arabert_cross_vocabulary_task4_fold6

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: 0.3809
• Qwk: 0.7212
• Mse: 0.3803

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	3.5915	0.0135	3.5774
No log	0.0580	4	1.6728	0.1066	1.6543
No log	0.0870	6	0.9207	0.3071	0.9146
No log	0.1159	8	0.9451	0.4183	0.9392
No log	0.1449	10	1.0092	0.4761	1.0019
No log	0.1739	12	0.8467	0.4077	0.8353
No log	0.2029	14	0.7589	0.4123	0.7471
No log	0.2319	16	0.6332	0.4728	0.6243
No log	0.2609	18	0.6097	0.5641	0.6051
No log	0.2899	20	0.5389	0.5141	0.5354
No log	0.3188	22	0.5073	0.5047	0.5034
No log	0.3478	24	0.4996	0.5613	0.4954
No log	0.3768	26	0.4724	0.5931	0.4689
No log	0.4058	28	0.4067	0.6836	0.4044
No log	0.4348	30	0.4385	0.8033	0.4374
No log	0.4638	32	0.4624	0.8118	0.4616
No log	0.4928	34	0.4385	0.7922	0.4378
No log	0.5217	36	0.4184	0.6922	0.4169
No log	0.5507	38	0.4874	0.6033	0.4851
No log	0.5797	40	0.6206	0.5584	0.6172
No log	0.6087	42	0.7056	0.5187	0.7018
No log	0.6377	44	0.6335	0.5490	0.6305
No log	0.6667	46	0.4625	0.6195	0.4610
No log	0.6957	48	0.3971	0.7571	0.3968
No log	0.7246	50	0.4158	0.7849	0.4158
No log	0.7536	52	0.4127	0.7821	0.4127
No log	0.7826	54	0.4163	0.7855	0.4162
No log	0.8116	56	0.4055	0.7826	0.4053
No log	0.8406	58	0.3891	0.7730	0.3888
No log	0.8696	60	0.3833	0.7639	0.3829
No log	0.8986	62	0.3800	0.7588	0.3795
No log	0.9275	64	0.3788	0.7357	0.3783
No log	0.9565	66	0.3800	0.7299	0.3795
No log	0.9855	68	0.3809	0.7212	0.3803

Framework versions

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