Update error

The engine.
shooltz[BayesFusion]
Site Admin
Posts: 1417
Joined: Mon Nov 26, 2007 5:51 pm

Re: Update error

Post by shooltz[BayesFusion] »

If your evolutionary algorithm produces very complex networks, switch inference algorithm to EPIS (inexact, sampling-based but quire reliable). You may want to adjust the number of samples, use Network.SampleCount property for this.
lablonsk
Posts: 32
Joined: Sat Jun 29, 2019 12:13 pm

Re: Update error

Post by lablonsk »

Good afternoon,
Do you suggest that I put Network.SampleCount = number of nodes?
Thank you.
lablonsk
Posts: 32
Joined: Sat Jun 29, 2019 12:13 pm

Re: Update error

Post by lablonsk »

shooltz[BayesFusion] wrote: Tue Apr 28, 2020 2:48 pm If your evolutionary algorithm produces very complex networks, switch inference algorithm to EPIS (inexact, sampling-based but quire reliable). You may want to adjust the number of samples, use Network.SampleCount property for this.
Good Morning,
I used this method and the error still persisted.
The problem is that I cannot further reduce the amount of information.
There are 42 nodes where each time I insert information at least 50 different information is multiplied by 800x. This is the least I can work on.
Unfortunately I was not able to work with all this data.
shooltz[BayesFusion]
Site Admin
Posts: 1417
Joined: Mon Nov 26, 2007 5:51 pm

Re: Update error

Post by shooltz[BayesFusion] »

I used this method and the error still persisted.
Are you sure you got the same error when inference algorithm was set to EPIS?

BTW, Network.SampleCount defaults to 10000 and can be made much larger. Increasing the sample count does not require more memory, but it makes the inference run longer and produce more accurate results.
lablonsk
Posts: 32
Joined: Sat Jun 29, 2019 12:13 pm

Re: Update error

Post by lablonsk »

shooltz[BayesFusion] wrote: Wed May 13, 2020 1:27 pm
I used this method and the error still persisted.
Are you sure you got the same error when inference algorithm was set to EPIS?

BTW, Network.SampleCount defaults to 10000 and can be made much larger. Increasing the sample count does not require more memory, but it makes the inference run longer and produce more accurate results.
unfortunately the error occurred yes, net.BayesianAlgorithm = Network.BayesianAlgorithmType.EpisSampling ;.
At first I don't need anything very precise, but rather fast. My file has 40000 pieces of information and for each sample I duplicate it 800x, 50 different samples being 50x800. I perform around 840000 update and inference, 400 Baysian network.
I'm running 1050 tests like this on a Gcloud server right now.
Post Reply