metadata

language:
  - en
license: mit
base_model: distilbert-base-uncased
tags:
  - low-resource NER
  - token_classification
  - biomedicine
  - medical NER
  - generated_from_trainer
datasets:
  - medicine
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: Dagobert42/distilbert-base-uncased-biored-augmented
    results: []

Dagobert42/distilbert-base-uncased-biored-augmented

This model is a fine-tuned version of distilbert-base-uncased on the bigbio/biored dataset. It achieves the following results on the evaluation set:

Loss: 0.5141
Accuracy: 0.8189
Precision: 0.6146
Recall: 0.5864
F1: 0.5983
Weighted F1: 0.8169

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: 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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Weighted F1
No log	1.0	25	0.5409	0.807	0.6881	0.5326	0.5615	0.7971
No log	2.0	50	0.5368	0.8108	0.7021	0.5447	0.5781	0.8012
No log	3.0	75	0.5383	0.8161	0.6921	0.5484	0.5835	0.8057
No log	4.0	100	0.5349	0.8131	0.6408	0.5885	0.6008	0.8103
No log	5.0	125	0.5436	0.8157	0.6275	0.606	0.6097	0.8136
No log	6.0	150	0.5488	0.8201	0.6805	0.5826	0.6043	0.8146

Framework versions

Transformers 4.35.2
Pytorch 2.0.1+cu117
Datasets 2.12.0
Tokenizers 0.15.0