finetuned_Roberta_newcode_5epoch-f1score
This model is a fine-tuned version of Youssef320/Reberta-emoji-finetuned-50label on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.7389
- Top 1 Macro F1 Score: 0.1910
- Top 1 Weighted F1score: 0.2434
- Top 3 Macro F1 Score: 0.3680
- Top3 3 Weighted F1 Score : 0.4515
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: 0.0002
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 2048
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Top 1 Macro F1 Score | Top 1 Weighted F1score | Top 3 Macro F1 Score | Top3 3 Weighted F1 Score |
|---|---|---|---|---|---|---|---|
| 3.0337 | 0.14 | 64 | 2.8439 | 0.1703 | 0.2206 | 0.3484 | 0.4290 |
| 2.9343 | 0.27 | 128 | 2.7976 | 0.1792 | 0.2333 | 0.3610 | 0.4415 |
| 2.8978 | 0.41 | 192 | 2.7960 | 0.1830 | 0.2353 | 0.3638 | 0.4416 |
| 2.8719 | 0.54 | 256 | 2.7718 | 0.1847 | 0.2376 | 0.3631 | 0.4456 |
| 2.8862 | 0.68 | 320 | 2.7410 | 0.1844 | 0.2363 | 0.3659 | 0.4496 |
| 2.8835 | 0.81 | 384 | 2.7556 | 0.1830 | 0.2372 | 0.3644 | 0.4484 |
| 2.8682 | 0.95 | 448 | 2.7389 | 0.1910 | 0.2434 | 0.3680 | 0.4515 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.1+cu102
- Datasets 2.0.0
- Tokenizers 0.11.0