Some error when using SM

[Alfred Yu]

Dear all,
Hope you can help me. I am trying to use the SM to analyse some
sequence data.But There are some types of errors.

First error: Discount must be in (0,1.0) (When num_samples is
greater than 40 ,errors will appear)
Second error:IntSamplingNode, "unseating in an empty seating
arrangment".

I don't know what will cause these two error. Would anybody give me
some hints what's reason here?
I really appreciate your help.

[Nicholas Bartlett]

I am working on these bugs and will send another email when they have been resolved

Nick

--
BARTLETT Nick
Department of Statistics
Columbia University, New York

[fw...@stat.columbia.edu]

As I'm sure Nick will announce soon, the sequence memoizer website has
been updated. The primary reason for this update is to release a new
version of the sequence memoizer java implementation. This version
has not been given a new version number but does include a significant
number of bug fixes, both those pointed out by Alfred (thank you by
the way) and also other bugs that we ourselves discovered at
Columbia.

Additionally, an R package for the sequence memoizer has been
developed. It is now available on the website as well. We're eager
to hear about it's use.

On Nov 12, 2:22 pm, Nicholas Bartlett <nbartlet...@gmail.com> wrote:
> I am working on these bugs and will send another email when they have been
> resolved
>
> Nick
>

[Jiangbo Yu]

I  really appreciate your help. Thank you so much.

Alfred.

--
Best Regards,
Jiangbo Yu
School of Computing,
National University of Singapore