On Thursday, May 8, 2014 5:59:10 PM UTC-7, William wrote: > > Do you recall if you handled the underflow problem in your > implementation? > > I believe it does not. > > > I haven't studied the code yet, but it seems like this could be the > culprit. > > I think you're right. You should implement it! >
I had a look at the code an it appears that it *is* already handling the underflow problem. The scaling factors are computed in _forward_scale_all() and used in both _forward_scale_all() and _backward_scale_all(). Also _viterbi_scale() is using log probabilities to avoid underflow in products of small probabilities. So I need to dig deeper. btw I am new to both sage and cython. I am eager to find the cause and fix this though. So here's my question: If I make a change to hmm.pyx, how do I get sage to pick up that change without having to rebuild all of sage from source? (that took a few hours). I read here that I can "attach" a .pyx file which should force a cython recompilation of hte file whenever the .pyx file is changed. Is that right? http://www.sagemath.org/doc/developer/coding_in_cython.html#attaching-or-loading-spyx-files I tried that and got a syntax error: sage: attach "/home/jhersch/bin/sage-6.2/src/sage/stats/hmm/hmm.pyx" File "<ipython-input-4-162f4bbc7027>", line 1 attach "/home/jhersch/bin/sage-6.2/src/sage/stats/hmm/hmm.pyx" ^ SyntaxError: invalid syntax What is the usual way sage developers go about making changes in cython code without rebuilding everything? Thanks! Jesse -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
