NCHLT Xitsonga RoBERTa language model
Please do not copy the URL from the browser for citation. The correct URL is 'https://hdl.handle.net/20.500.12185/642'
| dc.contact.email | Roald.Eiselen@nwu.ac.za | |
| dc.contact.name | Roald Eiselen | |
| dc.contributor.author | Roald Eiselen | |
| dc.contributor.other | Rico Koen | |
| dc.contributor.other | Albertus Kruger | |
| dc.contributor.other | Jacques van Heerden | |
| dc.date.accessioned | 2023-07-28T08:11:48Z | |
| dc.date.accessioned | 2023-05-01 | |
| dc.date.available | 2023-07-28T08:11:48Z | |
| dc.date.available | 2023-05-01 | |
| dc.date.issued | 2023-05-01 | |
| dc.description | Contextual masked language model based on the RoBERTa architecture (Liu et al., 2019). The model is trained as a masked language model and not fine-tuned for any downstream process. The model can be used both as a masked LM or as an embedding model to provide real-valued vectorised respresentations of words or string sequences for Xitsonga text. | |
| dc.format.extent | Training data: Paragraphs: 360,698; Token count: 7,357,764; Vocab size: 30,000; Embedding dimensions: 768; | |
| dc.format.size | 235.98MB (Zipped) | |
| dc.identifier.uri | https://hdl.handle.net/20.500.12185/642 | |
| dc.language.iso | ts | |
| dc.languages | Xitsonga | |
| dc.media.category | Language model | |
| dc.media.type | Text | |
| dc.project | NCHLT Text IV | |
| dc.publisher | North-West University; Centre for Text Technology (CTexT) | |
| dc.rights.license | Creative Commons Attribution 4.0 International (CC-BY 4.0) | |
| dc.software.requirements | Python | |
| dc.source | Web | |
| dc.source | Government Documents | |
| dc.title | NCHLT Xitsonga RoBERTa language model | |
| dc.type | Modules |
Files
Original bundle
1 - 1 of 1