Time that corresponds to a time interval that takes for a machine learning algorithm to make prediction after applying a learned model.



References

2019

2017

2012

2008

kNN with preprocessing of training set
training $\Theta(\vert\mathbb{D}\vert L_{ave})$
testing $\Theta( L_{a} + \vert \mathbb{D} \vert M_{ave} M_{a})= \Theta(\vert\mathbb{D}\vert M_{ave} M_{a})$
kNN without preprocessing of training set
training $\Theta(1)$
testing $\Theta( L_{a} + \vert \mathbb{D} \vert L_{ave} M_{a}) = \Theta(\vert\mathbb{D}\vert L_{ave} M_{a})$
Table 14.3: Training and test times for kNN classification.$ M_{ave}$ is the average size of the vocabulary of documents in the collection.