YouTube

Got a YouTube account?

New: enable viewer-created translations and captions on your YouTube channel!

English subtitles

← Coverage Buckets cs348 unit4

Get Embed Code
1 Language

Showing Revision 1 created 03/06/2013 by Cogi-Admin.

  1. Certain coverage items have the potential to generate a ton of data. Let's think
  2. about large scalars, like a 32-bit integer. If we would just sample it, all its
  3. possible values, we would generate a ton of data. That would burden the system,
  4. the computers. Consume a lot of memory. Further on, when you have to analyze all
  5. these values, it would be impossible. You don't want to analyze 4 million
  6. potential values. The answer is to introduce groups. These are called buckets of
  7. bins. You can declare these groups implicitly, let's say over a 32-bit integer,
  8. you want to have 10 different groups for values or you can do it explicitly,
  9. where you can define these ranges and even give them names. This is a great
  10. mechanism to manage the data overload and to be analyzed to make sense of these,
  11. lower levels of granularity.