restar_v1.0_model
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5218
- Accuracy: 0.8348
- F1 Macro: 0.6848
- Confusion Matrix: [[82465, 26863, 3426], [18567, 49076, 11286], [15442, 82987, 669888]]
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: 3e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- 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: 0.06
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Confusion Matrix |
|---|---|---|---|---|---|---|
| 0.2499 | 1.0 | 4688 | 0.6122 | 0.8257 | 0.6741 | [[24622, 12042, 1110], [4536, 17557, 3968], [4031, 30080, 222054]] |
| 0.3252 | 2.0 | 9376 | 0.5233 | 0.8339 | 0.6838 | [[27696, 8969, 1109], [6104, 16171, 3786], [5149, 28027, 222989]] |
| 0.2379 | 3.0 | 14064 | 0.6115 | 0.8312 | 0.6797 | [[27286, 9322, 1166], [5976, 16152, 3933], [5360, 28268, 222537]] |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2
- Downloads last month
- 7