Just want know if there is any difference between way 1 and way 2
way 1 (set and update)
set note evidence -> update
set next note evidence -> update
. . .
way 2 (set all of evidence then update once)
set note evidence
set next note evidence
. . .
update
please advise
set and update VS set all evidence and update once
Re: set and update VS set all evidence and update once
Hi
In the end, it doesn't make a difference unless you are interested in the intermediate probabilities after setting evidence one by one.
With either method 1 or 2 , if you check the posteriors of the variables after setting all evidence the result is the same.
Best,
Martijn
In the end, it doesn't make a difference unless you are interested in the intermediate probabilities after setting evidence one by one.
With either method 1 or 2 , if you check the posteriors of the variables after setting all evidence the result is the same.
Best,
Martijn
-
- Site Admin
- Posts: 1417
- Joined: Mon Nov 26, 2007 5:51 pm
Re: set and update VS set all evidence and update once
In any case, you'll get better performance with your method 2. In an extreme case, with very large network, method 1 may not complete (SMILE will be able to perform inference with all evidence, while the subset may prove intractable).