A RoBERTa model for Masked Language Modeling (MLM). The base RoBERTa model is initialized from https://huggingface.co/HooshvareLab/roberta-fa-zwnj-base.
Usage
pip install hezar
from hezar import Model
model = Model.load("hezarai/roberta-fa-mlm")
inputs = ["ایشالا از <mask> شروع میکنم"]
outputs = model.predict(inputs)
print(outputs)
{'filled_texts': ['ایشالا از اینجا شروع میکنم'], 'filled_tokens': [' اینجا']}