HusseinEid
/

bert-finetuned-ner

Token Classification

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

Edit model card

bert-finetuned-ner

This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:

Loss: 0.0636
Precision: 0.9343
Recall: 0.9498
F1: 0.9420
Accuracy: 0.9861

Model description

This is a model for Named entity recognition NER

Intended uses & limitations

Open source

Training and evaluation data

The conll2003 dataset

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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0757	1.0	1756	0.0638	0.9215	0.9362	0.9288	0.9833
0.0352	2.0	3512	0.0667	0.9360	0.9482	0.9421	0.9858
0.0215	3.0	5268	0.0636	0.9343	0.9498	0.9420	0.9861

Framework versions

Transformers 4.40.2
Pytorch 2.2.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1

Downloads last month: 11

Safetensors

Model size

108M params

Tensor type

F32

·

Inference Examples

Token Classification

This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for HusseinEid/bert-finetuned-ner

Base model

google-bert/bert-base-cased

Finetuned

(1939)

this model

Dataset used to train HusseinEid/bert-finetuned-ner

Evaluation results

Precision on conll2003
validation set self-reported

0.934
Recall on conll2003
validation set self-reported

0.950
F1 on conll2003
validation set self-reported

0.942
Accuracy on conll2003
validation set self-reported

0.986

View on Papers With Code