hugginglearners/malayalam-blurr-xlm-roberta-base

Blurr x Casual Machine Learning Model trained on Malayalam (മലയാളം) text. (Working in Progress)

---

malayalam-blurr-xlm-roberta-base (base-sized model)

malayalam-blurr-xlm-roberta-base model is pre-trained on xlm-roberta-base using the library blurr Language Model using fastai x huggingface frameworks.

Ref: Causal Language Modeling.

Usage

!pip install -Uqq huggingface_hub["fastai"] ohmeow-blurr

from huggingface_hub import from_pretrained_fastai
learner = from_pretrained_fastai(repo_id)

learner.blurr_generate("ബ്‌ളൂർ പഠിക്കാൻ വളെരെ എളുപ്പമാണ് എന്തുകൊണ്ട് എന്നാൽ", max_length=50, do_sample=True, top_k=25)

Intended uses & limitations

It's not fine tuned to the state of the art accuracy

Training and evaluation data

Wiki 2020 Malayalam Dataset