ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k1_task3_organization

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: 0.9988
• Qwk: 0.0261
• Mse: 0.9988
• Rmse: 0.9994

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: 8
• eval_batch_size: 8
• 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	Rmse
No log	0.2857	2	3.4390	-0.0053	3.4390	1.8544
No log	0.5714	4	1.8180	-0.0390	1.8180	1.3483
No log	0.8571	6	1.1737	0.0038	1.1737	1.0834
No log	1.1429	8	0.7425	0.2239	0.7425	0.8617
No log	1.4286	10	0.6278	0.0	0.6278	0.7923
No log	1.7143	12	0.7388	-0.0794	0.7388	0.8596
No log	2.0	14	0.7504	-0.0746	0.7504	0.8662
No log	2.2857	16	0.7916	-0.0370	0.7916	0.8897
No log	2.5714	18	0.8110	-0.0370	0.8110	0.9006
No log	2.8571	20	0.7747	-0.0370	0.7747	0.8802
No log	3.1429	22	0.6741	-0.0081	0.6741	0.8210
No log	3.4286	24	0.6633	-0.0233	0.6633	0.8144
No log	3.7143	26	0.6875	0.0448	0.6875	0.8292
No log	4.0	28	0.6947	0.0448	0.6947	0.8335
No log	4.2857	30	0.6838	-0.0909	0.6838	0.8269
No log	4.5714	32	0.6912	-0.0496	0.6912	0.8314
No log	4.8571	34	0.7184	0.0137	0.7184	0.8476
No log	5.1429	36	0.7700	0.0843	0.7700	0.8775
No log	5.4286	38	0.7980	0.0939	0.7980	0.8933
No log	5.7143	40	0.7711	0.0361	0.7711	0.8781
No log	6.0	42	0.7594	0.0698	0.7594	0.8715
No log	6.2857	44	0.7746	0.0698	0.7746	0.8801
No log	6.5714	46	0.8247	0.0455	0.8247	0.9081
No log	6.8571	48	0.8304	0.0769	0.8304	0.9113
No log	7.1429	50	0.9194	0.0370	0.9194	0.9588
No log	7.4286	52	1.1944	0.1429	1.1944	1.0929
No log	7.7143	54	1.3631	0.0681	1.3631	1.1675
No log	8.0	56	1.3001	0.0491	1.3001	1.1402
No log	8.2857	58	1.2407	0.0938	1.2407	1.1139
No log	8.5714	60	1.2025	0.0569	1.2025	1.0966
No log	8.8571	62	1.1216	0.1008	1.1216	1.0590
No log	9.1429	64	1.0577	0.0569	1.0577	1.0284
No log	9.4286	66	1.0190	-0.0295	1.0190	1.0095
No log	9.7143	68	1.0026	0.0261	1.0026	1.0013
No log	10.0	70	0.9988	0.0261	0.9988	0.9994

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1