Tools and Services
Stały URI dla kolekcjihttps://hdl.handle.net/11321/1014
Przeglądaj
Ostatnie zgłoszenia
Item type: Pozycja , Item type: Pozycja , HerBERT Large Pre-trained on KGR10 Data(Wrocław University of Science and Technology, 2021-10-22) Janz, Arkadiusz; Wątorski, PiotrHerBERT-large model fine-tuned on KGR10 data. The model was trained using DeepSpeed technology.Item type: Pozycja , Speech tools plugin for Annotation Pro(ClarinPL, 2021-10-14) Klessa, Katarzyna; Korzinek, DanijelThis resource describes the Annotation Pro plugin containing various tools for automatic processing of speech data. The initial tool provides only a speech aligner, but more are planned in the future.Item type: Pozycja , PoLitBert_v32k_tri_50k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v32k_linear_50k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v32k_linear_125k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v32k_cos1_5_50k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v32k_tri_125k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v50k_linear_50k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , PoLitBert_v32k_cos1_2_50k - Polish RoBERTa model(Ermlab, 2021) Sopyła, Krzysztof; Sawaniewski, ŁukaszPolish RoBERTa model trained on Polish Wikipedia, Polish literature and Oscar.Item type: Pozycja , KGR10-RoBERTa(Wrocław University of Technology, 2020-12-09) Janz, Arkadiusz; Ropiak, NorbertPolish RoBERTa model pre-trained on KGR10 corpora.Item type: Pozycja , XLM-RoBERTa-LARGE events relation recognition(Wrocław University of Science and Technology, 2020-11-12) Kocoń, JanA set of basic language tools for the Polish language. Z4.2a Improving the quality of recognition of relations between events using Transformer-type deep networks.Item type: Pozycja , XLM-RoBERTa event relations recognition(Wroclaw University of Science and Technology, 2020-07-09) Kocoń, JanEvent relations recognition models for the Polish language, based on the XLM-RoBERTa language model.Item type: Pozycja , KPWr n82 NER model (on Polish RoBERTa base)(Wrocław University of Science and Technology, 2020-06-18) Marcińczuk, MichałThe named entity recognition model for fine-grained categories of entities (82 types) was trained on the KPWr corpus using Polish RoBERTa base language model. Details can be found on the following page: https://github.com/mczuk/xlm-roberta-nerItem type: Pozycja , XLM-RoBERTa events recognition(Wrocław University of Science and Technology, 2020-05-22) Kocoń, JanEvent recognition models for the Polish language, based on the XLM-RoBERTa language model.Item type: Pozycja , Smyrna(Daniel Janus, 2019-10-16) Janus, DanielSmyrna is a tool for building and searching own Polish corpora from HTML files.Item type: Pozycja , Liner2 events model(ClarinPL, 2019-07-15) Kocoń, JanLiner2 model for event and event relation recognitionItem type: Pozycja , Liner2 temporal expressions model(ClarinPL, 2019-07-15) Kocoń, JanLiner2 model for temporal expression recognition and normalisationItem type: Pozycja , ELMo Embeddings for Polish(Wroclaw University of Science and Technology, 2019-06-26) Janz, Arkadiusz; Miłkowski, PiotrA model of ELMo embeddings for Polish language trained on large textual corpora (KGR10). To retrain the model please use the checkpoint and vocabulary files available at: https://nextcloud.clarin-pl.eu/index.php/s/EV96cjfflKwt2xHItem type: Pozycja , Tagger SentiOne - version 2(SentiOne, 2019-03-20) Walentynowicz, WiktorThis is the second version of the morpho-syntactic tagger for the Polish language, adapted to UGC-processing. It has been enriched with some heuristics to improve its accuracy and a tokenizer.