prueba4
This model is a fine-tuned version of PlanTL-GOB-ES/bsc-bio-ehr-es-pharmaconer on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2044
- Precision: 0.7288
- Recall: 0.6853
- F1: 0.7064
- Accuracy: 0.9752
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: 2.75e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 57 | 0.2361 | 0.6504 | 0.6892 | 0.6692 | 0.9694 |
No log | 2.0 | 114 | 0.2441 | 0.6190 | 0.6733 | 0.6450 | 0.9671 |
No log | 3.0 | 171 | 0.2064 | 0.6013 | 0.7211 | 0.6558 | 0.9699 |
No log | 4.0 | 228 | 0.2241 | 0.7004 | 0.6335 | 0.6653 | 0.9720 |
No log | 5.0 | 285 | 0.1992 | 0.6578 | 0.6892 | 0.6732 | 0.9727 |
No log | 6.0 | 342 | 0.2149 | 0.6073 | 0.7331 | 0.6643 | 0.9694 |
No log | 7.0 | 399 | 0.2099 | 0.7466 | 0.6574 | 0.6992 | 0.9755 |
No log | 8.0 | 456 | 0.2039 | 0.7293 | 0.6653 | 0.6958 | 0.9747 |
0.0017 | 9.0 | 513 | 0.2185 | 0.7342 | 0.6494 | 0.6892 | 0.9742 |
0.0017 | 10.0 | 570 | 0.2074 | 0.688 | 0.6853 | 0.6866 | 0.9732 |
0.0017 | 11.0 | 627 | 0.2010 | 0.7073 | 0.6932 | 0.7002 | 0.9745 |
0.0017 | 12.0 | 684 | 0.2030 | 0.7126 | 0.7012 | 0.7068 | 0.9749 |
0.0017 | 13.0 | 741 | 0.2045 | 0.7173 | 0.6773 | 0.6967 | 0.9745 |
0.0017 | 14.0 | 798 | 0.2040 | 0.7185 | 0.6813 | 0.6994 | 0.9747 |
0.0017 | 15.0 | 855 | 0.2044 | 0.7288 | 0.6853 | 0.7064 | 0.9752 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
- Downloads last month
- 10
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.