Creative Commons Attribution 4.0 International (CC-BY 4.0)Roald EiselenRico KoenAlbertus KrugerJacques van Heerden2023-07-282023-05-012023-07-282023-05-012023-05-01https://hdl.handle.net/20.500.12185/617Contextual word/string embeddings for the forward flavour of the FLAIR architecture (Akbik et al., 2018). The embedding provides real-valued vector representations for isiNdebele text.Training data: Paragraphs: 247,926; Token count: 3,633,845; Vocab size: N/A; Embedding dimensions: 2,048;nrModules181.83MB (Zipped)