This model is a fine-tuned version of om-ashish-soni/pos-ner-tagging on the conll2003 dataset. It achieves the following results on the evaluation set:

Loss: 0.4015
Precision: 0.8992
Recall: 0.8762
F1: 0.8876
Accuracy: 0.9021

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6746	1.0	1756	0.6405	0.7932	0.7137	0.7513	0.8002
0.5604	2.0	3512	0.5519	0.8312	0.7655	0.7970	0.8333
0.482	3.0	5268	0.4807	0.8555	0.8066	0.8303	0.8595
0.3909	4.0	7024	0.4579	0.8635	0.8201	0.8413	0.8672
0.3353	5.0	8780	0.4545	0.8648	0.8245	0.8442	0.8695
0.2641	6.0	10536	0.4518	0.8725	0.8367	0.8542	0.8773
0.2315	7.0	12292	0.4082	0.8924	0.8639	0.8779	0.8951
0.215	8.0	14048	0.4131	0.8932	0.8666	0.8797	0.8956
0.1662	9.0	15804	0.4064	0.8951	0.8719	0.8833	0.8986
0.1532	10.0	17560	0.4015	0.8992	0.8762	0.8876	0.9021

Framework versions

Transformers 4.33.1
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3