arabert_baseline_relevance_task2_fold1

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1509
• Qwk: 0.1747
• Mse: 0.1416

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: 16
• 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	Qwk	Mse
No log	0.3333	2	0.7029	-0.0252	0.7093
No log	0.6667	4	0.1462	-0.1951	0.1344
No log	1.0	6	0.2440	0.0	0.2429
No log	1.3333	8	0.1359	0.0	0.1325
No log	1.6667	10	0.1162	0.0219	0.1100
No log	2.0	12	0.1085	0.0345	0.1016
No log	2.3333	14	0.1197	0.0	0.1139
No log	2.6667	16	0.1465	0.0	0.1427
No log	3.0	18	0.1499	0.0	0.1462
No log	3.3333	20	0.1428	0.0	0.1389
No log	3.6667	22	0.1187	0.0483	0.1125
No log	4.0	24	0.1323	0.1217	0.1220
No log	4.3333	26	0.1627	-0.1667	0.1498
No log	4.6667	28	0.1511	0.2075	0.1392
No log	5.0	30	0.1317	0.1217	0.1222
No log	5.3333	32	0.1331	0.0637	0.1256
No log	5.6667	34	0.1407	0.0105	0.1341
No log	6.0	36	0.1544	0.0105	0.1484
No log	6.3333	38	0.1639	0.0219	0.1582
No log	6.6667	40	0.1595	0.0483	0.1529
No log	7.0	42	0.1480	0.0808	0.1402
No log	7.3333	44	0.1440	0.1000	0.1358
No log	7.6667	46	0.1462	0.1000	0.1382
No log	8.0	48	0.1490	0.1747	0.1404
No log	8.3333	50	0.1535	0.1747	0.1441
No log	8.6667	52	0.1570	0.0094	0.1473
No log	9.0	54	0.1553	0.0094	0.1456
No log	9.3333	56	0.1529	0.1747	0.1436
No log	9.6667	58	0.1515	0.1747	0.1422
No log	10.0	60	0.1509	0.1747	0.1416

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1