output
This model is a fine-tuned version of PlanTL-GOB-ES/bsc-bio-ehr-es on the Rodrigo1771/multi-dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.8606
- Precision: 0.4151
- Recall: 0.8521
- F1: 0.5583
- Accuracy: 0.8647
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10.0
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.2596 | 0.9997 | 1701 | 0.3842 | 0.3410 | 0.8170 | 0.4811 | 0.8444 |
0.1853 | 2.0 | 3403 | 0.3449 | 0.4112 | 0.8165 | 0.5469 | 0.8731 |
0.1254 | 2.9997 | 5104 | 0.5721 | 0.3961 | 0.8359 | 0.5375 | 0.8522 |
0.0823 | 4.0 | 6806 | 0.6648 | 0.3852 | 0.8364 | 0.5275 | 0.8523 |
0.0597 | 4.9997 | 8507 | 0.7122 | 0.4074 | 0.8238 | 0.5452 | 0.8582 |
0.0446 | 6.0 | 10209 | 0.7774 | 0.4082 | 0.8335 | 0.5480 | 0.8572 |
0.0325 | 6.9997 | 11910 | 0.8606 | 0.4151 | 0.8521 | 0.5583 | 0.8647 |
0.022 | 8.0 | 13612 | 0.9536 | 0.4056 | 0.8493 | 0.5490 | 0.8598 |
0.017 | 8.9997 | 15313 | 1.0069 | 0.4062 | 0.8454 | 0.5488 | 0.8610 |
0.0135 | 9.9971 | 17010 | 1.0186 | 0.4072 | 0.8488 | 0.5504 | 0.8615 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 1
Finetuned from
Evaluation results
- Precision on Rodrigo1771/multi-datasetvalidation set self-reported0.415
- Recall on Rodrigo1771/multi-datasetvalidation set self-reported0.852
- F1 on Rodrigo1771/multi-datasetvalidation set self-reported0.558
- Accuracy on Rodrigo1771/multi-datasetvalidation set self-reported0.865