distilroberta-base-finetuned-3d-sentiment
This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7236
- Accuracy: 0.7476
- Precision: 0.7515
- Recall: 0.7476
- F1: 0.7474
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: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 6381
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.7918 | 1.0 | 1595 | 0.7835 | 0.6718 | 0.6877 | 0.6718 | 0.6697 |
| 0.6103 | 2.0 | 3190 | 0.7777 | 0.6923 | 0.7151 | 0.6923 | 0.6917 |
| 0.5534 | 3.0 | 4785 | 0.6858 | 0.7132 | 0.7250 | 0.7132 | 0.7108 |
| 0.4998 | 4.0 | 6380 | 0.6715 | 0.7333 | 0.7398 | 0.7333 | 0.7325 |
| 0.4327 | 5.0 | 7975 | 0.6745 | 0.7421 | 0.7463 | 0.7421 | 0.7420 |
| 0.3534 | 6.0 | 9570 | 0.7236 | 0.7476 | 0.7515 | 0.7476 | 0.7474 |
| 0.2926 | 7.0 | 11165 | 0.7916 | 0.7456 | 0.7510 | 0.7456 | 0.7457 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu117
- Datasets 2.10.1
- Tokenizers 0.13.3