distilbert-base-multilingual-cased-v4
This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0888
- Accuracy: 0.7301
- F1: 0.6450
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: 7e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
1.6878 | 1.0 | 29 | 0.8023 | 0.6280 | 0.5411 |
0.4075 | 2.0 | 58 | 0.2881 | 0.7261 | 0.6418 |
0.164 | 3.0 | 87 | 0.1518 | 0.7524 | 0.6743 |
0.09 | 4.0 | 116 | 0.1167 | 0.7584 | 0.6812 |
0.0704 | 5.0 | 145 | 0.1150 | 0.7585 | 0.6813 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.6.0+cu126
- Datasets 3.3.2
- Tokenizers 0.19.1
- Downloads last month
- 20
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.