metadata

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

Dagobert42/mobilebert-uncased-biored-augmented

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

Loss: 0.5602
Accuracy: 0.8057
Precision: 0.6022
Recall: 0.4577
F1: 0.5007
Weighted F1: 0.7931

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	0.6641	0.7644	0.5563	0.3256	0.3768	0.7254
No log	2.0	50	0.6506	0.7728	0.5735	0.369	0.4057	0.7457
No log	3.0	75	0.6352	0.7779	0.5671	0.413	0.4527	0.768
No log	4.0	100	0.6147	0.7886	0.6221	0.4133	0.4523	0.7723
No log	5.0	125	0.6223	0.7885	0.6176	0.4051	0.4416	0.7684
No log	6.0	150	0.6020	0.7913	0.584	0.4496	0.4928	0.7811
No log	7.0	175	0.6081	0.7913	0.5508	0.4964	0.5183	0.7876
No log	8.0	200	0.5935	0.799	0.6333	0.4461	0.4926	0.7826
No log	9.0	225	0.6110	0.7941	0.5998	0.4656	0.5078	0.7825
No log	10.0	250	0.5990	0.798	0.6332	0.4564	0.5099	0.7777

Framework versions

Transformers 4.35.2
Pytorch 2.0.1+cu117
Datasets 2.12.0
Tokenizers 0.15.0