metadata
library_name: transformers
license: mit
base_model: microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract
tags:
- generated_from_trainer
datasets:
- source_data
metrics:
- precision
- recall
- f1
model-index:
- name: SourceData_GeneprodRoles_v1_0_0_PubMedBERT_base
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: source_data
type: source_data
config: ROLES_GP
split: validation
args: ROLES_GP
metrics:
- name: Precision
type: precision
value: 0.9208418258520169
- name: Recall
type: recall
value: 0.9266396374910565
- name: F1
type: f1
value: 0.9237316342541962
SourceData_GeneprodRoles_v1_0_0_PubMedBERT_base
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract on the source_data dataset. It achieves the following results on the evaluation set:
- Loss: 0.0135
- Accuracy Score: 0.9948
- Precision: 0.9208
- Recall: 0.9266
- F1: 0.9237
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 128
- seed: 42
- optimizer: Use adafactor and the args are: No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 1.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy Score | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.0146 | 1.0 | 864 | 0.0135 | 0.9948 | 0.9208 | 0.9266 | 0.9237 |
Framework versions
- Transformers 4.46.3
- Pytorch 1.13.1+cu117
- Datasets 3.1.0
- Tokenizers 0.20.3