metadata
license: mit
base_model: microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext
tags:
- generated_from_trainer
datasets:
- ncbi_disease
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: checkpoint-1000
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: ncbi_disease
type: ncbi_disease
config: ncbi_disease
split: test
args: ncbi_disease
metrics:
- name: Precision
type: precision
value: 0.8456973293768546
- name: Recall
type: recall
value: 0.890625
- name: F1
type: f1
value: 0.8675799086757991
- name: Accuracy
type: accuracy
value: 0.9850593950279626
checkpoint-1000
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the ncbi_disease dataset. It achieves the following results on the evaluation set:
- Loss: 0.0543
- Precision: 0.8457
- Recall: 0.8906
- F1: 0.8676
- Accuracy: 0.9851
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 340 | 0.0596 | 0.7778 | 0.875 | 0.8235 | 0.9795 |
| 0.0787 | 2.0 | 680 | 0.0416 | 0.8246 | 0.8865 | 0.8544 | 0.9851 |
| 0.0202 | 3.0 | 1020 | 0.0494 | 0.8385 | 0.8812 | 0.8593 | 0.9846 |
| 0.0202 | 4.0 | 1360 | 0.0543 | 0.8457 | 0.8906 | 0.8676 | 0.9851 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0