AraBERT_token_classification_AraEval24_trunc_rand_single_fixed

This model is a fine-tuned version of aubmindlab/bert-base-arabertv2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 5.1581
• Precision: 0.0264
• Recall: 0.0986
• F1: 0.0416
• Accuracy: 0.1239

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
1.2132	1.0	2469	3.1465	0.0195	0.0532	0.0285	0.1276
0.9749	2.0	4938	3.1130	0.0289	0.1198	0.0465	0.1600
0.7693	3.0	7407	3.4202	0.0276	0.1106	0.0442	0.1504
0.602	4.0	9876	3.8632	0.0308	0.0844	0.0451	0.1193
0.4902	5.0	12345	4.3511	0.0290	0.0805	0.0426	0.1166
0.3857	6.0	14814	4.3332	0.0352	0.1166	0.0540	0.1253
0.3151	7.0	17283	4.7986	0.0225	0.0737	0.0345	0.1135
0.2764	8.0	19752	4.9586	0.0266	0.1035	0.0424	0.1275
0.2238	9.0	22221	5.0601	0.0266	0.0939	0.0415	0.1202
0.1967	10.0	24690	5.1581	0.0264	0.0986	0.0416	0.1239

Framework versions

• Transformers 4.30.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.13.3