metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- source_data
metrics:
- precision
- recall
- f1
model-index:
- name: SourceData_NER_v_1-0-0_BioLinkBERT_large
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: source_data
type: source_data
args: NER
metrics:
- name: Precision
type: precision
value: 0.821286546432172
- name: Recall
type: recall
value: 0.8565732978940912
- name: F1
type: f1
value: 0.8385588676817499
SourceData_NER_v_1-0-0_BioLinkBERT_large
This model is a fine-tuned version of michiyasunaga/BioLinkBERT-large on the source_data dataset. It achieves the following results on the evaluation set:
- Loss: 0.1311
- Accuracy Score: 0.9583
- Precision: 0.8213
- Recall: 0.8566
- F1: 0.8386
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: 32
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adafactor
- lr_scheduler_type: linear
- num_epochs: 2.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy Score | Precision | Recall | F1 |
---|---|---|---|---|---|---|---|
0.1058 | 1.0 | 863 | 0.1321 | 0.9554 | 0.7953 | 0.8670 | 0.8296 |
0.0737 | 2.0 | 1726 | 0.1311 | 0.9583 | 0.8213 | 0.8566 | 0.8386 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0a0+bfe5ad2
- Datasets 2.10.1
- Tokenizers 0.12.1