Improved-xlm-roberta-base
This model is a fine-tuned version of Anwaarma/XLM-roberta-sentiment on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4061
- Accuracy: 0.7
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: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.6928 | 0.55 | 50 | 0.6880 | 0.47 |
0.5811 | 1.1 | 100 | 0.6209 | 0.64 |
0.4842 | 1.65 | 150 | 0.5061 | 0.78 |
0.4212 | 2.2 | 200 | 0.5500 | 0.78 |
0.3867 | 2.75 | 250 | 0.5485 | 0.78 |
0.3556 | 3.3 | 300 | 0.5235 | 0.77 |
0.2987 | 3.85 | 350 | 0.7389 | 0.72 |
0.3037 | 4.4 | 400 | 0.5235 | 0.84 |
0.2443 | 4.95 | 450 | 0.6659 | 0.75 |
0.2283 | 5.49 | 500 | 0.7582 | 0.77 |
0.1989 | 6.04 | 550 | 0.7092 | 0.78 |
0.1753 | 6.59 | 600 | 0.6941 | 0.78 |
0.1558 | 7.14 | 650 | 0.7594 | 0.78 |
0.15 | 7.69 | 700 | 1.0366 | 0.73 |
0.1268 | 8.24 | 750 | 1.1879 | 0.75 |
0.1467 | 8.79 | 800 | 0.8781 | 0.77 |
0.1263 | 9.34 | 850 | 1.1063 | 0.77 |
0.1098 | 9.89 | 900 | 1.4061 | 0.7 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.7
- Tokenizers 0.14.1
- Downloads last month
- 13
Invalid base_model specified in model card
metadata. Needs to be a model id from
hf.co/models.