metadata
license: mit
base_model: microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: PubMedBERT_BioNLP13CG_NER
results: []
PubMedBERT_BioNLP13CG_NER
This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2094
- Precision: 0.8591
- Recall: 0.8404
- F1: 0.8497
- Accuracy: 0.9510
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 0.99 | 95 | 0.3540 | 0.7748 | 0.7481 | 0.7612 | 0.9170 |
| No log | 2.0 | 191 | 0.2264 | 0.8430 | 0.8356 | 0.8393 | 0.9467 |
| No log | 2.98 | 285 | 0.2094 | 0.8591 | 0.8404 | 0.8497 | 0.9510 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0