metadata
library_name: transformers
license: mit
base_model: roberta-large
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: wikidyk-scope-clf-deberta-v3-large-semantic_3_clusters
results: []
wikidyk-scope-clf-deberta-v3-large-semantic_3_clusters
This model is a fine-tuned version of roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5560
- Accuracy: 0.9132
- F1: 0.7401
- Precision: 0.8592
- Recall: 0.6501
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
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 128
- total_eval_batch_size: 128
- optimizer: Use 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: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.0494 | 1.0 | 902 | 0.3398 | 0.8673 | 0.5190 | 0.8346 | 0.3766 |
| 0.0428 | 2.0 | 1804 | 0.3267 | 0.8902 | 0.6328 | 0.8696 | 0.4973 |
| 0.0292 | 3.0 | 2706 | 0.3243 | 0.9071 | 0.7333 | 0.8077 | 0.6714 |
| 0.0208 | 4.0 | 3608 | 0.4366 | 0.8926 | 0.6603 | 0.8284 | 0.5488 |
| 0.0096 | 5.0 | 4510 | 0.4438 | 0.8943 | 0.6829 | 0.7948 | 0.5986 |
| 0.0182 | 6.0 | 5412 | 0.5225 | 0.8950 | 0.6594 | 0.86 | 0.5346 |
| 0.0062 | 7.0 | 6314 | 0.4717 | 0.9078 | 0.7117 | 0.8776 | 0.5986 |
| 0.0021 | 8.0 | 7216 | 0.5226 | 0.9095 | 0.7254 | 0.8571 | 0.6288 |
| 0.0026 | 9.0 | 8118 | 0.5323 | 0.9173 | 0.7528 | 0.8715 | 0.6625 |
| 0.0008 | 10.0 | 9020 | 0.5560 | 0.9132 | 0.7401 | 0.8592 | 0.6501 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1