/SpanishTransformerXL

Language model trained on wiki corpus (500M tokens) with fastai v1 acc>42.3% len(vocab)=60K

Primary LanguageJupyter Notebook

Issues