Differences between revisions 5 and 11 (spanning 6 versions)
|
Size: 339
Comment:
|
Size: 1184
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 14: | Line 14: |
| == Models == | == Download == You may check out the source code from [[http://git.nlp.ipipan.waw.pl/cothec/Corneferencer|git repository]]. == Models == * [[attachment:model_1147_features.h5|1147 features (word embedding vectors size 50)]] * [[attachment:model_6647_features.h5|6647 features (word embedding vectors size 300)]] * [[attachment:model_1190_features.h5|1190 features (word embedding vectors size 50)]] * [[attachment:weights_siamese_model.h5|Siamese Networks model (word embedding vectors size 50)]] * [[attachment:model_with_singletons.h5|1190 features (5 times more negative examples used for training, word embedding vectors size 50)]] * [[http://mozart.ipipan.waw.pl/~axw/models/lemma/|word embedding vectors]] == Evaluation == * [[attachment:test.zip|test data set (TEI)]] * [[attachment:train.zip|training data set (TEI)]] |
Corneferencer
Corneferencer is a neural network based tool for performing Coreference Resolution (CR).
Principal developer: Bartłomiej Nitoń
Authors: Bartłomiej Nitoń, Paweł Morawiecki
Download
You may check out the source code from git repository.
Models