metadata
language:
- mn
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
base_model: cardiffnlp/twitter-roberta-base-sentiment
model-index:
- name: mongolian-twitter-roberta-base-sentiment-ner
results: []
mongolian-twitter-roberta-base-sentiment-ner
This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-sentiment on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1674
- Precision: 0.7560
- Recall: 0.8395
- F1: 0.7955
- Accuracy: 0.9540
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.4091 | 1.0 | 477 | 0.2507 | 0.5166 | 0.6789 | 0.5868 | 0.9162 |
| 0.2467 | 2.0 | 954 | 0.2363 | 0.6415 | 0.7465 | 0.6900 | 0.9243 |
| 0.2051 | 3.0 | 1431 | 0.1921 | 0.6732 | 0.7857 | 0.7251 | 0.9374 |
| 0.1738 | 4.0 | 1908 | 0.1746 | 0.6965 | 0.8038 | 0.7463 | 0.9440 |
| 0.1475 | 5.0 | 2385 | 0.1680 | 0.7217 | 0.8172 | 0.7665 | 0.9472 |
| 0.1305 | 6.0 | 2862 | 0.1736 | 0.7209 | 0.8228 | 0.7685 | 0.9483 |
| 0.1116 | 7.0 | 3339 | 0.1621 | 0.7337 | 0.8296 | 0.7787 | 0.9518 |
| 0.099 | 8.0 | 3816 | 0.1684 | 0.7353 | 0.8318 | 0.7806 | 0.9508 |
| 0.0882 | 9.0 | 4293 | 0.1666 | 0.7625 | 0.8417 | 0.8002 | 0.9547 |
| 0.0799 | 10.0 | 4770 | 0.1674 | 0.7560 | 0.8395 | 0.7955 | 0.9540 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3