metadata

tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: AraBERT_token_classification__AraEval24_merged_rassd
    results: []

AraBERT_token_classification__AraEval24_merged_rassd

This model is a fine-tuned version of aubmindlab/bert-base-arabert on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.8706
Precision: 0.0867
Recall: 0.0203
F1: 0.0329
Accuracy: 0.8593

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
0.6887	1.0	2878	0.7951	0.0	0.0	0.0	0.8632
0.6215	2.0	5756	0.7865	0.0667	0.0010	0.0020	0.8635
0.5597	3.0	8634	0.7852	0.0901	0.0025	0.0048	0.8634
0.5221	4.0	11512	0.7851	0.1001	0.0115	0.0206	0.8622
0.4521	5.0	14390	0.7992	0.0772	0.0063	0.0117	0.8627
0.4411	6.0	17268	0.8035	0.0873	0.0084	0.0154	0.8625
0.4185	7.0	20146	0.8330	0.0714	0.0092	0.0162	0.8619
0.3954	8.0	23024	0.8511	0.0943	0.0158	0.0271	0.8619
0.3688	9.0	25902	0.8527	0.0936	0.0158	0.0271	0.8608
0.3575	10.0	28780	0.8706	0.0867	0.0203	0.0329	0.8593

Framework versions

Transformers 4.30.2
Pytorch 1.12.1
Datasets 2.13.2
Tokenizers 0.13.3