[CS241] standardized format?

[Skip]

I'm not used to spreadsheets, sorry about the formatting nonsense.
CSV should be fine.

I'd find it easier to work with and to generalize to other problems if path
were in column 3, instead of splitting each path into a new sheet or a set
of rows.  It isn't hard to transform, but is there a reason we have it the
way it is?

On 10/28/07, Tim St. Clair <timothy_st_clair at brown.edu> wrote:
>
> Sorry, I just got this email.  The format I just sent out is somewhat
> similar.  It is a giant spread sheet: paths across the top, divided up by
> the probablity we're looking for, and tenses (vp0, vp1) down along the left
> side.
>
> Skip: I see how your proposed format is more machine readable, if we want
> to use other viewers for it.  However, if you just want to look at raw
> arrays of numbers, I find mine more usefull.  I will send out an array in
> your format by the end of the night, if you still want it.  (let me check
> over my numbers a little first).
>
> p.s.  Where does everybody else stand on this project / these issues?
>
> On 10/28/07, Skip < dave.hirshberg at gmail.com > wrote:
> >
> > I agree that we should standardize.
> > CSV doesn't work with 3+D data (P(path|vp0,vp1) is 3D) and we'll need to
> > specify more than a data format to have our elements match.
> >
> > A proposal
> >
> > What is the 6th path? What is the 6th tense?
> >  1. Order paths by occurrence in 2-9, most common first.  We'll have to
> > report what they are, too, because we may disagree here.
> >  2. Order the 16 tenses that make sense ([01][01][1-4]) by tree.vals[8],
> > least to greatest
> >      Clump all other tenses together as the 17th tense.
> >
> > Printing this information:
> > Given an order like this, we can trade nD arrays of numbers, and they'll
> > match.
> > Suppose we're trading P(x | y, z) -- think of it as P[i][j][k], an MxNxO
> > array where M is #x, N is #y, O is #z.
> > Print it like this
> > M N O
> > P[0][0][0]
> > P[0][0][1]
> > P[0][0][2]
> > ...
> > P[0][1][0]
> > ...
> >
> > What do you think?
> >
> > On 10/28/07, Tim St. Clair < timothy_st_clair at brown.edu> wrote:Do you
> > think we could agree on some kind of standardized, condensed method of
> > formatting / displaying our results, to make them easier to read and
> > compare?
> >
> > Might I suggest csv
> > I suggest
> >  on the first line, give the size of the array
> >  on each subsequent line, put one element
> >
> > For example:
> >
> > 1 2
> > 1
> > 2
> >
> > or:
> >
> > 1 2 3
> > 1
> > 2
> > 3
> > 4
> > 5
> > 6
> >
> > On 10/28/07, Tim St. Clair < timothy_st_clair at brown.edu> wrote:
> >
> > > Do you think we could agree on some kind of standardized, condensed
> > > method of formatting / displaying our results, to make them easier to read
> > > and compare?
> > >
> > > Might I suggest csv (comma separated value files -- extension .csv,
> > > each rows separated by lines, columns separated by ',' characters), so that
> > > they can easily be imported into any spreadsheet program?
> > >
> > > Thanks.
> > >
> > > --
> > > Tim St. Clair
> > >
> > > (617) 460 - 6497
> > > _______________________________________________
> > > CS241 mailing list
> > > CS241 at list.cs.brown.edu
> > > http://list.cs.brown.edu/mailman/listinfo/cs241
> > >
> > >
> >
>
>
> --
> Tim St. Clair
>
> (617) 460 - 6497
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.cs.brown.edu/pipermail/cs241/attachments/20071028/0f120a2f/attachment.html