gf-core.git/treebanks, branch master Unnamed repository; edit this file 'description' to name the repository. https://git.adelon.net/gf-core.git/atom?h=master 2015-11-18T10:16:55+00:00 some more conversion patterns for Susanne 2015-11-18T10:16:55+00:00 krasimir krasimir@chalmers.se 2015-11-18T10:16:55+00:00 urn:sha1:fe7397efe4403f94797203e6bf00daa95d749618 a new version of the conversion script for the Susanne corpus which covers a bit less than half of the data 2015-11-13T13:05:21+00:00 krasimir krasimir@chalmers.se 2015-11-13T13:05:21+00:00 urn:sha1:6011fbc033b95ed3e703a61cf1162f4d032ad4ca a refactoring in the Susanne converter which for some reason was not pushed before 2014-10-10T12:10:44+00:00 kr.angelov kr.angelov@gmail.com 2014-10-10T12:10:44+00:00 urn:sha1:2b881397c554d9aa17877c9686bf3bbfd6f4020f partial reconstruction for sentence structures in Susanne 2013-12-06T15:05:56+00:00 kr.angelov kr.angelov@gmail.com 2013-12-06T15:05:56+00:00 urn:sha1:e2fe50e5859cb6ef359c1a08e3bceb3080cd2159 more on the Susanne treebank 2013-12-05T10:05:33+00:00 kr.angelov kr.angelov@gmail.com 2013-12-05T10:05:33+00:00 urn:sha1:05854280181f5ad30939a4fc533d42560103d23f added script for reading the Susanne treebank 2013-12-04T12:11:41+00:00 kr.angelov kr.angelov@gmail.com 2013-12-04T12:11:41+00:00 urn:sha1:106b41a2cb562b26a477af5372ddf09fe8f79ca7 the content of ParseEngAbs3.probs is now merged with ParseEngAbs.probs. The later is now retrained. Once the grammar is compiled with the .probs file now it doesn't need anything more to do robust parsing. The robustness itself is controlled by the flags 'heuristic_search_factor', 'meta_prob' and 'meta_token_prob' in ParseEngAbs.gf 2013-11-06T10:21:46+00:00 kr.angelov kr.angelov@gmail.com 2013-11-06T10:21:46+00:00 urn:sha1:2483dc772897eb0909664f1a88cc7f8ec50ebd5b work on the translation dictionaries and the statistical model 2013-09-20T12:09:54+00:00 kr.angelov kr.angelov@gmail.com 2013-09-20T12:09:54+00:00 urn:sha1:9e148280c5afbb06802a258d3c47108195785eea added simple script for estimating the coverage on the PennTreebank 2013-03-28T09:15:38+00:00 kr.angelov kr.angelov@gmail.com 2013-03-28T09:15:38+00:00 urn:sha1:74a16273b99632caf94912a4a4510d17cc88abac added a few more multiword expressions in DictEng and a few words in the abstract syntax are not tagged with their senses. There is a new statistical model too 2013-03-27T20:46:42+00:00 kr.angelov kr.angelov@gmail.com 2013-03-27T20:46:42+00:00 urn:sha1:17fc938c20db58e65b005320db39c91f210fabac