distilbert
This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1273
- Accuracy: 0.969
- F1: 0.9689
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
0.4515 | 1.0 | 141 | 0.1685 | 0.9215 | 0.9242 |
0.1561 | 2.0 | 282 | 0.1402 | 0.955 | 0.9532 |
0.0658 | 3.0 | 423 | 0.1033 | 0.9645 | 0.9641 |
0.0475 | 4.0 | 564 | 0.1081 | 0.9685 | 0.9683 |
0.0167 | 5.0 | 705 | 0.1273 | 0.969 | 0.9689 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 2.14.4
- Tokenizers 0.21.1
- Downloads last month
- 3
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support