metadata
license: apache-2.0
base_model: Dr-BERT/DrBERT-7GB
tags:
- generated_from_trainer
datasets:
- quaero
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: drbert-7gb-finedtuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: quaero
type: quaero
config: medline
split: validation
args: medline
metrics:
- name: Precision
type: precision
value: 0.5151515151515151
- name: Recall
type: recall
value: 0.5749888740542947
- name: F1
type: f1
value: 0.5434279705573082
- name: Accuracy
type: accuracy
value: 0.8052467462039046
drbert-7gb-finedtuned-ner
This model is a fine-tuned version of Dr-BERT/DrBERT-7GB on the quaero dataset. It achieves the following results on the evaluation set:
- Loss: 1.0979
- Precision: 0.5152
- Recall: 0.5750
- F1: 0.5434
- Accuracy: 0.8052
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 105 | 0.7366 | 0.4295 | 0.4815 | 0.4540 | 0.7706 |
No log | 2.0 | 210 | 0.7073 | 0.4751 | 0.5256 | 0.4990 | 0.7844 |
No log | 3.0 | 315 | 0.7970 | 0.5277 | 0.5087 | 0.5180 | 0.7960 |
No log | 4.0 | 420 | 0.9072 | 0.4960 | 0.5287 | 0.5118 | 0.7926 |
0.4094 | 5.0 | 525 | 0.9859 | 0.4765 | 0.5696 | 0.5190 | 0.7899 |
0.4094 | 6.0 | 630 | 1.0399 | 0.5093 | 0.5750 | 0.5401 | 0.8039 |
0.4094 | 7.0 | 735 | 1.0778 | 0.5245 | 0.5723 | 0.5474 | 0.8065 |
0.4094 | 8.0 | 840 | 1.0979 | 0.5152 | 0.5750 | 0.5434 | 0.8052 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2