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: 1.3454
Macro F1: 0.7875
Precision: 0.7834
Recall: 0.7949
Kappa: 0.6913
Accuracy: 0.7949

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
1.1521	1.0	697	0.7730	0.7668	0.7492	0.7976	0.6831	0.7976
0.7985	2.0	1395	0.7075	0.7817	0.7674	0.8003	0.6965	0.8003
0.6333	3.0	2092	0.7101	0.7840	0.7793	0.8078	0.7023	0.8078
0.5307	4.0	2790	0.7471	0.7797	0.7779	0.7989	0.6929	0.7989
0.447	5.0	3487	0.7967	0.7826	0.7765	0.7951	0.6916	0.7951
0.304	6.0	4185	0.8912	0.7884	0.7836	0.7976	0.6961	0.7976
0.2597	7.0	4882	0.9286	0.7872	0.7820	0.7962	0.6925	0.7962
0.1859	8.0	5580	1.0321	0.7887	0.7845	0.7996	0.6963	0.7996
0.1542	9.0	6277	1.0918	0.7840	0.7801	0.7926	0.6879	0.7926
0.135	10.0	6975	1.1611	0.7884	0.7825	0.8035	0.6988	0.8035
0.0894	11.0	7672	1.2353	0.7866	0.7862	0.7911	0.6871	0.7911
0.084	12.0	8370	1.2618	0.7875	0.7832	0.7965	0.6920	0.7965
0.0595	13.0	9067	1.3147	0.7847	0.7836	0.7879	0.6844	0.7879
0.0472	14.0	9765	1.3424	0.7872	0.7839	0.7942	0.6897	0.7942
0.0422	14.99	10455	1.3454	0.7875	0.7834	0.7949	0.6913	0.7949

Framework versions

Transformers 4.36.0
Pytorch 2.1.0+cu118
Tokenizers 0.15.0