July 2017
Beginner to intermediate
486 pages
13h 49m
English
Lossy counting is very useful; especially when you have a large corpus and you don't want to consider the words that appear very rarely.
At this time, lossy counting is very useful because the user can set the minimum word frequency count as a threshold, so words that occur less than the threshold frequency count won't be included in our vocabulary.
So, if you have a large corpus and you want to optimize the speed of training, then we can use this algorithm.
In other words, you can say that by using this algorithm you narrow down your vocabulary size, thus, you can speed up the training process.
Read now
Unlock full access