arabert_cross_vocabulary_task4_fold4

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: 0.6439
• Qwk: 0.8180
• Mse: 0.6439

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: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.0290	2	3.9935	0.0078	3.9935
No log	0.0580	4	2.2937	0.0536	2.2937
No log	0.0870	6	1.2377	0.1729	1.2377
No log	0.1159	8	1.3353	0.1826	1.3353
No log	0.1449	10	1.1216	0.4773	1.1216
No log	0.1739	12	0.8392	0.5499	0.8392
No log	0.2029	14	0.6966	0.6127	0.6966
No log	0.2319	16	0.6473	0.6969	0.6473
No log	0.2609	18	0.6961	0.7720	0.6961
No log	0.2899	20	0.5180	0.8112	0.5180
No log	0.3188	22	0.4069	0.7973	0.4069
No log	0.3478	24	0.3749	0.7620	0.3749
No log	0.3768	26	0.3945	0.7235	0.3945
No log	0.4058	28	0.3796	0.7580	0.3796
No log	0.4348	30	0.4995	0.8231	0.4995
No log	0.4638	32	0.6489	0.7303	0.6489
No log	0.4928	34	0.6807	0.6611	0.6807
No log	0.5217	36	0.5950	0.7894	0.5950
No log	0.5507	38	0.4549	0.8319	0.4549
No log	0.5797	40	0.3601	0.7833	0.3601
No log	0.6087	42	0.3472	0.7513	0.3472
No log	0.6377	44	0.3541	0.716	0.3541
No log	0.6667	46	0.3486	0.7411	0.3486
No log	0.6957	48	0.3480	0.7678	0.3480
No log	0.7246	50	0.3984	0.8135	0.3984
No log	0.7536	52	0.5146	0.8349	0.5146
No log	0.7826	54	0.6619	0.8250	0.6619
No log	0.8116	56	0.7304	0.7951	0.7304
No log	0.8406	58	0.7284	0.7985	0.7284
No log	0.8696	60	0.7335	0.8019	0.7335
No log	0.8986	62	0.7134	0.8053	0.7134
No log	0.9275	64	0.6805	0.8133	0.6805
No log	0.9565	66	0.6611	0.8180	0.6611
No log	0.9855	68	0.6439	0.8180	0.6439

Framework versions

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