metadata
license: mit
base_model: roberta-base
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: roberta-finetuned-domains
results: []
roberta-finetuned-domains
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5388
- F1: 0.2879
- Roc Auc: 0.5757
- Accuracy: 0.2533
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: 12
- eval_batch_size: 12
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 768
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.3789 | 0.97 | 31 | 0.4545 | 0.0004 | 0.4997 | 0.0002 |
| 0.2729 | 1.97 | 63 | 0.4684 | 0.2193 | 0.5479 | 0.1668 |
| 0.1905 | 2.98 | 95 | 0.5180 | 0.2491 | 0.5566 | 0.2100 |
| 0.1568 | 3.98 | 127 | 0.5263 | 0.2633 | 0.5636 | 0.2236 |
| 0.1398 | 4.98 | 159 | 0.5298 | 0.2862 | 0.5751 | 0.2492 |
| 0.131 | 5.83 | 186 | 0.5388 | 0.2879 | 0.5757 | 0.2533 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3