arabert_baseline_mechanics_task5_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: 0.6569
• Qwk: 0.7179
• Mse: 0.6569

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	1.9122	0.0290	1.9122
No log	0.6667	4	0.8359	0.0	0.8359
No log	1.0	6	0.5865	0.3478	0.5865
No log	1.3333	8	0.5525	0.1739	0.5525
No log	1.6667	10	0.5082	0.25	0.5082
No log	2.0	12	0.5342	0.3636	0.5342
No log	2.3333	14	0.5194	0.3478	0.5194
No log	2.6667	16	0.5004	0.5714	0.5004
No log	3.0	18	0.5279	0.4848	0.5279
No log	3.3333	20	0.5491	0.4848	0.5491
No log	3.6667	22	0.6345	0.6154	0.6345
No log	4.0	24	0.7154	0.6111	0.7154
No log	4.3333	26	0.8157	0.5882	0.8157
No log	4.6667	28	0.8595	0.6667	0.8595
No log	5.0	30	0.7144	0.6842	0.7144
No log	5.3333	32	0.6153	0.6	0.6153
No log	5.6667	34	0.6974	0.6154	0.6974
No log	6.0	36	0.7895	0.7	0.7895
No log	6.3333	38	0.8393	0.6667	0.8393
No log	6.6667	40	0.8578	0.6667	0.8578
No log	7.0	42	0.8027	0.6667	0.8027
No log	7.3333	44	0.7182	0.6667	0.7182
No log	7.6667	46	0.6647	0.6842	0.6647
No log	8.0	48	0.6573	0.6	0.6573
No log	8.3333	50	0.6541	0.6842	0.6541
No log	8.6667	52	0.6499	0.6842	0.6499
No log	9.0	54	0.6477	0.7179	0.6477
No log	9.3333	56	0.6501	0.7179	0.6501
No log	9.6667	58	0.6545	0.7179	0.6545
No log	10.0	60	0.6569	0.7179	0.6569

Framework versions

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