 krasimir
|
be9ad26aea
|
refactor the API for random generation again. Now PGF contains probabilities in the abstract syntax
|
2010-10-02 13:03:57 +00:00 |
|
|
krasimir
|
2069d9c9eb
|
Yay!! Direct generation of PMCFG from GF grammar
|
2010-06-18 12:55:58 +00:00 |
|
|
krasimir
|
5e2feb4b60
|
since now we don't do common subexpression elimination for PGF we could simplify the PMCFG generation
|
2010-05-26 09:37:32 +00:00 |
|
|
krasimir
|
0b6b30d4a8
|
use the native unicode support from GHC 6.12
|
2010-04-19 09:38:36 +00:00 |
|
|
krasimir
|
ce99935812
|
now for every category we store, in PGF, the list of functions for it in source-code order. The order matters for the termination of the exhaustive generation with dependent types.
|
2010-03-24 11:11:55 +00:00 |
|
|
krasimir
|
9e547710f5
|
bugfix in the PGF typechecker and more test cases
|
2010-01-29 21:10:14 +00:00 |
|
|
krasimir
|
3685595ece
|
cleanup the code of the PGF interpreter and polish the binary serialization to match the preliminary specification
|
2010-01-27 09:39:14 +00:00 |
|
|
krasimir
|
a039808141
|
PGF is now real synchronous PMCFG
|
2010-01-17 21:35:36 +00:00 |
|
|
krasimir
|
15ddc283d4
|
rename some modules that had GFCC in the name to PGF+something
|
2009-12-14 10:10:58 +00:00 |
|