arabert_baseline_mechanics_task8_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: 0.5999
• Qwk: 0.7490
• Mse: 0.5999

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	1.4400	0.2581	1.4400
No log	1.0	4	1.3435	0.3959	1.3435
No log	1.5	6	0.7948	0.6199	0.7948
No log	2.0	8	0.8359	0.5911	0.8359
No log	2.5	10	0.5994	0.5641	0.5994
No log	3.0	12	0.5724	0.7742	0.5724
No log	3.5	14	0.5128	0.7153	0.5128
No log	4.0	16	0.5177	0.7153	0.5177
No log	4.5	18	0.5100	0.7153	0.5100
No log	5.0	20	0.5127	0.6784	0.5127
No log	5.5	22	0.5088	0.6854	0.5088
No log	6.0	24	0.5211	0.7490	0.5211
No log	6.5	26	0.5074	0.6784	0.5074
No log	7.0	28	0.5120	0.6784	0.5120
No log	7.5	30	0.5312	0.7871	0.5312
No log	8.0	32	0.5850	0.7871	0.5850
No log	8.5	34	0.6173	0.7586	0.6173
No log	9.0	36	0.6118	0.7586	0.6118
No log	9.5	38	0.6068	0.7449	0.6068
No log	10.0	40	0.5999	0.7490	0.5999

Framework versions

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