NER-finetuning-BERT-UNCASED-BIOBERT

This model is a fine-tuned version of google-bert/bert-base-uncased on the biobert_json dataset. It achieves the following results on the evaluation set:

Loss: 0.1163
Precision: 0.9432
Recall: 0.9668
F1: 0.9548
Accuracy: 0.9765

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.4431	1.0	612	0.1173	0.9250	0.9596	0.9420	0.9709
0.139	2.0	1224	0.1097	0.9276	0.9724	0.9495	0.9728
0.0933	3.0	1836	0.0957	0.9451	0.9686	0.9567	0.9776
0.0751	4.0	2448	0.0972	0.9392	0.9733	0.9559	0.9771
0.0536	5.0	3060	0.0978	0.9445	0.9705	0.9573	0.9770
0.0468	6.0	3672	0.1044	0.9427	0.9661	0.9543	0.9766
0.0392	7.0	4284	0.1080	0.9396	0.9691	0.9541	0.9765
0.0376	8.0	4896	0.1151	0.9390	0.9696	0.9540	0.9761
0.0293	9.0	5508	0.1128	0.9429	0.9674	0.9550	0.9766
0.0274	10.0	6120	0.1163	0.9432	0.9668	0.9548	0.9765

Framework versions

Transformers 4.45.1
Pytorch 2.4.0
Datasets 3.0.1
Tokenizers 0.20.0

Loading…

Criser2013/NER-finetuning-BERT-UNCASED-BIOBERT

NER-finetuning-BERT-UNCASED-BIOBERT

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NER-finetuning-BERT-UNCASED-BIOBERT

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions