---
language:
- mn
base_model: bayartsogt/mongolian-roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: roberta-base-ner-demo-turshilt1
  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-base-ner-demo-turshilt1

This model is a fine-tuned version of [bayartsogt/mongolian-roberta-base](https://huggingface.co/bayartsogt/mongolian-roberta-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1003
- Precision: 0.8136
- Recall: 0.8747
- F1: 0.8430
- Accuracy: 0.9695

## Model description

More information needed

## 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: 3e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.8083        | 0.9958  | 119  | 0.2969          | 0.4887    | 0.5017 | 0.4951 | 0.9058   |
| 0.192         | 2.0     | 239  | 0.1250          | 0.7392    | 0.8056 | 0.7709 | 0.9568   |
| 0.1209        | 2.9958  | 358  | 0.1057          | 0.7715    | 0.8421 | 0.8052 | 0.9632   |
| 0.0955        | 4.0     | 478  | 0.0979          | 0.7868    | 0.8558 | 0.8198 | 0.9655   |
| 0.0801        | 4.9958  | 597  | 0.0949          | 0.7889    | 0.8633 | 0.8244 | 0.9669   |
| 0.0692        | 6.0     | 717  | 0.0951          | 0.8031    | 0.8662 | 0.8335 | 0.9679   |
| 0.0617        | 6.9958  | 836  | 0.0985          | 0.8037    | 0.8671 | 0.8342 | 0.9674   |
| 0.0547        | 8.0     | 956  | 0.0992          | 0.8072    | 0.8696 | 0.8373 | 0.9680   |
| 0.0512        | 8.9958  | 1075 | 0.0968          | 0.8070    | 0.8727 | 0.8385 | 0.9689   |
| 0.0458        | 10.0    | 1195 | 0.0976          | 0.8082    | 0.8748 | 0.8402 | 0.9691   |
| 0.0437        | 10.9958 | 1314 | 0.0980          | 0.8163    | 0.8753 | 0.8447 | 0.9694   |
| 0.0414        | 12.0    | 1434 | 0.0982          | 0.8146    | 0.8755 | 0.8439 | 0.9694   |
| 0.0404        | 12.9958 | 1553 | 0.1002          | 0.8145    | 0.8745 | 0.8434 | 0.9695   |
| 0.0382        | 14.0    | 1673 | 0.1010          | 0.8114    | 0.8735 | 0.8414 | 0.9692   |
| 0.0385        | 14.9372 | 1785 | 0.1003          | 0.8136    | 0.8747 | 0.8430 | 0.9695   |


### Framework versions

- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1