arabert_baseline_mechanics_task6_fold0

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: 1.0047
• Qwk: 0.4859
• Mse: 1.0047

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	0.5	2	1.3188	0.1646	1.3188
No log	1.0	4	0.9104	0.3453	0.9104
No log	1.5	6	0.9681	0.3684	0.9681
No log	2.0	8	0.9649	0.3053	0.9649
No log	2.5	10	0.8511	0.3684	0.8511
No log	3.0	12	0.8161	0.3869	0.8161
No log	3.5	14	0.8312	0.5333	0.8312
No log	4.0	16	1.0768	0.4129	1.0768
No log	4.5	18	1.1227	0.4277	1.1227
No log	5.0	20	1.0800	0.4277	1.0800
No log	5.5	22	0.9594	0.5648	0.9594
No log	6.0	24	0.8494	0.4400	0.8494
No log	6.5	26	0.8298	0.4413	0.8298
No log	7.0	28	0.8737	0.4400	0.8737
No log	7.5	30	0.9941	0.6244	0.9941
No log	8.0	32	1.0585	0.5088	1.0585
No log	8.5	34	1.0554	0.5088	1.0554
No log	9.0	36	1.0367	0.4385	1.0367
No log	9.5	38	1.0146	0.5556	1.0146
No log	10.0	40	1.0047	0.4859	1.0047

Framework versions

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