---
language: 
  - tg
widget:
- text: "Пойтахти <mask> Душанбе"
- text: "<mask> ба ин сайти шумо медароям."
- text: "Номи ман Акрам <mask>"

tags:
- generated_from_trainer
model-index:
- name: RoBERTa-tg
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# RoBERTa-tg

This model is a fine-tuned version of [Tajik-Corpus](https://huggingface.co/datasets/muhtasham/tajik-corpus) dataset which is based on Leipzig Corpora.


## Model description

You can use model for masked text generation or fine-tune it to a downstream task.

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 128
- eval_batch_size: 8
- seed: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1

### Training results



### Framework versions

- Transformers 4.20.0.dev0
- Pytorch 1.11.0+cu113
- Tokenizers 0.12.1