metadata
library_name: transformers
license: mit
base_model: microsoft/mdeberta-v3-base
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: mdeberta-domain_fold2
results: []
mdeberta-domain_fold2
This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3673
- Accuracy: 0.8759
- Precision: 0.8300
- Recall: 0.8274
- F1: 0.8284
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 1.0329 | 1.0 | 19 | 0.8877 | 0.5931 | 0.8644 | 0.3333 | 0.2482 |
| 0.8422 | 2.0 | 38 | 0.6863 | 0.5931 | 0.8644 | 0.3333 | 0.2482 |
| 0.6622 | 3.0 | 57 | 0.5863 | 0.6552 | 0.7538 | 0.4333 | 0.3988 |
| 0.5772 | 4.0 | 76 | 0.5274 | 0.8 | 0.7533 | 0.6823 | 0.7003 |
| 0.4933 | 5.0 | 95 | 0.4355 | 0.8552 | 0.8229 | 0.7792 | 0.7964 |
| 0.3899 | 6.0 | 114 | 0.3974 | 0.8552 | 0.8143 | 0.7800 | 0.7857 |
| 0.2837 | 7.0 | 133 | 0.3670 | 0.8759 | 0.8428 | 0.8274 | 0.8347 |
| 0.2585 | 8.0 | 152 | 0.3569 | 0.8690 | 0.8232 | 0.8312 | 0.8268 |
| 0.1997 | 9.0 | 171 | 0.3879 | 0.8483 | 0.8155 | 0.8256 | 0.8144 |
| 0.2828 | 10.0 | 190 | 0.3673 | 0.8759 | 0.8300 | 0.8274 | 0.8284 |
Framework versions
- Transformers 4.46.0
- Pytorch 2.3.1
- Datasets 2.21.0
- Tokenizers 0.20.1