metadata

base_model: aubmindlab/bert-base-arabertv02
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: Type_of_relation
    results: []

Type_of_relation

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.7828
Macro F1: 0.9002
Precision: 0.8999
Recall: 0.9007
Kappa: 0.8082
Accuracy: 0.9007

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: 128
seed: 25
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Precision	Recall	Kappa	Accuracy
0.492	1.0	697	0.3226	0.8904	0.8920	0.8923	0.7872	0.8923
0.3283	2.0	1395	0.3107	0.9000	0.9001	0.9010	0.8071	0.9010
0.2371	3.0	2092	0.3224	0.8959	0.8973	0.8949	0.8020	0.8949
0.1817	4.0	2790	0.3469	0.9000	0.9007	0.9000	0.8082	0.9000
0.1372	5.0	3487	0.4185	0.8966	0.8980	0.8962	0.8034	0.8962
0.0779	6.0	4185	0.4717	0.8989	0.8991	0.8992	0.8059	0.8992
0.0676	7.0	4882	0.5415	0.8962	0.8972	0.8958	0.8014	0.8958
0.042	8.0	5580	0.6031	0.8984	0.8988	0.8982	0.8057	0.8982
0.0335	9.0	6277	0.6551	0.9017	0.9019	0.9025	0.8106	0.9025
0.0239	10.0	6975	0.7116	0.8975	0.8974	0.8980	0.8029	0.8980
0.0168	11.0	7672	0.7130	0.8976	0.8976	0.8984	0.8029	0.8984
0.019	12.0	8370	0.7464	0.9011	0.9007	0.9018	0.8101	0.9018
0.0121	13.0	9067	0.7709	0.9006	0.9003	0.9010	0.8091	0.9010
0.0089	14.0	9765	0.7790	0.9002	0.9002	0.9007	0.8081	0.9007
0.0068	14.99	10455	0.7828	0.9002	0.8999	0.9007	0.8082	0.9007

Framework versions

Transformers 4.36.0
Pytorch 2.1.0+cu118
Tokenizers 0.15.0