About weights

[Denis Maurel]

Hi,

I try different experiments with weights. It seems to me two points:

1) weights have priority from longest match

2) there is not addition of weights when one use a subgraph with weights. weights seem computed only after the box where are...

Do you agree with these two points?


Best regards,

Denis Maurel


____________________________________
Professor Denis Maurel
Université François Rabelais Tours
LI (Computer Science Research Laboratory)
EPU-DI
64 avenue Jean-Portalis
37200 Tours
France
Phone: 33-2.47.36.14.35
Fax: 33-2.47.36.14.22
mailto:denis....@univ-tours.fr

http://www.univ-tours.fr/maurel

http://www.li.univ-tours.fr
http://tln.li.univ-tours.fr/

[eric.laporte]

Dear Denis,

On Tuesday, 4 February 2014 10:27:08 UTC+1, Denis MAUREL wrote:
<<

1) weights have priority from longest match

>>
No, longest match has precedence over weights: this is documented in the current version of the manual: 'With weights, you can define a priority among paths that match the same sequence. You cannot define a priority among embedded matching sequences' (Section 5.2.4, p.104).
There is already a topic (Suggestions for Unitex and Gramlab, 24 October 2013) with a posting (Precedence between paths in local grammar graphs) suggesting a new option where weights would have precedence over longest match. It has been suggested by Gaëlle Recourcé during the Unitex Gramlab Workshop.
Best,
Eric

[Denis Maurel]

Hi Eric,

Thanks. I think these precedence is good.

What about the possibility of more than one weight in a path?

Best regards,

Denis Maurel


____________________________________
Professor Denis Maurel
Université François Rabelais Tours
LI (Computer Science Research Laboratory)
EPU-DI
64 avenue Jean-Portalis
37200 Tours
France
Phone: 33-2.47.36.14.35
Fax: 33-2.47.36.14.22
mailto:denis....@univ-tours.fr

http://www.univ-tours.fr/maurel

http://www.li.univ-tours.fr
http://tln.li.univ-tours.fr/

Dear Denis,

On Tuesday, 4 February 2014 10:27:08 UTC+1, Denis MAUREL wrote:
<<

>>
Yes, weights have precedence over longest match: this is documented in the current version of the manual, Section 5.2.4, p.104.