Age | Commit message (Expand) | Author |
2013-01-08 | add header files to sources to create correct distributions | Chris Dyer |
2012-12-16 | add grammar convert back in | Chris Dyer |
2012-12-15 | enable kenlm compression | Chris Dyer |
2012-11-20 | Merge branch 'master' of github.com:redpony/cdec | Chris Dyer |
2012-11-20 | fixes for 2011 optimizer | Chris Dyer |
2012-11-19 | merge upstream/master | Patrick Simianer |
2012-11-19 | merge upstream/master | Patrick Simianer |
2012-11-18 | forgotten makefile | Chris Dyer |
2012-11-18 | major restructure of the training code | Chris Dyer |
2012-10-22 | Update search, make it compile | Kenneth Heafield |
2012-10-16 | c++11 compatibility | Chris Dyer |
2012-10-11 | Merge branch 'master' of https://github.com/redpony/cdec | Chris Dyer |
2012-10-11 | produce alignments for a test set (which can be stdin) | Chris Dyer |
2012-10-02 | cdec cleanup, remove bayesian stuff, parsing stuff | Chris Dyer |
2012-09-20 | rename model1 to fast_aligner, since it does way more than model1 but is most... | Chris Dyer |
2012-08-13 | another fix | Chris Dyer |
2012-07-29 | fix grammar converter to remove edges that cannot exist in any valid derivation | Chris Dyer |
2012-07-02 | [python] Explicit candidate set; memory leak fixes | Victor Chahuneau |
2012-06-29 | add option for entropy optimization | Chris Dyer |
2012-06-24 | minimum risk training, not completely ready for primetime | Chris Dyer |
2012-06-24 | compute risk / gradient of risk | Chris Dyer |
2012-06-22 | single processor fix | Chris Dyer |
2012-06-19 | compute held-out ppl in mpi_batch_optimize | Chris Dyer |
2012-06-18 | add non-const iterators to sparse vector, speed up model1 code | Chris Dyer |
2012-06-01 | fix potential build break | Chris Dyer |
2012-05-29 | add support to rampion for accumulating k-best lists | Chris Dyer |
2012-05-28 | cache metric computation in pro | Chris Dyer |
2012-05-27 | fix mapper to use common candidate set code | Chris Dyer |
2012-05-27 | kbest repo | Chris Dyer |
2012-05-20 | Rename decoder library, assimilate mira directory | Kenneth Heafield |
2012-05-17 | Training jamfile | Chris Dyer |
2012-05-17 | jamfiles for creg | Chris Dyer |
2012-05-13 | put creg in its own top-level folder | Chris Dyer |
2012-05-13 | fast creg training code for univariate linear and logistic regression | Chris Dyer |
2012-05-08 | better logging | Chris Dyer |
2012-05-08 | switch to liblbfgs for pro | Chris Dyer |
2012-05-08 | add liblbfgs, which is much less crappy than the current one | Chris Dyer |
2012-03-20 | make c++11 compatible | Chris Dyer |
2012-03-07 | more mpi fixes | Chris Dyer |
2012-03-07 | better logging with MPI | Chris Dyer |
2012-02-29 | mpi fixes | Chris Dyer |
2012-02-29 | cleanup, mpi-ify lblmodel | Chris Dyer |
2012-02-24 | load embeddings from file | Chris Dyer |
2012-02-22 | add regularization | Chris Dyer |
2012-02-21 | use lbfgs | Chris Dyer |
2012-02-21 | basic lbl model, nothing to see here | Chris Dyer |
2012-02-19 | lbl preliminary clean up | Chris Dyer |
2012-02-18 | initial lbl_model stub | Chris Dyer |
2012-02-18 | fix diagonal model | Chris Dyer |
2012-02-17 | boost version checking, check for Eigen, get rid of old digamma stuff | Chris Dyer |